•proceedings of % 
^ n t e n r a i u m a l ^ R u b b e r C o n f e r e n c e 
"75 / f e a r s o f ; E u b t a r ^ s e a r c h i n ^ r i l l a t t f e a " 
Colombo, 17-19 S e p t e m b e r , 1984. 
VOL. 1 
PART (IT) 
RUBBER RESEARCH INSTITUTE OF SRI LANKA 
AGALAWATTA 
PROCEEDINGS OF THE INTERNATIONAL RUBBER CONFERENCE 
SRI LANM 
Vol. 1 (Part II) 1 9 8 4 
C O N T E N T S 
PAGE 
SESSION 7. SMALLHOLDINGS 
INTERCROPPING HEVEA RE PL A M I N O S DURING THE IMMATURE 
PERIOD — L. B. Chandrasekara .... .... .... .... 389 
PROGRESS AND DEVELOPMENT OF THE SMALLHOLDING SECTOR 
OF THE NATURAL RUBBER INDUSTRY IN MALAYSIA — Raja 
Badrul Shah Kobat .... .... .... .... .... 395 
STAGES OF DEVELOPMENT AND GOVERNMENT POLICY IN 
RUBBER SMALLHOLDING AGRICULTURE — Colin Barlow and 
S. K. Jayasuriya .... .... .... .... .... 409 
SESSION 8. ROOT DISEASES 
TOWARDS A BETTER LIVING STANDARD OF THE RUBBER SMALL­
HOLDER — R. P. M. de Zoysa and L. K. Meegahawatta .... .... 427 
SOCIO-ECONOMIC CHARACTERISTICS OF RUBBER SMALLHOL­
DERS IN SRI LANKA : IMPLICATIONS FOR DEVELOPMENT — 
H. M. G. Herath and W. G. Jayasena .... .... .... .... 437 
PATHOGENICITY VARIABILITY OF RIGIDOPORUS LIGNOSUS 
AND PHELLINUS NOXIUS — 3. P. Geiger, D . Nandris, M. Nicole and 
B . R i o .... _ .... .... .... „ 449 
A NEW METHOD OF DIRECT CONTROL OF FOMES LIGNOSUS, 
CAUSING AGENT OF WHITE ROOT ROT OF HEVEA — Tran Van 
Canh .... .... .... .... .... .... 473 
RUBBER ROOT ROTS : CELLULAR AND MOLECULAR ASPECTS 
OF HOST-PARASITE INTERACTIONS — M. Nicole, J. P. Geiger and 
D. Nandris .... .... .... .... .... 493 
AN INTEGRATED APPROACH TO CONTROL OF WHITE ROOT 
DISEASE IN SRI LANKA — A . de S. Liyanage, O. S. Peries, S. S. 
Warnapura, E. A. T. Senadeera and W. Arnaratunga _o « i 499 
SESSION 9. SOIL MANAGEMENT 
PAGE 
MANAGEMENT OF GROUND COVERS FOR OPTIMUM PRODUC­
TION — N. Yogaratnam, A. M. A. Perera and G. J. de Mel .... .... 521 
SOME ASPECTS OF MOISTURE AND SOIL CONSERVATION IN 
RUBBER PLANTATIONS — Lalani Samarappuli and N. Yogaratnam .... 529 
APPROACHES TOWARDS LAND EVALUATION SYSTEMS FOR 
HEVEA BRASILIENSIS CULTIVATION IN PENINSULAR MALAY­
SIA — H. Y. Chan, F. K. Yew and E. Pushparajah .... .... 545 
NODULATION OF PUERARIA PHASEOLOIDES BY INTRODUCED 
RHIZOBIA IN COMPETITION WITH NATURALIZED STRAINS 
IN THREE DIFFERENT SOILS — C. K. Jayasinghe, C. A. Parker, D. K. 
Kidby and S. A. Kulasooriya .... .... .... .... 563 
SESSION 10. MARKETING 
INVESTMENT OPPORTUNITIES IN THE RUBBER INDUSTRIES IN 
SRI LANKA — M. Nadarajah and G. T. Unamboowe .... .... 569 
THE SPECIAL PROBLEMS OF MARKETING R U B B E R - G u y Siri-
manne 
NATURAL RUBBER : PAST PROBLEMS, FUTURE PROSPECTS 
P. W. Allen 
THE IMPACT OF GOVERNMENT POLICIES O N INVESTMENT IN 
THE RUBBER INDUSTRY OF SRI LANKA — S. K. Jayasuriya and 
C. Barlow 
579 
BROKERS ROLE IN THE MARKETING OF SRI LANKA RUBBER — 
Ranjith Perera .... .... .... 5gj 
SESSION 11. ECONOMICS OF RUBBER 
589 
NATURAL RUBBER SUPPLY IN SRI LANKA : ANALYSIS AND PRO­
JECTIONS USING VINTAGE APPROACH — H. P. Smh .... 597 
THE INTERNATIONAL NATURAL RUBBER AGREEMENT . AN 
A S S E S S M E N T — C . Suan Tan .... ' .... .... .... 625 
COMPLETING THE CENTURY . LONG RUN PRODUCTION CYCLES 
AND THE SRI LANKA RUBBER INDUSTRY OVER THE NEXT 
25 YEARS — D. M. Etherington and W. M. Premachandra .... 651 
675 
PAGE 
EFFEdTVITY OF RESEARCH AND ITS IMPLEMENTATION IN THE 
INDUSTRY — D. Lamb .... 707 
THE PROBLEMS AND PROSPECTS OF THE RUBBER INDUSTRY 
EM SRI LANKA — M. R. C. Peiris .... .... ... •• 7 1 1 
GROWING OF RUBBER WITH SPECIAL REFERENCE TO REPLAN­
TING CYCLES — D. E. C. Wijesinghe 
CLOSING SESSION. 
SEVENTY FIVE YEARS OF RUBBER RESEARCH IN SRI LANKA — 
O. S. Peries 
SESSION 12. PLANTERS' VIEWS 
COMPARATIVE VIEW OF MANAGEMENT IN THE PLANTATION 
INDUSTRY IN SRI LANKA AND MALAYSIA — M . P. G. S. 
Jayawardena .... .... .... .... .... .... 689 
RESEARCH REQUIREMENTS FOR THE RUBBER INDUSTRY — C. N. 
M. Rodrigo .... .... .... .... .... 703 
SMALLHOLDINGS 
INTERCROPPING HEVEA REPLANTINGS DURING THE 
IMMATURE PERIOD 
By 
L. B. CHANDRASEKARA 
(Rubber Research Institute of Sri Lanka, Agalawatta) 
Importance of intercropping 
Most of the Hevea plantations in Sri Lanka are located in the low country wet dis­
tricts which receive rainfall during both the North East and South West monsoon periods. 
Under such conditions, a plantation of Hevea budgrafts normally takes around 6 years 
to reach tappable girth. Limited acreages of Hevea plantations are also found in the 
drier parts of the country such as in the Matale, Kurunegala, Badulla and Moneragala 
districts. Due to limited rainfall conditions in these districts, a plantation of Hevea 
budgrafts takes much longer, sometimes up to 9 years to reach tappable girth. During 
this period of immaturity, it has been possible to utilize the inter-row space for growing 
a selection of crops that would give an income to the estate owner until the Hevea trees 
are brought into tapping in about the seventh year from planting. This is particularly 
important for the smallholder who had so far been reluctant to replant his old and low 
yielding rubber due to the long period of time he has to wait until he gets some return 
from his land. 
In Sri Lanka, particularly in the major rubber growing districts, rubber is generally 
replanted on an approximately 33 year cycle. If such replanting is carried out regularly, 
approximately 3 percent of the total rubber acreage in Sri Lanka will be replanted each 
year. On a national level therefore, approximately 15 percent of the rubber acreage 
would be immature at any given time and would have the potential to be intercropped. 
This represents an extent of about 30,000 hectares. 
Type of crops 
In Sri Lanka, the high annual rainfall renders the steep terrain, in which most rubber 
is grown, subject to soil erosion if the land is clean weeded. This factor narrows down 
the scope for intercropping to mostly perennial crops that do not require clean weeding 
and regular cultivation of soil. Apart from this, a suitable crop should ideally have the 
following characteristics. 
(a) Non-competitive with rubber 
(b) Suitable for the climatic and soil conditions of the area 
(c) Less labour intensive 
(d) N o necessity for irrigation or regular watering 
(e) Freedom from diseases, particularly those that affect rubber 
(f) Should have ready markets 
Bearing these in mind, the Rubber Research Institute of Sri Lanka has carried out 
trials with a wide range of crops and the following have crystallized out as being suitable 
for interplanting among young rubber in the wet districts. 
3 8 9 
Banana, passion fruit and pineapple 
These crops require only circle weeding round each plant, just like in rubber, while 
the rest of the land can remain under ground covers. Pineapple however, is an exception 
that requires clean weeding and is therefore suitable for cultivation on flat or gently sloping 
land. For intercropping, rubber is best planted with a minimum inter-row distance of 6 m. 
Effects -of intercrops on rubber 
We have carried out three large scale experiments involving the inter-planting of 
banana, passion fruit and coffee among newly planted rubber. In each experiment the 
rubber is planted at an initial density of 450 points per hectare with a spacing of 6 m bet­
ween rows. A single row of banana, passion fruit or coffee is planted centrally between 
each pair of rubber rows with the following spacing along the row. 
Banana — 3-7 m 
Coffee — 3-7 m 
Passion fruit — 7 • 5 m 
All three crops are included in each experiment and the three experiments are sited 
in three districts, namely Ratnapura, Kalutara and Kegalla. The experimental design 
is a randomised layout with each treatment replicated six times. The average girth 
measurements of rubber recorded at 3 years of age are given in Table 1. 
Table 1. Mean girth (cm) of Hevea budgraftsat 3 yrs of age in three field experiments 
Crop Experiment I Experiment II Experiment III 
Banana 21-75 21-18 29-82 
Passion fruit 21-61 19-93 29-58 
Coffee 21-77 21-53 30-50 
Control 21-15 19-13 31-22 
The girth differences between the treatments are statistically not significant. This 
means that interplanting these crops has not affected the growth of rubber up to the 
3rd year from planting. 
In the case of pineapple, there was a large scale experiment sited on flat land in the 
Kalutara District. The treatments consisted of 1 row, 2 rows, 3 rows and 4 rows of pine­
apple planted centrally between each.pair of rubber rows and a control without pineapple. 
The five treatments were replicated six times. 
The pineapple plants were spaced 45-7 cm along the row and 0-6.m between rows. 
Where 4 rows of pineapple are planted, there was a central ;path 1 - 5 m wide between each 
pair of pineapple rows to permit weeding and other cultural operations. At the above 
390 
spacing, four rows of pineapple plants between each pair of rubber rows would result in 
an approximate stand of 9,400 pineapple plants per hectare when no allowance is made 
for rocks and paths. The girth measurements of rubber budgrafts in the experimental 
area were taken at 3 years of age and are summarised in Table 2. 
Table 2. Mean girth (cm) of Hevea budgrafts intercropped with pineapple at 3 years of age 
Treatment Mean girth (cm) 
Pineapple — 1 row 16-57 
— 2 rows 16-43 
„ — 3 rows 16-32 
— 4 rows 18-10 
Control 13-78 
LSD — 2-51 
The girth differences between the control plots and the intercropped plots have been 
statistically significant. This means that the growth of rubber in the intercropped plots 
has been better than the control plots. 
Bananas 
The cultivation of bananas as an intercrop offers attractive prospects as it has the 
following advantages : 
1. grows well in most parts of the country 
2. less labour intensive than other crops 
3. requires less capital investment than others 
4. has ready markets in Sri Lanka. 
However, experiments that have been carried out so far show a marked influence of 
rainfall on yield. One difficulty in estimating the yield of bananas in experimental plots 
had been due to thefts which are difficult to prevent. Therefore in experimental areas 
the bunches were numbered serially and before they are fully mature the number of fruits 
per bunch were counted and the value of bunches estimated. Table 3 gives a summary 
of the estimated mean value of bunches in three rubber growing districts. 
Table 3. Mean value (Rs.) per bunch of bananas in three field experiments 
District No. of trees Mean value per bunch 
1st year 2nd year 
Ratnapura 180 20-63 23-98 
Kalutara 180 17-53 17-95 
Kegalla 240 7-30 5-95 
391 
The mean value reflects the size of bunches harvested, the largest of which were recor­
ded from Ratnapura District. Poorest yields were recorded in the relatively dry Kegalla 
District. 
Table 4 gives the estimated income in rupees per hectare for the three districts accor­
ding to which the lowest income has been recorded for the Kegalla district. 
Table 4. Estimated income in Rs. per hectare of bananas in three field experiments 
District Number of plants 1st year 2nd year 3rd year 
Ratnapura 180 5,362 10,030 7,225 
Kalutara 180 3,092 3,290 4,815 
Kegalla 240 382 340 232 
Passion fruit 
Although passion fruit can be grown in most parts of Sri Lanka, cultivation on a 
commercial scale is at present confined to four districts, namely Kalutara, Ratnapura, 
Galle and Avissawella the reasons being the following. 
1. Availability of marketing facilities 
2. Suitability of climatic conditions for optimum growth of vines and yield of fruit. 
Here again, due to thefts, the yield of vines in experimental plots were evaluated by 
counting the immature fruits at suitable intervals. The counted fruits were marked with 
a dot to prevent a re-count. At harvesting time, a random sample of 100 fruit were 
drawn, weighed and from the resulting value, the weight of the total crop was estimated. 
Table 5 gives a summary of the mean yield of passion fruit in three field experiments 
located in three rubber growing districts. 
Table 5. Mean yield of passion fruit in kg per plant in three field experiments 
-
District No . of plants 1st year 2nd year 
Ratnapura 109 6-39 12-00 
Kalutara 99 5-81 14-50 
Kegalla 142 2-18 2-50 
Ratnapura and Kalutara Districts with their well distributed rainfall have recorded 
the highest yield of fruit per vine. 
392 
Table 6 gives the estimated income from growing passion fruit as an intercrop in 
three experimental areas. 
Table 6. Estimated income in Rs.per hectare of passion fruit crop in three field experiments 
Estimated income 
Districts N o . of plants 1st year 2nd year 
3rd year 
Ratnapura 109 11,510 21,255 
7,885 
Kalutara 99 10,452 21,407 
— 
Kegalla 142 3,927 4,257 
3,695 
According to data in Table 6, passion fruit had so far been the most profitable crop. 
Pineapple 
The cultivation of pineapple as an intercrop has not been very popular so 
smallholders. This may probably be due to the following reasons. 
1. It requires a higher capital investment 
2. It is more labour intensive 
In one large scale experiment involving an extent of approximately 1.25 hectares, 
an average income of Rs. 11,752 per hectare was recorded at the end of the second year 
in 1983. In certain locations, particularly in the Kalutara district, pineapple plants were 
also observed to the affected with a dry rot caused by the fungus Ceratocystis paradoxa. 
393 
PROGRESS AND DEVELOPMENT OF THE SMALLHOLDING 
SECTOR OF THE NATURAL RUBBER INDUSTRY IN 
MALAYSIA 
By 
RAJA BADRUL SHAH KOBAT 
(Rubber Research Institute, Malaysia, Kuala Lumpur) 
ABSTRACT 
The paper reviews the progress of the rubber smallholding sector and its importance 
to the Malaysian economy. Current development efforts undertaken by the Government 
aim at further increasing the productivity and income of rubber smallholders by way of a 
dynamic replanting program which incorporates various recent technological inputs and 
innovative extension efforts. The development of the Malaysian rubber smallholding sector 
today is more and more geared to the group and integrated agricultural development approach 
where technology and development package derived from research findings on rubber and 
other viable agricultural enterprises are consciously adapted and transferred to smallholders 
at both the immature and mature phases of rubber growth. Such technologies include, 
among other things, rational felling policy for rubber wood utilisation, replanting with high 
yielding rubber clones, efficient exploitation systems, modern farm management techniques 
and profitable integrated and mixed farming systems. Wider application and extension of 
this integrated technology and farming systems approach in smallholder development can 
ensure the continuity and socioeconomic viability of the smallholding sector as the most 
important and efficient production unit in the Malaysian agricultural economy. 
INTRODUCTION 
Total world demand for rubber ( N R + S R ) today amounts to about 13 million metric 
tonnes annually. From this total, about 30% of the world supply comes from N R , 
41 % of which from Malaysia mainly from the smallholding sector. Table 1 shows that 
in Peninsular Malaysia the smallholdings occupied 73% of land under rubber, or about 
two-thirds of the land area under agricultural crops, and produced 63 % of the total rubber 
production in 1983. N R thus contributes to a major source of the country's employment 
and foreign exchange earnings. Numbering about half a million farms, the smallholding 
sector with an average family size of 5 5 persons supported directly or indirectly about 
three million people or 30 % of Peninsular Malaysia's population and N R alone contri­
butes to 15% of total export earnings in 1983. 
Progress of the smallholding sector 
In terms of hectarage and production, the rubber smallholder sector has made a tre­
mendous expansion especially smallholdings under RISDA. Planted hectarage increased 
from 1 077 300 hectares in 1970 to 1 205 700 hectares in 1980 while production increased 
from 594 757 tonnes in 1970 to 877 090 tonnes in 1980 and yield increased from 753 
kg/ha/year to 1 104 kg/ha/year over the same period (Table 2). 
395 
Table 1. Rubber hectarage and production in Malaysia 1981 — 1983 
vo 
Regions and Sectors 
Total Malaysia 
Peninsular Malaysia 
Sabah & Sarawak 
Estates 
Smallholdings 
Area under rubber (Hectarage) 
1981 
2 030 000 
1 720 000 
310 000 
490 000 
1 230 000 
1982 
2 010 135 
1 700 700 
309 435 
473 200 
1 227 500 
1983 
2 009 900 
1 702 400 
307 500 
464 000 
1 238 400 
%(1983) 
100 
85 
15 
27 
73 
Rubber production (Tonnes) 
1981 
565 000 
500 000 
63 000 
597 800 
967 200 
Sources : Rubber Research Institute of Malaysia Annual Reports 1981, 1982, 1983 
Rubber Monthly Statistics, Department of Statistics, Malaysia 
1982 
1 516 585 
1 478 982 
37 603 
576 721 
902 261 
1983 
1 561 992 
1 521 617 
40 375 
560 114 
961 503 
% (1983> 
100 
97 
3 
37 
63 
Table 2. Planted hectarage, production and yield on smallholdings in Peninsular Malaysia (1970 —1980) 
('000 Hectares) 
Total Estimated 
Year FELDA RISDA FELCRA Others3 Total Productions Av. Yield 
(tonnes) (kg/ha) 
1970 59-8 436-0 5-7 575-8 1077-3 594 757 753 
1971 67-6 459-2 8-3 551-4 1086-5 608 865 735 
1972 76-2 482-6 1 5 0 518-2 1092 0 598 815 703 
1973 88-8 511-2 21 5 483-1 1104-6 791 519 927 
1974 94-3 534-8 23-6 464-9 1117-6 800 972 1039 
1975 105-1 555-9 26-1 444-5 1131-6 817 505 1068 
1976 114-7 570-2 26-2 419-6 1130-7 884 542 1093 
1977 124-6 583 0 28-5 409-2 1145-3 883 919 1102 
1978 145-2 596-3 31-2 403 0 1175-7 888 047 1104 
1979 156-8 610-6 35-1 392-6 1195-1 890 053 1105 
1980 168-9 626-3 41 8 368-7 1205-7 877 090 1104 
1981 — — — 1226-2 882 300 — 
a Include independent smallholdings (Including those assisted by RISDA) and organised smallholding 
other than those in FELDA and FELCRA Schemes, 
b Source: MRRDB 
Source : Department of Statistics, West Malaysia, Rubber Statistics Handbook. 
The replanting rate as shown in Table 3 indicates that in 1970, 90% of the estates 
have been replanted compared to only 46% in the smallholdings. By 1980, the rate 
has increased to 99% for the estate and 72% for the smallholdings. The replanting rate 
for smallholdings in 1982 was estimated at 78%. In terms of replanted hectarage, Table 4 
shows hectarage replanted since 1975 and estimated replanting programmes up to 1985. 
An annual average of 17 000 hectares of rubber were replanted between 1975 — 82. The 
last decade has thus seen a revitalising of the trees in the smallholdings through replanting. 
Past trends have also shown replanting activities to increase during periods of low rubber 
prices where the industry took stock of itself and reinvested in replanting. Indications 
are that during the current low rubber price a similar reinvestment will occur as indicated 
by the replanting projections for 1983 — 85 in Table 4. 
Table 3. Percentage under replanting (1970 — 1982) 
Year Estate % Smallholding % 
1970 90 46 
1971 92 49 
1972 93 52 
1973 95 56 
1974 95 59 
1975 97 62 
1976 98 64 
1977 98 66 
1978 98 68 
1979 99 70 
1980 99 72 
1981 — 75 
1982 - 78* 
* Up to 1982, an estimated 791,969 hectares (78%) on smallholdings 
have been replanted. 
397 
Table 4. Hectarage replanted since 1975 andestimatedreplanting up to 1985 
Hectarage Hectarage Total hectarage Estimated 
Year replanted replanted replanted replanting 
(rubber only) (other crops) (all crops) (ha all crops) 
1975 20 709 12 470 33 179 
1976 14 250 5 485 19 735 
1977 12 789 5 419 18 208 
1978 13 255 5 943 19 198 „ 
1979 14 186 8 205 22 391 — 
1980 15 300 7 685 22 985 
1981 22 624 6 827 29 451 26 710 
1982 23 699 7 484 31 183 30 271 
1983 — — — 33 355 
1984 — — — 36 422* 
1985 — — — 38 446* 
136 812 59 518 196 330 165 204 
* Including replanting on Felda Schemes : 
1984— + 3643 hectares 
1985 — + 4048 hectares 
Source: RISDA 
The expected achievement in replanting for the period 1981 — 1985 is as shown in 
Table 5. A total of about 152 000 hectares are expected to be replanted at a total cost 
of nearly $457 million by the end of the Fourth Malaysia Plan period. 
Table 5. Expected achievement in replanting program for the period 1981 —1985 in 
Peninsular Malaysia 
Year Hectare Cost (?) 
1981* 29 550 81 621 402 
1982 30 467 96 715 581 
1983 33 355 121 908 783 
1984 28 651 76 567 000 
1985 29 697 80 081 000 
* Actual achievement 
Source : RISDA 
Problems in the smallholding sector 
Though smallholding rubber acreages have increased significantly over the years 
vis-a-vis the estate sector, its productivity still lags behind that of the latter due mainly 
to structural problems. The average rubber yield from the smallholdings is relatively 
398 
lower at 1 104 kg/ha/yr, compared to the estates yield of 1 507 kg/ha/yr (Table 2). Socio­
economic institutional and technological inadequacies which include the uneconomic 
sized holdings, ageing labour, inefficient production and management techniques together 
with the vagaries of the market on smallholders'produce, account for the low productivity 
and the high incidence of poverty among rubber smallholders at 41 - 3 % in 1980. They 
still form the single largest poverty group in the country numbering about 175 900 poor 
households. Their percentage among the total poor was the highest in the country at 
about 26 4% in 1980. In terms of average monthly family income, individual rubber 
smallholders were earning about M$ 240/-. However, rubber smallholders in land deve­
lopment schemes enjoyed a much higher level of net income of M$ 450/- monthly, owing 
to their larger farm size in 1983. Those with even larger land holdings of 12 acres 
earned a net average income of M$ 1041/- per month during the same period. 
In view of these structural problems which resulted in low productivity and income 
particularly among the individual smallholders, poverty alleviation programs in the rubber 
smallholdings have to be made more comprehensive. Such programs cover the total 
agricultural activities that can be developed to raise the general level of farm productivity 
and income from the rubber holdings. 
Government efforts to raise farm productivity and income in smallholdings 
Various steps were instituted by the Malaysian Government through its development 
agencies to raise farm productivity and income among rubber smallholders. These in­
clude developing them on organised group basis through various development strategies 
like Group Replanting; Schemes and mini estates in addition, to the established and orga­
nised land development schemes of FELDA, FELCRA, State Schemes and so forth. 
Future development for the individual smallholders, where the problem of poverty is 
most acute, is increasingly being organised on the basis of group replanting and integrated 
agricultural development projects in line with the National Agricultural Policy. Such 
organised and integrated farming systems approach enable an organised transfer of 
technology both for rubber and other related feasible agricultural activities. This multi-
activity approach can ensure future viability and efficiency of smallholdings as enduring 
production units at par with the organised schemes and mini estates in achieving its socio­
economic objectives. 
Smallholders development through the Integrated Agricultural Development (IAD) 
approach in Rubber Group Replanting Schemes 
The Government rubber replanting program has progressed significantly since 1970. 
By 1982, about 78% of smallholdings have been replanted. In order to enhance further 
this replanting activity the IAD approach is advocated to achieve higher productivity and 
income in the smallholder sector. The IAD approach in group replanting embarks on 
the consolidation of the 1 — 2 ha individual rubber smallholdings into more or less larger 
contiguous block or area of 10 — 20 ha for efficient adoption of various modern and 
appropriately developed and adapted technologies at the immature and mature phases. 
The rationale for the selection of this size group is based on the fact that in Peninsular 
399 
Malaysia more than 70% of the smallholdings are owner-operated and fall under the 
12 — 2 0 ha, group. These entities fall out of the mini estate development concept and 
more often than not enjoy only peripheral effects of new technologies. 
The planning, implementation and coordination o f the IAD projects are done by 
the sponsoring authority, which in this case is the RRIM with assistance forthcoming from 
RISDA and other related development agencies. Project financing comes from the 
RISDA replanting grant for rubber and production credit from the sponsoring authority 
for other agricultural projects. The project cycle lasts through the immaturity period 
of rubber from which point other mature rubber technologies take on. The objectives 
of this multi-activity concept of technology transfer to smallholders are : 
© To provide the smallholders with ' package-deal' technologies in order that 
they might benefit from all new technologies available from research and deve­
lopment in the RRIM ; 
Q To raise general farm productivity by increasing farm production and family 
income in more diverse and efficient ways ; 
© To narrow the productivity gap between the smallholdings and estates sectors ; 
@ To serve as demonstration plots to other surrounding smallholders on the bene­
fits of adopting new technologies available to them. 
Smallholders development — Recommended technologies available at the pre-planting, 
replanting, immature and mature phases of rubber growth. 
The recommended technologies to increase the productivity and income of rubber 
smallholders at the various growth phases of rubber trees are as follows : 
(i) Pre-Replanting Phase : Yield stimulation 
(ii) Replanting Phase : Sale of rubber wood, the use of high yielding clones and 
advanced planting materials, efficient field maintenance 
programmes such as the growing of cover crops, fertili­
ser application and weed control. 
(iii) Immature Phase : Planting with cash-crops and/or poultry rearing 
(iv) Mature Phase : Poultry rearing and/or sheep rearing. 
Yield stimulation 
Stimulation with Ethephon can substantially increase yield in a wide range of rubber 
cultivars. This approach to increase yield is now a routine practice in the commercial 
plantations with rubber trees fifteen years and older. 
400 
Appropriate stimulation practices are now available to improve the economic yield 
of rubber smallholdings. The procedures involved minimal use of stimulant and judicious 
methods of application. These have been geared to suit smallholdings tapping and cul­
tural practices. It is estimated that about 60 % of the potential holdings with less than 
10 hectares can be stimulated with Ethephon. Additional income of $ 488/- ha/yr for 
mature holdings and $ 1245/- ha/yr for old holding can be realised from yield stimulation 
after deduction of input costs. 
Sale of rubber wood 
From replanting the rubber wood can be sold to generate extra income for small­
holders. It is estimated that 100 — 120 tonnes of rubber wood are obtainable from one 
hectare. At a price of $ 20/- per tonne waste wood, an income of $ 200/ $ 240/- can 
be expected from one hectare of smallholding if there is a buyer for the wood. 
Studies show that 40 % of the waste wood can be recovered for sawn timber. Table 6 
shows the estimated tonnage of sawable rubber wood obtainable per hectare from some 
clones on smallholdings. A greater output of wood can be expected from the GT 1, 
PR 107 and PB 5/51 clones when they are due for replanting at the age of around 30 years. 
On the average, about 40 tonnes of sawable rubber wood per hectare can be obtained at 
the time of replanting. 
Table 6. Estimated tonnage of sawable rubber wood obtainable per hectare from some 
clones 
Approximate Mean girth Tonnage of 
Clone Age N o . of stands/ 150 cm above sawable wood 
(years) hectare ground (cm) per hectare 
RRIM 600 26 200 103 66 
Tjir 1 25 250 83 42 
GT 1 18 250 79 37 
PR 107 18 250 82 32 
PB 5/51 19 250 83 32 
Clones 
For smallholdings to be replanted it is mandatory for these holdings to replant with 
high yielding clones. The clones recommended for smallholdings are shown below and 
their yield profiles are shown in Table 7. 
(a) Recommended clones for individual smallholdings 
CLASS I : GT 1 ; PR 217 ; PR 255 ; PR 261 ; RRIM 600 ; RRIM 712 
CLASS II : PB 280 ; RRIM 729 ; PB 235 
401 
Table 7. Mean yields* of recommended clones* (Classes I and If) for smallholdings 
CLONE 
CLASS I 
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 
RRIM 600 720 1210 1600 1860 2310 2320 2350 2470 2700 2360 2190 2040 2660 2940 3260 
RRIM 712 690 1490 2010 2330 2230 2290 2610 2290 2560 2760 2250 _ 
PR 255 1170 1500 1850 2250 1920 2070 2300 2140 2110 2050 2380 2210 2140 2100 2120 
PR 261 860 1290 1610 1840 1830 2240 2360 2420 2260 2120 1860 1600 1690 1870 1720 
PB 217 570 1050 1380 1520 1580 2200 2200 2270 1950 2020 2110 2210 2050 _ 
GT 1 
CLASS n 
700 1180 1410 1640 1570 1960 2280 2340 2310 1880 2040 1700 1530 1670 1640 
PB 235 1370 1870 2280 2300 2000 2060 3230 2530 2560 2530 2170 3290 
RRIM 729 800 1360 1920 2030 2210 2090 
PB 280 1090 1500 1890 2180 2240 2160 
a. In kilogram per hectare per year 
b. Data from large-scale done trials 
(b) Recommended clones for block-planted smallholdings 
CLASS I : GT 1 ; PB 217 ; PR 255 ; PR 261 ; RRIM 600 ; RRIM 712 
CLASS H : AVROS 2037 ; PB 28/59 ; PB 255 ; PB 260 ; PB 280 ; PB 235 ; 
RRIM 527 ; RRIM 729. 
Planting materials 
Two-whorl polybag planting materials are recommended since they are suitable for 
smallholding conditions. These materials can be prepared by the smallholders themselves 
to prevent problems of logistics when their supplies are organised through tenders. The 
use of other advanced planting materials such as maxi stumps is also recommended for 
smallholdings. These planting materials can reduce the immaturity period to 4 — 4 J 
years in the case of maxi stumps and 4-J — 5 years in the case of polybag materials. Be­
sides their use also result in higher planting success, uniform trees and high percentage of 
tappability. 
Maintenance 
In addition to recommended clones and planting materials, other agricultural prac­
tices have to be followed to ensure the long term productivity of replanted holdings. 
This includes efficient fields maintenance programmes namely, cover cropping, fertilizer 
application and weed control. 
(a) Cover crops 
Legume covers are recommended in the interrows of rubber in smallholdings. 
The advantages are that cover crops protect soil from erosion, add organic 
matter to the soil, keep the soil cool, fix atmospheric nitrogen and control weeds. 
(b) Fertiliser application and weed control 
Good management practices particularly judicious application of fertiliser 
coupled with efficient weed control can lead to increases in yield of 25 — 30 % 
and overall productivity of rubber trees in the smallholdings. Continued U6e 
of fertilisers based on the present discriminatory fertiliser recommendation 
also result in residual benefits through better bark regeneration. 
Integrated farming as a source of income in rubber smallholdings 
At the immature and mature phases of rubber growth, income of smallholders can 
be increased through integrated farming in rubber holdings. Integrated farming can 
be carried out along the following combinations of farming systems namely (a) cash crop 
— rubber, (b) poultry — rubber, (c) sheep — rubber and (d) poultry — sheep — rubber. 
(a) Cash - Crop - Rubber 
During the first 3 years of rubber cultivation about 80 % of the land, 
under rubber is suitable for planting with intercrops. Crops that require less 
than 3 years to mature may be cultivated in the rubber interrows. Thus maize, 
403 
groundnut, soybean and banana can be successfully cultivated on flat to undu­
lating smallholdings, while banana is suitable on hilly areas with slopes 
greater than 8°. 
(b) Poultry - Rubber 
Other than for intercrops, poultry sheds may be constructed in the rubber 
interrows for poultry rearing. Where day-old chicks, feed and market are 
available, poultry rearing especially broiler production is suitable for small­
holders. 
(c) Sheep - Rubber 
After about three years of growth, the leaves of the rubber trees normally 
shade the interrows, and the cultivation of intercrops is not economical. At 
this stage sheep can be integrated under rubber. Results of integrating sheep, 
particularly Dorset Horn crossbreds, with rubber have been encouraging and 
acceptable to smallholders since they provide a steady income stream to small­
holders and also feed on the weeds thus reducing the cost of weed control in 
the holdings. 
(d) Poultry - Sheep - Rubber 
This combination in the farming system is the most beneficial to smallholders 
who have limited land and capable o f raising a few heads of sheep and some 
poultry. These animals can serve as important additional income earners. 
Farming practices 
The above farming systems have their own merits and demerits and their benefits 
are now discussed under types of farming practices as follows : 
(a) Cash crops 
Table 8 shows the comparative advantage of planting with the recommended 
crops. Groundnut requires the highest cost input while maize gives the highest 
net return to investment provided all the fresh cobs are sold at the right time. 
Land preparation, labour and fertiliser account for a large portion of the pro­
duction cost. 
(b) Poultry 
Broilers are more suitable than egg production for rubber smallholders because 
o f quick turnover and less management problems. The average production 
cost (Table 9) was slightly higher than commercial but lower than smallholders' 
farms receiving no supervision. The average income per smallholder per year 
is shown in Fig. 1. Although these figures showed lucrative returns, the 
net family income varied among smallholders and time of year. The most 
important factors that could influence income are : 
404 
B u k i t S e t t i n g U l u 
Cheraka U l u T iram 
N o . o f b i r d b a t c h " 1 : 5 0 0 5 0 0 - 1 0 0 0 * 1 5 0 - 3 0 0 
( B a t c h y e a r " 1 ) ; ( 4 ) ( 6 ) ( 5 ) 
T o t a l b r o i l e r s p r o d u c e d : 5 , 9 5 0 1 1 , 9 4 0 4 , 8 0 0 
Av, Income y e a r " 1 ? $ 1 , 9 9 0 $ 2 , 3 8 0 $ 5 3 0 
Fig. 1. Average yearly income from broiler project. 
40$ 
$ Supply of high quality day-old chicks when needed, 
% Supply of unadulterated and high quality feed, 
9 High cost of day-old chick and feed, and 
Q Availability of market and suitable market price. 
Table 8. Production costs and return (%j Hajcrop) from 4 types of intercrops planted during 
the first 3 years of rubber cultivation 
Types of crops 
Items Groundnut Maize Soybean Banana* 
($) ($) (?) ($) 
Land preparation 340 340 340 130 
Planting materials — seeds 450 60 75 160 
Fertilizer 250 440 260 600 
Weedicides 40 40 50 45 
Pesticides 120 30 125 20 
Labour (Family) 600 330 540 25 
Total production cost 1,800 1,240 1,390 1,205 
Yields (kg) 3,500 25,000 1,200 6,274 
cobs 
Price (farm-gate) 75c/kg lOc/cob H-30/kg $ 0 30/kg 
Gross income 2,625 2,500 1,560 1,882 
Profit 825 1,260 170 677 
* Mean of 3 croppings 
Table 9. Average cost of broiler production under smallholdings 
Item (M$) bird- 1 
(a) Day-old chicks 1 00 
(b) Feed : Starter 0-85 
Grower 2 40 
(c) Veterinary medicines 0 10 
(d) Other costs (electricity, water etc.) 0 18 
(e) Depreciation on fixed assets (shed, feed and 
water utensils) 0 • 32 
Total 4-85 
406 
(c) Sheep rearing 
Integrating sheep with rubber can bring direct and indirect benefits to the 
rubber smrllholders. The direct income is through the sale of livestock as 
mutton, and with the crossbred, the potential for wool and pelt is bright. The 
Economic viability of small scale sheep rearing under rubber is given in Table 10. 
Table 10. Estimated cost of production and income for local sheep and 50% Dorset Horn 
crossbred 
Initial Stock plus 1 Stud 
Items 5 10 15 20 
1. *Shed built with looal materials 300 530 780 1000 
(0 7 mr/'earning) (J) (8 4 m s ) (17 5 m") (26 0 m a ) (35 0m») 
2. Fencing if necessary. Cost is Depends on the need and area. More accurate est 
too high. mates can be obtained at each locality 
3. Animal cost ($ 80 for female 
and % 100 for male) 450 800 1100 1000 
4. Cost of planting grass 80 120 160 180 
(if necessary) ( 0 1 ha) (0-15 ha) (0-20 ha) (0 225 ha) 
5. Cost/year (?) 154 276 403 578 50 
6. No. of animal sold 4 11 18 25 
7. **Income after \ \ years a. 62 0 318 0 568 0 831 5 
Income/year (p) b. 266 0 879 0 1486 0 2106 5 
* Include area for new born lamb. 
** Local breed : a. 18 kg @ 3 00 per kilo live-weight "] To deduct the depre­
ciation cost of shed, 
b. Crossbred 50 % Dorset Horn. 1 year 
- 35 kg 
Vet. medicine and 
others. 
The average cost of vet. medicine, salt lick etc. @ ? 2 50 per animal per year. 
CONCLUSION 
The total land area replanted since the launching of the replanting scheme in 1953 
now constitute about 75% of the total registered rubber area. Under the Dynamic 
Production Policy, rubber replanting with higher yielding and earlier maturing planting 
materials together with better exploitation techniques can bring about further improve­
ment in productivity and income of rubber smallholders. Through appropriately deve­
loped technology packages and agricultural support facilities, namely efficient supply of 
407 
farm input6 and extension of farm management techniques for rubber and related non-
rubber activities smallholders' productivity and income base can be broadened and im­
proved. 
The current economic recession and inflation which resulted in the simultaneous 
decline in the rubber price and increase in the cost of inputs and general cost of living 
have bitten hard into the rural sector resulting in migration of rural youth to the cities 
and other growth areas. This continuing rural-urban migration which brings about the 
attendant problems of labour shortages in the rural rubber sector suggests the need to 
continuously reallocate scarce resources and increasingly mechanise various farm opera­
tions in order- to make it more attractive for rural youth to remain behind and operate 
farms as on-going viable enterprises. The rubber-based rural economy thus needs to be 
stimulated with additional economic activities through the IAD to supplement that of 
rubber. 
Being inelastic in nature the rural smallholders' rubber economy cannot be readily 
diversified (nor is it deemed desirable to do so) in the face of the current unfavourable rub­
ber price trend. Thus under this present conditions it is only through the increase in the 
volume of rubber production by yield stimulation both at the old tree phase (3 years to 
felling) and mature phase (Panels C & D) can smallholders' income from rubber be in­
creased in the short-term. In addition, other non-rubber farm activities like intercropping 
and cash-cropping (at the immature rubber phase) and mixed farming (at the mature 
rubber phase) can help broaden the income base of rubber smallholders in the medium 
term. As a long-term measure to increase the productivity and income of rubber small­
holders, effective replanting programme using the best yielding clones and early maturing 
planting materials and best management techniques together with the latest technology 
in integrated farming of inter-crops, cash crops and animal husbandry at the various 
phases of rubber planting should be advocated. 
Thus at all the phases of rubber planting there are important investment decisions 
to be made. Methods of discovering the optimum combination in the farming-mix at 
the various planting phases and the scale of operation will ensure the long-term viability 
of farming rubber on the smallholding model. The crops grown and animals reared by 
such smallholders should fulfil the subsistence as well as the cash needs of smallholders 
and supplement their cash income from rubber. All these aspects should form part 
and parcel of the processes in the Dynamic Production Policy. To facilitate and accelerate 
these processes various subsidies and credits need to be instituted and prudently managed 
and the profits reinvested for sustained and long-term benefits of rubber smallholders in 
this country. 
408 
STAGES OF DEVELOPMENT AND GOVERNMENT POLICY IN 
RUBBER SMALLHOLDING AGRICULTURE 
BY 
COLIN BARLOW A N D S . K . JAYASURIYA 
(AUSTRALIAN NATIONAL UNIVERSITY, AUSTRALIA.) 
Smallholdings have been a major and dynamic component o f the natural rubber 
industry from its very early years (Table 1). By the I980's they covered about three quar­
ters of total world rubber lands, and contributed over two thirds of total output. The 
balance of area and output came mainly from large estates (footnote 1). This great expan­
sion of smallholdings contributed significantly to economic growth in the main producing 
countries, and gave an important source of cash income to participating farmers. 
In light o f this vigorous advance, it is somewhat paradoxical that the economic via­
bility o f rubber smallholdings has often been underestimated by government. Thus 
in the early years the burgeoning expansion of their planted area and output surprised 
the British and Dutch colonial regimes, whose policies generally favoured the estate mode 
of organisation conducted primarily by their own expatriate nationals. Again in the 
1980's, the ability of smallholdings to cope with economic change is widely questioned in 
official circles (Malaysian Rubber Research and Development Board, 1983 ; Mubyarto, 
1983), and improvement policies frequently assume an estate type of organisation to be 
superior. 
The approach in this paper is to analyse the process of smallholding growth over 
time by Stages, beginning with the initial Stage o f expansion into rubber from a basically 
subsistence economy, and proceeding to the final Stage of rubber in a highly commercia­
lised environment including substantial manufacturing and services sectors. The factors 
effecting economic growth and smallholder welfare at each Stage are investigated, and the 
main government policies impinging on these developments reviewed. An attempt is 
made to identify policies chiefly responsible for facilitating or impeding growth. 
The assessment resulting from this process is thought to negate the view of those seeing 
little future for smallholdings, which appear well capable under appropriate circumstances 
o f adjusting to changed economic conditions. While such adjustment has not always 
occurred, this may in fact be partly attributed to unsuitable official policies. Given cer­
tain policy alterations, it is considered that further dynamic responses of the smallholding 
sector can be anticipated. 
The Stages 
The three Stages of development are each characterized by rather different relative 
endowments of land, labour, capital and other factors, by consequently different prices 
of these resources, and by a different basic technology. Indeed the latter may itself be 
regarded as having responded dynamically to price changes since the previous Strge, 
through the actions of research workers in attempting to improve it in directions econo­
mically profitable to the farmers concerned. This is the expected response of technology 
409 
Table 1. World rubber areas, 1900 — 80 ('000 hectares)" 
Indonesia Malaysiab Thailand Sri T .anIra 
Year Smallh.c Estates Total Smallh. Estates Total Smallh. Estates Total Smallh. Estatef Total 
1900 
1910 6 104 110 
109 352 461 
729 573 1,302 
731 626 1,357 
2 
219 
883 
1930 729 573 ,  483 763 1,246 
1940 731 626 ,  547 843 1,390 
S 1950 643 795 1,438 
° 1960 766 783 1,549 
1970 1,077 647 1,724 
1980 1,862 466 2,328 1,206^ 492 1,698 1,579 46 1,620 122 106 228 
Source : International Rubber Study Group (190 — 80) ; sources quoted by Batlow and Drabble (1983). 
Notes : (a) This table is incomplete, and contributions of relevant data are welcomed. 
(b) Peninsular Malaysia only. 
(c) Mainly in the Outer Island of Sumatra. 
(d) Including 391,000 hectares under block new planting schemes. 
in the induced innovation framework of Hayami and Ruttan(1971), which empirical tests 
have shown to be broadly correct (footnote 2). The prices of resources and their relative 
usage by smallholders and estates in producing rubber in each stage distinguished in 
Malaysia are detailed in Table 2. 
Stage I 
This is the Stage of emergence from an essentially subsistence agriculture, practiced 
at the frontiers of cultivation where land is abundant and labour relatively plentiful. The 
market for capital is both poorly developed and fragmented, and credit thus extremely 
scarce. Under such circumstances the real prices of land and labour are low, whilst that 
of capital is very high (Table 2, footnote 3). 
Table 2. Prices, yields and resource use In smallholding and estate rubber production, Malaysia, Stages I, 
II, and III (prices and costs all In 1978 M%Y 
Smallholdings 
Stage I 
(1922) 
Stage II 
(1964) 
Stage III 
(1981) 
Prices 
Rubber (MS/kg, f.o.b.) b 
Land (MS/lha) 
Labour (MS/day, all in) 
Capital (per cent) 
Wage : interest rate ratio 
1-50 
904 
4-52 c 
30 — 4 0 e 
12-9 
2-58 
3,872 
7-74c 
25 — 30 
2 8 1 
1 80 
13,000 
11•66 d 
15 — 20 
66-7 
Yield (kg/tapped ha, all grades) 
Resource use 
Labour (MS/ha) 
Capital (M$/ha) g 
518 f 
518f (97)h 
18 (3) 
1,173 
1.478 (82) 
315 (18) 
1,300 
1,166(85) 
205 (15) 
Total (MS/ha) 536 f 1,793 1,371 
Total unit costs 
(MS/1000 kg/ha) 1,035 f 1,529 1,055 
Physical labour 
(man days/a lOOOkg/ha) 222 f 163 77 
Estates (1963) 
Prices 
Rubber (MS/kg, f.o.b.) b 
Capital (per unit) 
1-87 
6 — 7 
2-72 
8 — 10 
1-92 
12 — 15 
Yield (kg/tapped ha, all grades) 300 1,308 1,450 
Resource Use 
Labour (MS/ha) 
Management (MS/ha) 
Capital (MS/ha) 
638(58) h 
259(24) 
193(18) 
1,082(65) 
192(11) 
398(24) 
1,057(60) 
229(13) 
476(27) 
Total 1,090 1,672 1,762 
Total unit costs 
(MS/lOOOkg/ha) 3,633 1,278 1,215 k 
Physical labour 
(man days/lOOOkg/ha) 503 108 62 
(Continued over) 
411 
(Table 2 Continued) 
Sources : (of date for 1922 and 1963/64): sources quoted by Barlow and Jayasuriya (forthcoming) 
(of data for 1981) : sources quoted by Barlow (1984). 
Notes: a. For " progressive " smallholdings and estates (which account for somewhat over one half 
of the total smallholding area and two thirds of the total estate area). To convert figures 
to 1978 P, nominal values are weighted according to an especially constructed consumer price 
index with 1978 = 100,1922 = 16-6,1963 = 58-8,1964 = 5 8 1 a n d l 9 8 1 = 130 (for sources, 
see above). 
b. For the balance of grades reported from smallholding/estate rubber output. 
c. Returns to tappers getting 50% of the crop at the yields shown, at concurrent task sizes, 
and at relevant farm gate prices. 
d. Valuing smallholding labour at the same rate as estate labour (average 1981 estate earnings 
per day, plus 20% for perquisites). 
e. Nominal, not real interest rate. 
f. South Sumatran physical inputs and outputs for 1981, valued at Malaysian smallholding 
prices for 1922. N o physical input figures were recorded for Malaysian smallholdings in 
the early 1920's, but the basic technology of smallholding production in South Sumatra of 
the 1980's, was the same. Technical efficiency could well have increased since that time, 
however, and absolute inputs of labour in 1922 could have been considerably higher than 
indicated (see also footnote 4). Physical output figures from other sources however, confirm 
yields somewhat above 500 kg/tapped hectare. 
g. All labour, including any management input involved. 
h. Figures in brackets are percents of total resource cost in each category. 
i. Costs of material inputs (fertilizers, agrocides, transport, etc), which may be regarded as 
capital. The costs of developing young rubber to maturity are not included. 
j . Prices of labour and land were very similar to those for smallholdings. 
k. Although unit costs on estates in 1981 were higher than on smallholdings, the unit value 
of rubber output was also higher (see f.o.b. rubber prices), so that net profit differed little. 
The technology of clearing jungle or scrub and of subsequent maintenance depends 
on simple hand tools, and the technology of rubber production centres on unimproved 
seed obtained free from neighbouring estates, or cheaply from itinerant traders. These 
technologies are actually well adapted to the ruling situation of plentiful labour and land. 
They are also divisible in nature, being applicable on a very small scale and further exhi­
biting no economies, as scale increases. There is thus little difference in their cost or 
profit between large and small units (footnote 4). The circumstances of this Stage I 
production phase are also such that the smallholder response to economic stimuli can be 
primarily endogenous, with no vital need for exogenous help as occurs in Stages II and III. 
While there are substantial economies of scale in the end-processing of rubber into various 
crepes and other products, these are secured through the medium to large scale processing 
factories set up by the private trading network, and juxtaposing effectively with the small­
holding production sector. 
Small wonder that with a cash cropping technology so well suited to their conditions, 
smallholders with their high marginal utility of cash took up the rubber cultivation tech­
nology first introduced by the estates with such enthusiasm, and expanded so dynamically 
(Table 1). The actual conditions of this cultivation varied somewhat, there being on the 
one hand the extremely land-using and capital-saving operations characteristic of the 
Indonesian Outer Islands, where land was virtually free and farmers fitted rubber culti-
4 1 2 
vation into a swidden agriculture which still involved the cultivation of subsistence crops 
(Whitford, 1931). There was on the other hand the somewhat less land-using and capital-
saving system true of smallholdings in early Malaysia, where land alienation was more 
restricted and Chinese smallholding families in particular often earned cash wages which 
enabled limited material purchases. The predominantly labour-using balance of resource 
use in production under the Stage I technology is indicated in Table 2. The early Sri 
Lankan smallholding advance was similar to that in Malaysia, while the rather later Thai 
development parallelled that in Indonesia. 
Under all these conditions however, and also indeed under the more management 
and capital-intensive circumstances of the estates (Table 2), the yields per tapped hectare 
and economic returns to the major factors employed were quite similar (Barlow 1978, 
footnote 4). Thus in writing of the circumstances of the time, de Silva (1982) aptly 
commented that " largeness in the scale of plantation operations is not an agronomic 
necessity ; plantation crops can be grown just as efficiently on smallholdings". In fact, 
during the Great Depression of the 1930's it became apparent that the huge areas o f 
Outer Islands smallholdings with their low fixed costs and alternative subsistence activities 
were more economically viable than the corporate estates, to which they now posed a 
distinct threat (Bauer, 1948 ; Thee, 1977). 
Stage II 
In this Stage substantial commercialization of the economy has occurred, although 
agriculture is still the dominant economic sector. With the growth of population and o f 
competing demands upon them, both land and labour have become scarcer and therefore 
more expensive. This same process, however, has reduced the previous fragmentation 
of the capital market, and credit has become cheaper. The roads and other basic infra­
structures of rural areas have been considerably enhanced, especially reducing the cost 
o f transport but also improving information flows. 
Much progress has also been made in modifying rubber technology, which has now 
moved from its primitive first stratum (footnote 5) to a second stratum involving mecha­
nized techniques for land clearing and much higher yielding planting materials responding 
positively to fertilizers in the phase of production. More sophisticated processing methods 
have also been developed. This modified technology reflects changed resource endowments 
in saving land and labour, and making greater use of capital. It also involves considerably 
more skill in effective application, and is thus management-using. The changed prices 
and resource use under this modified technology are again illustrated in Table 2. 
Yet while this second stratum technology is still divisible in that mechanized tech­
niques can be applied on small units through contract services, while all other maintenance 
and production components can be employed by farmers themselves on a very small scale, 
some major problems of adjustment to it have emerged. Thus although the advance 
through replanting or even newplanting into the higher level technology is usually a pro­
fitable long term investment for him, the smallholder does not have the cash needed to 
hire contract services for clearing, or to purchase the high yielding materials and their 
complementary material inputs (footnote 6). The private capital market, while better 
developed then before, is still nowhere near providing long term credit of this nature. 
413 
In such circumstances smallholdings cannot advance without exogenous financial sup­
port supplied by government. In contrast, however, there is little problem for small­
holdings in adjusting to better processing technology, which is easily undertaken by large 
scale firms with ample capital resources in the juxtaposed rubber trading network. 
In this situation the move of farmers in Stage II to a more appropriate production 
technology depends not only on resource price changes and technological development, 
but also on exogenously provided replanting and newplanting grants. These were sup­
plied in Malaysia and Sri Lanka of the 50's, and in Thailand of the 60's (Association of 
Natural Rubber Producing Countries, 1976). Financial support was not provided in 
Indonesia until the 1970's, however, and then in a very restricted manner. (Koestono 
and Gwyer, 1976). Indeed, in the Outer Islands in the 1980's, the vast majority of far­
mers still use the primitive first stratum technology, albeit with some increase in technical 
efficiency from earlier years (Barlow and Muharminto, 1982). Yet it should also be 
added that the move into Stage II of relative resource endowments in this remote region 
has even now not been completed. In many areas land is still quite abundant although at 
increasing distances from rivers and roads, and in the absence of alternative enterprises 
most farmers have profitably continued to expand with the old technology in their tra­
ditional land-using manner. 
It should further be noted that in Stage II the situation of rubber estates differs from 
that of smallholdings, in that the access of the former to adequate finance generally renders 
an endogenous adjustment to new technology through replanting quite possible (foot­
note 7). The estates have a further endogenous advantage in that their management 
structure enables them to apply new technologies efficiently at field level without outside 
advice. These features are pecuniary and management economies of scale, which were 
not relevant with the first stratum technology. Yet given suitable endogenous support, 
the rubber smallholding in this stage can still adjust well and remain very economically 
viable, as demonstrated in all countries where official replanting grants and advice have 
been supplied. 
Stage HI 
In this stage commercialisation of the economy has advanced much further. Agri­
culture is no longer dominant, being increasingly dwarfed by manufacturing, services, 
and other sectors. The prices of land and labour have again advanced rapidly, the latter 
being accompanied by widespread migration from rural areas to what are often more 
remunerative urban occupations. The cost of capital however, has continued its down­
ward trend. Infrastructures are further improved, and rising levels of general education 
in particular have both demanded a much improved flow of information and imparted 
superior ability to assess it. This is really the stage of " industrial revolution ", and 
amongst natural rubber producing countries has so far only been reached by Malaysia. 
Technology has once more responded to the changed resource endowment, although 
in reflection of diminishing returns its savings of scarce resources and additions to yield 
are not so dramatic as between Stages I and II. Bigger machines for clearing have been 
developed, and chemical and fertilizer imputs have been improved. Somewhat higher 
yielding varieties with other desirable attributes have been introduced, and methods of 
414 
processing have been enhanced. The bias of technology is again land and labour-saving 
and capital-using, with a further management-using element. The prices and resource 
use in Stage III are once more detailed in Table 2. 
In this Stage, and despite the further development of the capital market, smallholders 
still require exogenous financial support in moving to renew their productive stock of trees 
through replanting. It is also very evident, although there were already signs of this in 
Stage II, that the management skills to apply improved technology effectively are only 
possessed by some more progressive smallholders. The balance of farmers cannot really 
manage the technology, and in their consequent failure to replant face a situation of low 
and diminishing returns. Most members of such less progressive families find they can 
secure higher earnings from unskilled employment in the modern sector, and in what is a 
major feature of such a commercialising economy accordingly abandon their holdings 
in favour of this alternative. It is further evident from Malaysian experience of this 
Stage that with the rising wage of labour, landowners no longer find it profitable to employ 
hired workers to tap their holdings for them. While the institution of sharing has proved 
flexible to change in other contexts (Chew,1981), this particular situation with its tightening 
labour market appears to contain very real difficulties of adjustment to a more dynamic 
system (footnote 8) ; substantial areas of holdings run by owner-employers of workers 
have accordingly been closed down. In Malaysia of the 1980's, perhaps half of the indi­
vidual smallholding area of 800,000 hectares is owned by less progressive farmers or small 
landlords, and considerable abandonment of rubber production in this portion has already 
taken place (Malaysian Rubber Research and Development Board, 1983). 
Against this, however, progressive and well informed owner occupying smallholders 
making regular usage of replanting grants have been earning returns per day from rubber 
easily commensurate with those from alternative enterprises (Barlow, 1984). Such families 
rarely earn all their income from rubber, however, since their planted areas are too small 
for this. Thus in another important feature usually parallelled in other commercialising 
agricultural economies (Shand,1983), and departing from that of the rather more specialist 
farms of Stages I and II, they usually have several distinct income sources including some 
of a non-agricultural nature. Such dovetailing of rubber with other activities not only 
stabilizes income, but also improves cash flow and enables further purchases of comple­
mentary and labour-substituting inputs in rubber production. Yet while the more dynamic 
elements of this progressive group might also be expected to expand on to the failing and 
abandoned rubber areas of less progressive elements and small landlords, in a process 
characteristic of agricultural development elsewhere, the constraints on land transfer 
outlined below make widespread change of this kind impossible. 
Once more the marketing and processing network in Stage III has adapted well to 
further economic change, undertaking further capital-intensive and labour-using innova­
tions coupled with quality improvements. The estate rubber sector has again continued 
to thrive in this changed environment, although its comparative advantages with lower 
interest rates and better information have declined with external economies in these items 
promoted by commercialisation. 
415 
Government policies 
The progression of rubber smallholdings through the three Stages has been sub­
stantially influenced by various government policies. The most major of these policies 
are now examined, and their effects on economic growth and the welfare of smallholders 
assessed. 
Research and extension 
The provision of rubber research facilities may be regarded as one of the most econo­
mically beneficial measures taken by government in all rubber producing countries (Pee, 
1977). It is manifestly impossible for individual small farmers to undertake substantial 
generation of new technology themselves, and in these circumstances government inter­
vention is the best means of bridging the private-social divergence (footnote 9). The 
early work of Thomas Ridley, the subsequent efforts of Dutch scientists in Indonesia, 
and the postwar efforts of researchers in the publicly funded institutes of the main rubber 
producing countries have all made important contributions to more economic rubber 
production, through enabling a steady improvement in output and quality, with major 
concomitant savings in usage of scarcer resources (Table 2). 
There is nonetheless evidence that especially in earlier research pertinent to the tran­
sition from Stage I to Stage II, there was some prejudice toward the relative resource 
endowment of estates with their lower price of capital. The technology generated at that 
time consequently had a more marked capital-using bias than was appropriate for the 
economic position of smallholdings, and appears to have reduced the profitability of their 
transition to Stage II technology in Malaysia (Barlow and Jayasuriya, forthcoming). 
While this Stage is past in Malaysia, whose prices now dictate similar technology for both 
rubber sectors, it is suspected that a similar bias operates against Indonesian smallholdings, 
which are only now facing the need for transition to Stage II. In that context there is a 
weighty economic argument for generating a more capital-saving technology specific to 
smallholding conditions (Barlow and Muharminto, 1982). Again in the current Stage ILT 
situation of Malaysia, there are good arguments for further concentrating on labour-
saving technology, so as to maintain profitability in face of expected increases in the real 
price of labour. These arguments apart, there is certainly a well justified case for further 
heavy emphasis on supporting public rubber research. 
The bringing of research technology to the notice of farmers through advisory ser­
vices is another well justified public activity, especially pertinent in the poorly developed 
information situation of transition to Stage II, and particularly complementing the crucial 
process of first replanting with high yielding materials. Despite its obvious importance, 
however, it is very hard to organize. Except in the notably successful recent case of 
Thailand, where an effective programme concentrating on crucial aspects of technology 
transfer in Stage II appears to have been successfully mounted (Phit Panyalaksharla and 
Blencowe, 1976), such extension has rarely had very wide impact. Thus in Indonesia 
in the late 1970's, rubber extension hardly existed outside the intensive rubber develop­
ment schemes (Barlow and Muharminto, 1982). In Malaysia the recent social emphasis 
of extension, together with its focus on restricted groups, has meant that relatively few 
416 
rubber farmers have had much assistance in adopting new technology (Malaysian Rubber 
Research Development Board, 1983). More broadly directed and technically oriented 
extension in these contexts can undoubtedly assist adoption, and is thus most desirable. 
Replanting grants 
The exogenous organisation by government of replanting grants has been of vital 
economic significance enabling independent smallholders to make a lumpy investment 
in higher yielding materials. The provision through an export tax transfer mechanism 
of necessary material inputs in cash or kind, together with additional cash to cover the 
cost of labour inputs (footnote 10), has successfully bridged the private-social divergence 
attributed to imperfect capital markets, which even in Malaysia of Stage III are not suffi­
ciently developed to provide credit for such a purpose. This approach to funding small 
plantation investment has great advantages in following what we have previously termed 
a " dispersal" strategy (Barlow and Jayasuriya, 1984). In contrast to the more specific 
and locationally concentrated situation of the land development schemes discussed below, 
it is available to all progressive elements throughout the farming community, and thus 
has maximum economic impact (footnote 11). 
As already mentioned, the economic benefits of replanting grant programmes have 
been amply demonstrated in all major natural rubber producing countries except Indonesia. 
It is strongly felt that where it is economically desirable, existing programmes should be 
maintained and new programmes extended into untouched areas, although the coverage 
of the measures must naturally depend on what material resources are available. Thus 
in contrast to the liberally funded programmes of Malaysia, Indonesia and Thailand, 
Barlow and Muharminto (1982) in the Indonesian context have suggested very restricted 
planting assistance, largely comprising selected seedling planting materials accompanied 
by advisory help in establishing them. Such materials are regarded as a " minimum 
technological base", which is then susceptible to further improvement through other 
complementary inputs as more cash becomes available. 
Difficulties have of course emerged in all replanting grant programmes, and basically 
concern the failure of less progressive farmers to make use of this investment opportunity. 
In Stage II where there are normally no major alternatives to rubber cultivation, these 
are best dealt with by special follow-up measures to assist adjustment (footnote 12). In 
Stage III they are better handled through freeing the land market in the manner, dis­
cussed below, together with direct training and other assistance to facilitate the transfer 
of persons who cannot benefit to urban areas. 
Land use regulation 
This has been applied in most rubber producing areas since the beginning of Stage I, 
normally with some form of social or political justification. 
Thus the colonial government of early British Malaysia employed what de Silva (1982) 
has termed " legal-institutional barriers " to reduce the Stage I smallholding expansion 
on to land suitable for estates and in Dutch Indonesia even more rigid measures effectively 
banished smallholding development to remote peripheries in the Outer Islands (Geertz, 
417 
1953). On the other hand, the further land use measure in Malaysia of reserving wide 
areas for Malays only under the Malay Reservations Enactment of 1913 did at least have 
the merit of preserving for indigenous peoples a major stake in the rubber economy of 
their own land. Another practice during the early colonial period was the reservation 
of land titles for particular crops only, and this was again largely applied against small­
holding interests to reduce their rubber cultivation in competition with estates (Lim, 1977). 
But the subsequent persistence of these regulations, often with significant political 
overtones, has usually had adverse long term effects in " locking " farmers in to economic 
activity not desirable in a dynamically changing economy. Thus the location of most 
Indonesian rubber smallholders in remote areas with few linkages to a wider economy 
has made their desirable transition from Stage I to Stage II that much more difficult. 
The existence in Malaysia of reservations and the specified cultivation of certain crops, 
together with bureaucratic difficulties in land transfers have rendered far harder the con­
version of lands to more economic purposes. Such conversions notably concern the 
sale of abandoned rubber lands to more progressive smallholders. Especially in Stage 
III with its much wider economic alternatives including wage employment in the urban 
sector, there is little justification for preserving the position of particular groups in agri­
culture through such regulations. 
It is accordingly believed that as development proceeds, the manifold land use regu­
lations in rubber smallholding sectors should carefully assessed, with a view to removing 
what are in fact serious constraints on more economic operation and long term small­
holder welfare. 
Land development schemes 
From the late 1950's in what was essentially the transition to Stage II, the newly inde­
pendent government of Malaysia commenced a substantial policy of organizing rubber 
improvement through new group land development schemes. In contrast to the " dis­
persed " policy of replanting grants, new planting in these schemes was focussed on conti­
guous blocks of individual settler parcels, organized in several ways. These ranged from 
small schemes of up to a few hundred hectares, with relatively loose guidance by manage­
ment and no ancillary services or special infrastructures, to generally very large schemes 
of one thousand or more hectares with tight management, central processing and other 
facilities, and well developed infrastructures including settler housing (Lim, 1976) (foot­
note 13). The latter case was epitomized by the great schemes of the Federal Land Deve­
lopment Authority, whose concept was one of arranging management and services so 
as to secure economies of scale in these items, while at the same time retaining quasi-
independent settler units for the production phase. Indeed, the institutional principle 
of such schemes may be described as that of a " guided yeomanry ", planned as combining 
the advantages of central supervision and certain other facilities at the top with inde­
pendence at the bottom (footnote 14). 
All the early land development schemes were launched in a situation of widespread 
landlessness within cultivated areas, where an official investment of capital to improve 
previously inaccessible land appeared both socially and privately profitable. Such schemes 
were often favoured by government over dispersed development, owing to their greater 
visibility and apparent amenability to central control. 
418 
Yet almost from the beginning, the smaller less organised schemes encountered mount­
ing problems arising from uneven development. Commonly they emerged as a patch­
work, with all too many bad parcels interspersed amongst good in direct reflection of the 
differing motivations and abilities of participating farmers. In contrast to this, the large 
coordinated schemes of the early years appeared to be developing well, and indeed to 
justify their far more elaborate nature and intensive supervision. Their approach was 
accordingly emphasized further in new developments (footnote 15). 
Then during the 70's the large schemes too entered a period of increasing trouble, 
partly owing to the wider developments of Stage III. Thus the departure of younger 
people left an ageing population, increasingly unable to exploit their parcels efficiently. 
There was also, however, an emerging conflict between settlers and management, where 
the former appeared to regard their individually allocated blocks as their own to be ma­
naged as they wished, especially where they had substantially repaid their debts (footnote 
16). Management, on the other hand, continued to press for certain husbandry practices 
which the settlers were not prepared to accept (footnote 17). This conflict undermined 
the whole process of technology transfer, leading to lower than expected incomes and 
thus to a further acceleration in departure of the younger people. Yet under the pre­
vailing tenure arrangements of these schemes, it was again not possible for progressive 
elements to expand their activities through purchase of adjoining parcels. Thus there 
was once more a prospect of settlers being locked into a situation of declining productivity, 
with economic returns much below those originally anticipated. It was largely owing 
to this that the recent group reviewing Malaysian rubber policies recommended that 
" all new land schemes should from now on be operated as commercial estate enterprises ", 
and that the individual parcel concept be abandoned (Malaysian Rubber Research and 
Development Board, 1983). 
While Indonesia, as the only other country with major group rubber land develop­
ments, did not begin these substantially until the mid 1970's, similar problems of uneven 
progress have already appeared in smaller less coordinated schemes, and been countered 
by more intensive supervision (World Bank, 1974 — 82). On the other hand the large 
co-ordinated " nucleus estate " schemes of Indonesia, which are now the preferred official 
method of smallholding development (Mubyarto, 1982), have barely reached the pro­
duction phase, and thus not faced any real problems in settler-management relations 
(footnote 18). Again the lack of alternatives to rubber in the Indonesian Outer Islands, 
where Stage II has hardly begun, means that outmigration will not be a difficulty in the 
near future. 
Yet the problems of uneven development and management relations would already 
seem quite widespread enough to raise serious questions about the long term viability 
of group land development schemes as an institutional mode of smallholding improvement, 
especially in the more commercialized environment of Stage III. While governments 
recognizing these difficulties may thus be predisposed to recommend the replacement of 
this mode by estates, preferably under the auspices of some government instrumentality 
(footnote 19), the latter condition especially should also be questioned. This is in a 
situation where the role of government in directly administering development may already 
be seen as overextended and counter productive. 
419 
The commercial estate development of new land is certainly one solution, where this 
is economically desirable. Another alternative is the opening up of such land by inde­
pendent smallholdings. Given suitable exogenous support, with flexibility for expansion 
or contraction of units according to economic circumstances, this would seem at least as 
viable an economic approach which could have important social advantages (footnote 20). 
Marketing improvement 
One sphere of active government intervention in all countries has been that of domestic 
rubber marketing. This reflects a widespread suspicion by all sections of the community, 
including some academics (e.g., Wharton, 1962), that pricing is manipulated by various 
middlemen to their own benefit in excess profits, and to the farmers' detriment in lower 
prices. Such suspicion has arisen despite the fact that the trading network generally 
seems to have adjusted quite efficiently to resource price and technology changes at suc­
cessive Stages. 
Although some " imperfections " in the domestic market undoubtedly exist, it is our 
impression that widespread exploitation of this nature has yet to be substantiated through 
rigorous empirical study (Lim, 1968). Thus the trading network in most countries appears 
to include sufficient competitive elements to prevent monopsonistic behaviour. It might 
also be hypothesized that the degree of exploitation would be greatest in Stage I, where 
the market for rubber products may be seen as most fragmented, with little price informa­
tion reaching producers. Even in this stage, however, it is apparent that except in remote 
places with small rubber areas which barely support one dealer the trading network is 
generally quite competitive (footnote 21). 
Certainly it also seems fair to comment that the manifold efforts made by govern­
ment in all rubber producing countries have rarely had much positive influence on the 
farm gate rubber price. In most cases, government market intervention has been effected 
at considerable cost in scarce managerial resources, and following expensive mistakes 
over a period of several years has usually been withdrawn (Thomas, 1957 ;• Barlow and 
Lim, 1965; Abdullah et al, 1972). In instances where government agencies have been 
granted some monopoly, it appears that farm prices have actually been depressed below 
levels which would otherwise have obtained in the " free " (albeit imperfect) market 
without such intervention (Thomas and Panglaykim, 1973). The most successful inter­
ventions over the years appear to have been those aiming to improve quality through 
small local " group processing centres " (Sribo Chaiprisit and Lim, 1976), although even 
in this case the primary continuing role has actually been taken by centres sponsored 
under private rather that government auspices (Barlow, 1978). 
Intervention in domestic marketing is thus regarded as a normally expensive measure, 
justified increasingly less as development proceeds into Stage III and deserving very careful 
prior assessment. Where it is thought warranted, the low cost alternative of attempting 
through appropriate measures to push the private sector towards greater competitiveness 
without expensive direct intervention should certainly be considered (footnote 22). Quite 
often indeed, much can be done by abolishing such constraints as licencing provisions 
420 
on private dealers, which effectively bolster their position by preventing the free entry of 
other competitors into the market. Certainly as a rubber economy moves into the com­
mercialized environment of Stage III, direct intervention is rarely likely to be worthwhile. 
Governments have also intervened in the international marketing of rubber, notable 
in the Stevenson and International Rubber Regulation Agreements of the 1920's and 
1930's to restrict the supply of rubber, and in the International Natural Rubber Agree­
ment of 1979 to iron out price fluctuations. The adverse long-term influence of output 
restrictions in reducing the competitiveness of natural rubber, and the discriminatory 
application of such restrictions to smallholdings, have already been well established (Bauer, 
1948 ; Lim, 1977). While the current international buffer stock scheme is particularly 
helpful to smallholdings to the extent that it reduces price fluctuations, the price of achieving 
this goal must also be weighed against alternative more direct means of stabilization (foot­
note 23). 
General measures 
One major group of more general economic measures concerns improvements to 
infrastructures, and especially to roads and other communications. While progress in 
this aspect to some extent parallels the Stages of rubber development, there have been 
considerable differences between countries within given Stages. Thus there was a large 
contrast in Stage I between the almost unserviced rubber growing regions of the Outer 
Islands, and the much better infrastructures accompanying and helping to explain the 
faster early growth of smallholding rubber planting in Malaysia and Sri Lanka. Again, 
the special emphasis on infrastructural improvement in Malaysia of the late 1950's and 
early 1960's much assisted the wide early success of replanting and other rubber improve­
ment measures in that country (Ness, 1967). 
Infrastructural investment promotes major external economies in rubber production, 
most notably in lowering transport costs, but also in a range of associated aspects already 
mentioned as accompanying commercialization. Substantial emphasis on such invest­
ment by government is accordingly favoured as a further major way of promoting small­
holding advance. 
The other major group of overall economic measures effecting rubber essentially 
comprise fiscal policies. These involve taxes on agricultural exportables including rubber, 
together with exchange rates, import duties, quotas, and price supports directly intended 
to improve the profitability of manufactured goods and food crops. Heavy taxes were 
quite justifiably applied to rubber in Stage I and early Stage II, when it was major amongst 
export items and one of the few sources of income to government, but their continuation 
during the later part of Stage II and in Stage III, when rubber is usually a declining industry 
should be seriously questioned (footnote 24). Indeed, we believe we have already de­
monstrated in our companion paper (Jayasuriya and Barlow, 1984) the undesirable impli­
cations for the whole economy of the fiscal measures towards rubber in Stage II of Sri 
Lanka. It is thus believed that fiscal policies towards rubber and other agricultural ex-
portables should be carefully reassessed, especially in light of the generally poor economic 
performance of the preferred alternatives, with a view to removing what appear to be 
very serious distortions in several national economies. 
421 
CONCLUSIONS 
The great expansion of rubber smallholdings over the first 80 years of this century 
illustrates the capacity of small farmers and the rural trading network which serves 
them to respond dynamically to appropriate economic incentives. This development 
may be usefully characterized into three distinct Stages, which have implications for the 
nature of development taking place. During this development towards commercializa­
tion, land and labour become increasingly scarcer and capital relatively cheaper, while 
the nature of technology basically follows this change in resource relationships. It is 
thought somewhat paradoxical that despite illustrating their capacity for change, small­
holdings have generally been underrated by governments, which have looked to the estate 
model as a more suitable form for any development or expansion. 
The argument in this paper is in part that of neoclassical free market economist, but 
it goes further than this. Thus on the one hand it is felt that particularly with the move 
into the full commercialization of Stage III, more rein must be given to the great potential 
of the progressive small farmer element. This can partly be done by removing various 
distorting measures of government which constrain this element, and at the same time 
seriously reduce the capacity of the rubber sector as a whole to respond to economic 
change. 
These distorting measures, which may at one time have had good political or social 
justification, include regulations on land use, inflexible institutional forms of land deve­
lopment, costly interventions in marketing, and various fiscal policies serving to stimulate 
other sectors at the expense of rubber and other prime exportables. 
It is also recognized, however, that the smallholding sector is inherently not able 
to respond to economic change without exogenous help, and requires strong positive 
assistance by government in certain key spheres. These comprise research and extension, 
planting grants, and infrastructures, all of which are probably most crucial in enabling 
the transition from Stage I to Stage II, but remain necessary in some degree as commer­
cialisation of the economy continues. Removing expensive and unnecessary government 
intervention in other spheres should enable these positive aspects of intervention t o be 
more effectively undertaken. 
It is believed that given a re-orientation of policies in the manner suggested, a much 
more vigorous smallholding response to change can be expected, including a more rapid 
transition to more advanced Stages in the different natural rubber producing countries. 
FOOTNOTES 
1., Most were corporately or publicly organized, and covered well over 1,000 hectares. There was also 
a minority of smaller proprietary concerns (Barlow, 1978). 
2. Thus the resource-using bias of new technologies will match the relative prices of these resources to 
farmers. 
3. In a subsistence agriculture there is little demand for capital, and thus no proper market for it. In 
contrast, the estate sector of the early natural rubber industry had access to the highly developed 
stock markets of Europe. 
422 
4. Although the resource cost in producing rubber on smallholdings Stage I is shown as being far lower 
than that on estates (Table 2), the smallholding figures actually refer to South Sumatra in 1981, by 
which time there had probably been a substantial increase in technical efficiency, even with the primi­
tive Stage I technology. 
The yields on estates in Stage I appear to be somewhat lower than on smallholdings (Table 2), and 
confirmation of this is provided from other available data (Barlow and Jayasuriya, forthcoming) 
5. Technology is somewhat crudely but conveniently regarded here as proceeding to higher more tech­
nically and economically efficient levels by successive " strata ", each of which represent the emergence 
of a new set of innovations (and their underlying production functions). In fact, the progress to 
higher levels tends to be a continuous affair. 
6. Although the production technology itself is divisible, there is considerable lumpiness in the invest­
ment in replanting, in that owing to shading and other agronomic problems it is usually not possible 
to consider the replanting of less than about O S hectares. 
7. While various income tax concessions and even direct grant provisions have been made to encourage 
estates with replanting, it is apparent that they can also undertake replanting investment using their 
own privately generated funds, or those borrowed on the private capital market (Barlow, 1978). 
8. In the tight labour supply situation with their consequent ease of moving to new jobs, workers had 
short term horizons in particular jobs, and purposively extracted maximum short-run yields from 
the trees, thereby seriously damaging their future productivity. On the other hands the hierarchical 
estate arrangemrents which enabled profitable large scale land ownership were not applicable in this 
small context. 
9. Although public facilities were also provided for the estates, this was not really necessary in their 
case, as demonstrated by the highly successful Dutch cooperative research stations organized by the 
estate companies (Dijkman, 1951). 
10. Thus in most cases replanting (and research) have been financed by a " cess " component of the export 
tax, which is then placed in a fund to be subsequently disbursed as indicated by a special administering 
agency (Edwards, 1970 ; Association of Natural Rubber Producing Countries, 1976 ; Booth, 1980). 
11. Under these conditions the more progressive farmers, who can be assumed to be the better managers 
with higher marginal value products to resource investment, will all have the opportunity cf secu­
ring grants. This contrasts with the locationally concentrated land developments, where a somewhat 
arbitrarily selected group of fanners containing both good and bad managers receives assistance. 
The total marginal product to investment will-be less in this latter case. 
12. Thus replanting grants may be increased for the smallest holdings, which may also receive other direct 
welfare assistance. 
13. In Federal Land Development Authority rubber schemes the settlers arrived about half way through 
the immaturity period, some time after the initial clearing and planting of land which was undertaken 
by contractors. They then worked under general supervisors on maintenance activities, and only 
commenced specific activities on their allocated parcels of about 4 hectares at the start of tapping. 
In new schemes since 1979, the individual parcel system has been replaced by one of 100 ha blocks, 
operated by groups of 24 settlers. 
14. This principle has been popular and widespread in South and Southeast Asia since the first years of 
Independence. It is also applied to other plantation crops, and to some giant rice cultivation schemes, 
such as the Mahaveli in Sri Lanka. 
15. As a result of their difficulties, many smaller schemes weie taken over and rehabilitated by the Federal 
Land Consolidation and Development Authority, which managed most of them in a highly centra­
lized fashion using hired tappers. 
16. Most land schemes involved the repayment of about two thirds of development costs wi»h interest 
by participating settlers, usually through deduction of proceeds from latex sales over a period of 
some 15 years. 
17. The husbandry practices put forward by management involved careful conservative exploitation 
over a long time period, while the settlers generally wished to minimize short term yields. The settlers 
barely considered the cost of replanting trees, which would be initially undertaken and paid for by 
the scheme. 
18. It was only in the production phase when settlers became actively involved with their parcels that 
conflicts began to rise. During immaturity the settlers were " newcomers ", and also worked in 
general gangs rather than on particular parcels. 
423 
19. Thus in the case of abandoned individual rubber smallholdings in Malaysia, consolidation and reha­
bilitation is being organised in " mini-estates " by the Rubber Industry Smallholders Development 
Authority. 
20. In the past the commercial estate system has been associated with the regimentation of workers and 
their families (Jain, 1970 ; de Silva, 1982). 
21. Thus in almost all cases of a sample of 250 rubber smallholders scattered through South Sumatra 
Province, farmers appeared to have a choice of several possible buyers (Barlow, 1981). 
22. Much can be done by supporting the establishment of better central commodity markets, through 
providing basic facilities etc. 
23. Thus these macro-measures of stabilization should be compared with the alternative of direct grants 
to the poorest farmers, preferably with training and investment components. 
24. Rubber in the late 1970's in Malaysia suffered significant negative effective production (Sahathavan 
Meyanathan and Wells, 1982). 
REFERENCES 
ABDULLAH B I N SBPIEN, CHEAM, S . T . , AND LAI , M. H . ( 1 9 7 2 ) . A report on the financial 
analysis of the M.R.D.C. central processing factory at Meru. Economics and planning 
Division, Rubber Research Institute of Malaysia, Kuala Lumpur. 
ASSOCIATION OF NATURAL RUBBER PRODUCING COUNTRIES ( 1 9 7 6 ) . The progress and deve­
lopment of rubber smallholders. Proceedings of the First Seminar held in Medan, 
Indonesia, Kuala Lumpur. 
BARLOW, C. ( 1 9 7 8 ) . The natural rubber industry. Oxford University Press, Kuala Lumpur. 
BARLOW, C. ( 1 9 8 1 ) . Personal observations of the rubber marketing channel in South 
Province, 1 9 7 6 — 8 1 . 
BARLOW, C. ( 1 9 8 4 ) . Institutional and policy implications of economic change : Malaysian 
rubber 1 9 5 0 — 8 3 . Department of Economics, Research School of Pacific Studies, 
Australian National University, Canberra. 
BARLOW, C. AND LIM, S. C. ( 1 9 6 5 ) . A report on the RIDA central latex processing factory 
at Meru. Economic Report No. 2, Statistics Division, Rubber Research Institute of 
Malaya, Kuala Lumpur. 
BARLOW, C. AND MUHARMINTO ( 1 9 8 2 ) . Smallholder rubber in South Sumatra. Towards 
economic improvement. Balai Penelitian Perkebunan Bogor and Australian National 
University. 
BARLOW, C. AND DRABBLE, J. H . ( 1 9 8 3 ) . Government and the emerging rubber industries 
in the Netherlands East Indies and Malaya, 1 9 0 0 — 4 0 . Paper presented to the 
Conference in Indonesian Economic History in the Dutch Colonial Period. December, 
Australian National University, Canberra. 
BARLOW, C. AND JAYASURIYA, S . K . ( 1 9 8 4 ) . " Problems of investment for technological 
advance : the case of Indonesian rubber smallholders ". Journal of Agricultual 
Economics, XXXV, 1 , 8 5 — 9 5 . 
424 
BARLOW, C. AND JAYASURIYA, S. K. (forthcoming). " Prejudice towards the large farm 
sector in agricultural research. The case of Malaysian rubber". Research and 
Development in Agriculture, London. 
BAUER, P. T. (1948). The rubber industry : a study in competition and monopoly. Long­
mans, Green and Co. London. 
BOOTH, A. (1980). " The burden of taxation in colonial Indonesia in the twentieth cen­
tury ". Journal of Southeast Asian Studies, XI, 1, 91 — 109. 
CHEW, T. A. (1981). The economics of Chinese rubber smallholdings in Peninsular Malaysia. 
Unpublished P h . D . thesis, University Pertanian Malaysia. 
DlJKMAN, M . J. (1951). Hevea : thirty years of research in the Far East. University of 
Miami Press, Florida. 
EDWARDS, C T. (1970). Public finances in Malaysia and Singapore. Australian National 
University Press, Canberra. 
GEERTZ, C. (1963). Agricultural involution. The process of ecological change in Indone­
sia. Australian National University Press, Canberra. 
HAYAMI, Y. AND RUTTAN, V. W. (1971). Agricultural development: an international 
prospective. The Johns Hopkins Press, Baltimore. 
INTERNATIONAL RUBBER STUDY GROUP (1970 — 8 0 ) . Rubber Statistical Bulletin. London. 
JAIN, R . K. (1970). South Indians on the plantation frontier in Malaya. Yale University 
Press, New Haven. 
JAYASURIYA, S. K . AND BARLOW, C. (1984). The impact of government policies on invest­
ment in the rubber industry of Sri Lanka. Paper presented to the International Rubber 
Conference, Colombo, September. 
KOESTONO AND GWYER, G . D . (1976). " Aspects of smallholder rubber replanting and 
newplanting, with special reference to organization, finance, and implementation ". 
In Association of Natural Rubber Producing Countries (1976). 
LIM, S. C. (1968). A study of the marketing of smallholders' rubber at the first trade 
level in Selangor. Report No . 4, Economics and Planning Division, Rubber Re­
search Institute of Malaysia, Kuala Lumpur. 
LIM, S. C. (1976). Land development schemes in West Malaysia : a study of benefits and 
costs. Rubber Research Institute of Malaysia, Kuala Lumpur. 
LlM.T. G. (1977). Peasants and their agricultural economy in colonial Malaya, 1984 — 1941. 
Oxford University Press, Kuala Lumpur. 
MALAYSIAN RUBBER RESEARCH AND DEVELOPMENT BOARD (1983). The Malaysian 
Rubber Industry, 1083 — 2000. Report of the Task Force of Experts, Kuala Lumpur. 
425 
MUBVARTO ( E D ) (1982). Growth aud equity in Indonesian agricultural development. 
Souvenir edition produced for the XVIII International Conference of Agricultural Econo­
mists, Jakarta. 
MUBVARTO (1983). Plantation agriculture in Indonesia ; its past and future. Paper 
delivered to the ARTEP-ILO meeting, Bangkok, October. 
NESS, G. D . (1967). Bureaucracy and rural development in Malaysia. University of 
California Press. Berkeley. 
PEE, T. Y. (1977). Social returns from rubber research in Peninsular Malaysia. Unpub­
lished Ph.D. thesis, University of Michigan. 
PHIT PANVALAKSHANA AND BLENCOWE, J. W . (1976). " Infrastructural support for the 
rubber industry in Thailand". In association of Natural Rubber Producing 
Countries (1976). 
SAHATHAVAN MEYANATHAN AND WELLS, R. J. G. (1982). " Malaysian economic develop­
ment and food trends ". Paper presented to the Second Western Pacific Food Trade 
Workshop. Jakarta. August. 
SHAND, R. T. (1983). Off-farm employment in the development of rural areas : issues. 
Paper presented to the Conference on Off-farm Employment in the Development of Rural 
Areas, Chiang Mai. 
DE SILVA, S . B. D . (1982). The political economy of underdevelopment. Routledge and 
Kegan Paul, London. 
SRIBO CHAIPRASIT AND LIM, P . L . (1976). "Organisation of group marketing". In 
Association of National Rubber Producing countries (1976). 
THEE, K. W . (1977). Plantation agriculture and export growth; an economic history of 
East Sumatra, 1863 — 1942. National Institute of Economic and Social Research, 
Jakarta. 
THOMAS, K. D . (1957). Smallholders' rubber in Indonesia. Institute for Economic and 
Social Research, University of Indonesia. 
THOMAS, K. D. AND PANGLAYKIM, J. (1973). The role of the Chinese in the South Sumatra 
rubber industry. Economics Discussion Paper 28/73, Economics Department, La 
Trobe University, Bundome. 
WHARTON, C. R. (1962). " Marketing, merchandizing and moneylending : a note on 
middleman monopsony in Malaya ". Maylayan Economic Review, 7, 24. 
WHITFORD, H. N. (1931). Estate and native plantation rubber in the Middle East, 1930. 
Rubber Manufacturers Association, Inc., New York City. 
W O R L D BANK (1974 — 82). Miscellaneous reports on Indonesian Rubber development. 
426 
SESSION 8. ROOT D I S E A S E S 
TOWARDS A BETTER LIVING STANDARD OF THE RUBBER 
SMALLHOLDER 
BY 
R. P. M . DE ZOYSA AND L. K. MEEGAHAWATTA 
(Rubber Research Institute, Sri Lanka) 
Agriculture in general is the single largest source of employment in Sri Lanka as 43 
percent of the economically active population is engaged in it, directly or indirectly. 
Rubber, one of the major plantation crops, has contributed significantly to this 
feature, over the last century. Rubber is a labour intensive crop which has played a very 
significant role in the economy as it has generated 500,000 employment opportunities, 
a majority of them in the rural sector. The extent under rubber is 205,700 ha, comprising 
20% of the total plantation crops. 
Economically, rubber continues to play an important role as the export earnings in 
1983 amounted to Rs. 2,852 million. 
As in all rubber producing countries, the smallholder sector predominates in Sri 
Lanka too. The ownership of the smallholder sector which is presently 68 % of the total 
area under rubber rose from 33% as a result of the land reform acts of 1972 and 1975. 
This area supports nearly 196,000 smallholders (Table 1). 
Table 1. Distribution of rubber land area 
Size group (ha) Ha 0 o Holdings % 
0 — 0 5 32651 16* 130604 67" 
0 6 — 1 0 31474 15 
•48% 
41965 21 
1 1 — 2 0 21446 10 14297 7 
2 1 — 4 0 14398 7 
J 
4800 3 
4-1 — 10 0 19941 10" 2849 r 
101 — 20 0 11425 5 . 20% 761 0 5 • 
Over 20 0 10043 5^ 335 0 2 
State 67300 32 32% n.a. 
1 7 % 
427 
Rubber is also an ideal smallholder crop as it is capable of growing under varying 
conditions of difficult terrain and can withstand different management standards. It 
also has a flexibility in the scale of production and marketing. 
The broad objectives of the institutions involved with the agricultural development 
is to bring about an increase in family income, generate employment opportunities and 
improve the general living standard of the population. The role of the Advisory Services 
Department (ASD) of the Rubber Research Board (RRB), is therefore to achieve these 
objectives for the rubber smallholder. A multidisciplinary approach has been found to 
be successful with rubber. 
Replanting 
Increased rate of replanting on an annual target of 3 % of the national stand was a 
primary objective since the inception of the Rubber Replanting Scheme in 1953. A 
considerable drop in performance in the early 1970s, to a very low level of 1 0% has re­
sulted in an accumulation of uneconomic rubber to the alarmingly high rate of 28 percent 
of the national stand in 1979, thus resulting in a very low national average yield in the 
smallholdings. Replacement of the above with high yielding and disease resistant 
varieties was considered a priority activity in the eighties. 
The smallholder Rubber Rehabilitation Project launched in 1980, under World Bank 
assistance, has played a vital role and the performance is considered satisfactory, despite 
climate constraints (See Table 2). Further marked improvement is shown in the new 
planting rate, which is considered essential in view of the large scale urbanisation and 
fragmentation of land. 
Table 2. Amount paid as rubber replanting subsidy for private sector 
Amount in Rs. per acre Date the payment commenced 
2000/- 1974 — 1976 
3000/- 16th November 1977 
4000/- 16th November 1978 
5000/- 1st September 1979 
6000/- 15th November 1979 
7500/- 13th November 1981 
9000/- 15th March 1983 
Source RCD 
428 
The generous subsidies given by Government for both replanting and new planting 
have encouraged rubber planting. The high yielding, disease resistant clones of the 
RRIC 100 series have been popular with smallholders. The new nursery unit, has led 
to proper nursery administration and effective distribution of plants to replanters 
(Table 21). 
Year 
1981 
1981 
Total for 1981 
Table 2 I. Planting material distribution under S.R.R.P. 
Season No . of SH No . of Plants 
May/June 
Oct ./November 
1779 
1417 
3196 
501,419 
376,538 
877,957 
Acreage 
2493 2 00 
1853 0 00 
4346-2 00 
1982 
1982 
Total for 1982 
May/June 
Oct./November 
2160 
1106 
3266 
698,000 
312,350 
1,010,350 
'3209 2 30 
1470 1 37 
4680 0 35 
1983 
1983 
Total for 1983 
May/June 
Oct./November 
2369 
1209 
3578 
797,312 
371,012 
1,168,324 
3507 0 16 
1643 3 16 
5 1 5 0 3 32 
1984 
Fertilizer 
May/June 2326 697,545 3232 0 05 
The present low yield per unit area has increased the cost of production of small­
holdings. In addition, the current high cost of fertilizer almost precludes its use in small­
holdings, unless it is given at subsidised rates. 
In order to achieve increased usage and also to obtain the best benefits, fertilizer 
mixtures based on soil series are being distributed as a part of the Smallholder Rubber 
Rehabilitation Project (Table 2 II). Fertilizer demonstration plots are located in 5 
rubber growing districts. Smallholders will also benefit from soil and foliar analysis 
recommendations. 
429 
tab le 2 II. Fertilizer distribution under the S.R.R.P. Kegalle, Kalutara, Ratnapura districts 
Year Season N o . of SH Acreage Quantity distributed 
Kg. 
1981 Aug./Sept. 1980 2923 0-32 73 415 
Tota l for l981 1980 2923 0 32 73 415 
1982 Feb./March 3597 5531-1 00 137 710 
Aug./Sept. 6530 9324-1 16 311 220 
Total for 1982 10487 14855 2 16 448 930 
1983 Feb./March 7932 11428 2 17 431 930 
Aug./Sept. 10398 15373 3-34 698-415 
T o t a l f o r l 9 8 3 18330 25802 2 11 1,130 345 
1984 
Total for 1984 
Feb./March 
Aug./Sept. 
11766 
13414 
25170 
17652 0-37 
19762-1-15 
37414 2-12 
866 990 
1,283 210 
2,150 200 
Intercropping 
The objective of the smallholder and particularly the smallest, is more to maintain 
an adequate cash income and to maximize returns per hectare. Intercropping offers the 
opportunity to do this during the immature period, and is therefore attractive to the 
smallholder. Although hilly terrain, absentee ownership and, inconsistent marketing 
conditions present impediments to the efforts made, fair success has been made in pro­
moting perennial and seasonal crops (Table 3). 
430 
fable 3. Smallholder Rubber Rehabilitation Project Intercropping —1983 
District Crop Replanting acreage Newplanting 
acreage 
Kegalle 
Ratnapura 
Kalutara 
Banana 2 5 0 1 - 3 8 148-1 20 
Vegetable 14 2 20 — 
Pineapple 15-1-00 8 0 00 
Paddy 0 1 00 — 
Coffee 1 1 20 1 
Passion fruit 0 2 00 
Others — 0-3 00 
i 
Banana 92 0 27 
i 
46-3-30 
Cocoa 1 - 2 - 0 0 I — 
Coffee 0 2 00 
1
 — 
Vegetables 5 3 30 — 
Passion fruit 13 3 00 17 2 00 
High land paddy 5 0 00 — 
Pineapple 1 1 00 — 
Others 8 0 00 6 0 00 
Banana 126 0 20 6 2 00 
Passion fruit 125 0 38 45 2-26 
Coffee 1 2 00 I — 
Pineapple 2 0 00 0 2 00 
Vegetables 84-2-23 4 1 01 
Others — 2-2 00 
The common observation is that rubber grows better whenever intercropping is 
practised. 
Processing and marketing 
The Rubber Research Institute (R.R.I.) encourages central collection of smallholder 
latex for latex based industries, centrifuging plants and to state owned crepe rubber and 
technically specified rubber factories ; these are of significant benefits to the smallholder. 
Further, by processing the rubber in large factories, where cost savings are made on scale 
o f production, the country's foreign exchange earnings are enhanced and at the same 
time a better price is paid to the smallholder. In remote and difficult areas the Advisory 
Services Department aids in establishing Group Processing Centres (G.P.CC) to process 
smallholder latex into superior Rubber Smoked Sheets (RSS). 
Training 
Special emphasis is being given to the aspect of training with a view to achieving 
better cultivation standards. 
431 
Broadly, the training activities offered could be categorised into :—• 
(a) Foreign training 
(b) Resident training (Local) 
(c) Field training 
Short visits to major rubber producing countries such as Malaysia, Thailand, Indo­
nesia are being made by rubber extension personnel, and this programme is continuing 
as a part of the Smallholder Rubber Rehabilitation Project. These short programmes 
have been extremely beneficial in teaching modern concepts, in all aspects of rubber 
growing, manufacture and marketing. 
A fully equipped training centre was established in 1982 to provide resident faci­
lities to the Advisory Services Department staff and all other categories involved in the 
rubber industry. Induction programmes, and annual refresher programmes to Advisory 
Services Department staff and smallholder training in cultivation and processing, are 
broad priorities (Table 4). 
Rubber research findings, cultivation standards, intercropping and processing pro­
cedures are the main areas of training given. Short programmes covering the areas of 
paddy cultivation, coconut growing, family health, environment and ecology, and 
forestry are also provided to enable the extension personnel to be more resourceful at 
the village level. The centralised training, programmes also include training of selected 
smallholders. Village leaders and selected youth, involved with rubber husbandry are 
given priority. As an inducement to the.smallholders, free board, lodging and transport 
from their residential areas are provided to them. 
The training centre has been extremely useful to other related institutions, such as 
the Co-operative Department, Sri Lanka State Plantation Corporation (S.L.S.P.C.), Janatha 
Estate Development Board (JEDB), National Fertilizer Secretariat (NFS) and the Rubber 
Control Department (RCD) who are involved with the rubber husbandry and marketing. 
Other government officers who are engaged with rural agricultural development at grass­
roots level are also accommodated in the programmes. 
In addition several field programmes are regularly conducted by the extension per­
sonnel and the objective of these programmes is farmer education and creation of aware­
ness. These include rubber cultivation training programmes for which an approved 
curriculum is followed by all the Rubber Extension Officers (R.E.OO). Certificates are 
presented to all successful participants. 
Other programmes such as village level group discussions, field days, field demonstra­
tions, are of immense educational value to all involved in the management of rubber. 
432 
Table 4. Progress of training programmes at the training centre established at Nivitigalakele 
4^ 
Type of Course 
01. Induction Course 
02. Induction Course 
03. Refresher Course 
04. Refresher Course 
05. Refresher Course 
06. Refresher Course 
07. Processing Advisory Training 
08. Rubber Husbandry Training 
09. Rubber cultivation 
10. Rubber Processing Training 
11. Rubber Processing 
Category of participants 
Rubber Extension Assistants 
Rubber Inspectors 
Rubber Extension Officers 
Rubber Extension Assistants 
Divisional Rubber Extension 
Officers 
Senior Processing Advisors/ 
Processing Advisors 
— do — 
Rubber Smallholders 
Field Officers (SPC) 
Rubber Smallholders/Group 
Processing Centre's members 
Factory Officers (SPC) 
12. Rubber Cultivation & Processing Co-operative Managers 
13. Fertilizer Demonstration 
Training 
No. of Training 
Programmes held 
1982 1983 1984 
01 
03 
01 
02 
01 
03 
01 
01 
05 
02 
01 
03 
01 
01 
— 01 
08 
01 
04 
01 
— 06 — 
No. of participants 
1982 1983 1984 
26 
38 
74 
16 
74 
30 
14 
12 
150 
60 
150 
16 
All Rubber Extension Officers 
(Kalutara District) 01 — 
74 
16 
16 
12 
240 
21 
120 
19 
35 
Publicity 
Active publicity has been an important component in the Advisory Services Depart­
ment, throughout the years. Activities covered are : 
(a) Programmes conducted with the national television network and the radio. 
(b) Press publicity by way of news, articles and advertisements, 
(c) Publicity material — posters, leaflets and hand-outs, 
(d) Field programmes such as field days, demonstrations, group discussions and 
village-level seminars, 
(e) Participation in exhibitions and competitions, 
(f) Village-level cinema programmes, 
(g) Participating in joint community programmes with the other institutions. 
Considering the increased demands and challenges of the smallholder sector, the 
Advisory Services Department (ASD) has made a conscientious effort to strengthen its 
services during recent times. This includes functional decentralisation, strengthening of 
the staff and the provision of necessary infrastructure. Smallholder Rubber Rehabilita­
tion Project is very helpful in obtaining the desired results having taken into consideration 
the benefits realised by the smallholder. The World Bank has recognised the work done 
by extending the project to all the rubber growing areas of the country. 
ACKNOWLEDGEMENT 
The authors are deeply indebted to Dr O. S. Peries, Vice Chairman, Rubber Research 
Board, for his valuable suggestions and guidance in the preparation. We are also grateful 
to Mr A. Dahanayake, Regional Advisory Officer, Kegalle, for helping us in preparation 
of the slides. 
REFERENCES 
Agriculture/Fertilizer Industry of Sri Lanka, September, 1979. 
COMMONWEALTH DEVELOPMENT CORPORATION. Report on the Rubber Industry Master-
plan study, Twelve volumes. September 1 9 7 9 , South Asia Records. 
JOHN KEELS, Rubber Statistics, 1 9 7 8 , — 1 9 8 2 . 
Proceedings of the fourth seminar held in Colombo, Sri Lanka. 
(a) Rubber Industries Smallholder's Development Authority (RISDA) Malaysia. 
— Socio—Economic factors that affect the development. , 
434 
(b) Sri Lanka Country Report by A. B. Dissanayake. 
The Rubber Research Institute of Sri Lanka. Annual Review, 1978, 1979, 1980. 
The Rubber Replanting Scheme (I.M.F. Report, Sri Lanka) 
WORLD BANK. Development in Sri Lanka, Issues and Prospects Report No. 1937, March 
22, 1978. 
J 
f 
435 
SOCIO-ECONOMIC CHARACTERISTICS OF RUBBER 
SMALLHOLDERS IN SRI LANKA : IMPLICATIONS FOR 
DEVELOPMENT 
By 
H . M. G. HERATH* AND W. G. JAYASENA** 
( University of Peradeniya, Sri Lanka* 
Agrarian Research and Training Institute, Sri Lanka**) 
INTRODUCTION 
The rubber industry occupies an important place in the economy of Sri Lanka. The 
total rubber area was estimated to be 476051 acres or 192655 hectares in 1979 representing 
20 percent of the total area under main plantation crops (CDC, 1979, vol. V. APP. 1). 
About 349203 acres or 68 percent of this area is owned by private rubber growers and 
48 percent of them are smallholders owing less than 4 hectares each. The rubber industry 
has provided employment for nearly 500,000 persons both directly and indirectly in the 
195,600 rubber holdings and in various stages of processing upto shipment. 
The rubber industry is classified into the estate sector and the smallholding sector. 
According to the original classification adopted by the Rubber Controller, lands below 
10 acres are classified as smallholdings and the lands above 10 acres to below 50 acres 
are classified as private estates. The smallholder Rubber Rehabilitation Project of the 
World Bank which sought to improve the smallholder rubber sector classified all rubber 
lands below 50 acre to be small holdings. 
Much of the rubber is now old and there is a need to improve the efficiency of the 
industry by increasing the rate of replanting with new technologies and also accepted 
management practices. The various policies and incentives offered with pious intent, 
manifestly failed to arrest a general deterioration of the industry. This is particularly so 
in respect of replanting. The structure and the problems of smallholders and estates differ 
significantly. It is an imperative to understand the structure and functioning of the 
smallholders and estates to establish effective policies. The objectives of this paper 
are : (a) to document some of the main socio-economic characteristics of rubber small­
holders in Sri Lanka, and (b) to draw implications of these characteristics for their 
development. 
METHOD OF STUDY 
This study is based mainly on a socio-economic study conducted by the authors in 
three rubber growing districts in Sri Lanka, namely: the Ratnapura, Kalutara and Kegalle 
Districts. These three areas represent the best rubber growing areas in the country account­
ing for 68 percent of the country's rubber acreage. Three hundred smallholders from 
each of the three districts were selected according to a two stage random sample. The 
information required for this study was collected mainly from those 900 smallholders 
* Head, Department of Agricultural Economics and Extension, University of Peradeniya, Sri Lanka. 
•* Research and Training Officer, Agrarian Research and Training Institute, Sri Lanka. 
437 
through detailed interviews. Information relating to household particulars, land use 
pattern, income profile, production and processing patterns in rubber are collected during 
the survey. These are discussed in the sections to follow. 
Socio-Economic characteristics 
Demographic characteristics and educational status 
The survey showed that the average size of the household is 5-8, 5*3 and 6 1 persons 
for the Ratnapura, Kalutara and Kegalle Districts respectively. Approximately 54-3, 
49 5 and 51 9 percent of the smallholder population in the Ratnapura, Kalutara and 
Kegalle Districts are males. A majority of them in all three districts is in the economically 
active age group of 15—64 years. The economically active group constitutes 69 9, 68 6 
and 71 2 percent of the total population for the Ratnapura, Kalutara and Kegalle 
Districts, respectively. 
The literacy rates among the farmers in (aged 5 years and over) Ratnapura, Kalutara 
and Kegalle Districts are 94-8, 96-7 and 96 8 percent, respectively, which is very high. 
However, the literacy rates among females in all three districts are slightly lower than 
those of males. There is also a marked variation in the educational attainments of the 
population. Majority of the population in the three districts had an educational level 
of 1 — IX. This group constituted 72 • 2,70 6 and 65 2 percent in the Ratnapura, Kalutara 
and Kegalle Districts, respectively. The percentages of those who have had education 
upto General Certificate of Education (Ordinary Level) are 14 0, 19 9 and 23 9 in the 
Ratnapura, Kalutara and Kegalle Districts, respectively. The number of people with 
higher qualifications (GCE A/L) and above ranged from 3 4 percent to 6 3 percent. 
Activities and characteristics of the labour force 
The activities of the smallholder rubber farmers are an important aspect in designing 
policies for them. Information relating to their activities is given in Table 1. 
Table 1. Distribution of the sample population by main activity (percentage) 
Activity Ratnapura Kalutara Kegalle 
Employed* 28-8 29-5 26 5 
Unemployed 1 0 0 1 1 8 14-4 
Family helpers 6 4 5 0 4-5 
Housewives 13 4 14-4 13 4 
Students 25-4 21-4 28-4 
Others* 15 2 17-9 12 8 
Total 1 0 0 0 100 0 100 0 
* Including employed children 
Source : Jayasena, W. G. and Herath, H. M. G. (1984) 
438 
Tfable 1 shows that the employed and student categories form the important activity 
groups in all three districts. The employed category comprises 28 8, 29-5 and 26 5 
percent in the Ratnapura, Kalutara and Kegalle Districts, respectively. The other two 
important activity categories are the family helpers and housewives which together consti­
tute 19-8, 19 4 and 17 9 percent of the total of the sample population in the Ratnapura, 
Kalutara and Kegalle Districts, respectively. 
The distribution of the labour force into the five main activity groups referred to 
earlier is given in Table 2. 
Table 2. Percentage distribution of labour force in the sample by activity 
Activity 
Male 
Ratnapura 
Female Total Male 
Kalutara 
Female Total Male 
Kegalle 
Female Total 
Employed* 33-3 7-2 40-5 3 4 0 8-3 42-3 32-6 4-2 36-8 
Unemployed 7-7 7-7 15-5 6-9 10 3 17-2 9-9 10-5 20-4 
Family helpers 3 0 6-3 9 3 0-6 6-7 7-3 0-6 5-7 6-3 
Housewives — 19-4 19-4 — 2 1 0 2 1 0 
— 
1 9 1 1 9 1 
Students** 7-9 7-4 15-3 5-7 6-5 12-2 8-9 8-5 17-4 
* Excluding employed children 
** Students aged 15 years and over 
Source : Jayasena, W. G. and Herath, H. M. G. (1984) 
The economically active population of the Ratnapura, Kalutara and Kegalle Districts 
are 85, 87 and 83 percent, respectively. The employed population in Ratnapura, Kalutara 
and Kegalle are 40 5, 42 3 and 36 8 percent, respectively. Majority of the employed 
in all three districts are males. Another important feature of the labour force is the high 
level of unemployment in the three districts. The unemployed population is 15 5, 17-2 
and 20 4 percent in the Ratnapura, Kalutara and Kegalle Districts, respectively. Family 
helpers are also fairly high in all three districts. Though they do not engage in any gainful 
employment, the services given by them for their family activities for rubber production 
especially tapping are important. The housewives population is also very high and in 
most smallholder families, the tapping and processing activities in rubber are done by 
the housewives, a part of their day to day activities. 
The employment pattern amongst the smallholder rubber farmers according to occu­
pation is given in Table 3. Table 3 shows that 51 6, 50 5 and 41 6 percent of the farmers 
in the Ratnapura, Kalutara and Kegalle Districts are employed in agriculture. Thus 
agriculture is the single largest source of employment and income for most farmers in 
the study area. The second important source of employment is government service which 
is classified here as ' white collar ' jobs. The white collar category is high in Kalutara 
amounting to 31 • 6 percent. This is due to the closeness to Colombo and the easy accessi­
bility. Petty trading and non-agricultural labour are also important sources of employ­
ment. A special feature of the Ratnapura District is the importance of the gem mining 
industry both for employment and as a source of income. 
439 
Table 3 Percentage distribution of employed* according to main occupation 
Occupation 
Male 
Ratnapura 
Female Total Male 
Kalutara 
Female Total Male 
Kegalle 
Female Total 
Agricultural operator 45-3 6-3 51-6 48-7 1-8 50-5 36-5 5 1 41-6 
Crafts 1-4 — 1-4 1-7 0-2 1 9 2-6 — 2 6 
Trader 5-3 0-2 5-5 5 0 — 5 0 8-8 — 8-8 
Gemming activities 1 1 6 — 1 1 6 
Ayurvedic Physician 0-6 0-2 0-8 0-8 — 0-8 0-9 0-2 1 1 
Agricultural Labourer 3-5 2-2 5-7 4 0 1 0 5 0 2-6 1 1 3-7 
Non-Agricultural 
Labourer 4 1 3 0 7 1 5 0 0-2 5-2 1 2 1 1-9 1 4 0 
White collar jobs 1 0 4 5 9 16-3 2 3 4 8-2 3 1 6 1 7 0 11-2 28-2 
Total 82-2 17-8 100-0 88-6 11-4 1 0 0 0 80-5 19-5 1000 
•Excluding employed children 
(Source : Jayasena, W. G. and Herath, H. M. G. (1984) 
Distribution and use of land 
Land forms the most basic resource in rubber production. The use of land is in­
fluenced by its size, ownership and distribution, which are examined below for the rubber 
smallholders, The number of holdings and the percentage of total land area by holding-
size are given in Table 4 for all three districts. 
Table 4. Distribution pattern of rubber smallholdings and extent of land by holding size 
Holding size Ratnapura Kalutara Kegalle 
(acres) 
No. of % of total No. of % of total No . of % of total 
holdings extent holdings extent holdings extent 
Below 1 36 1-9 43 3 5 39 2 0 
1 to below 2 86 8 7 101 17 0 91 10 7 
2 to below 4 92 19 7 104 35-7 79 19 0 
4 to below 10 61 30 2 46 3 4 0 71 39 4 
10 to below 25 19 23 1 6 9 8 18 22 6 
25 to below 50 6 1 6 1 — — 2 6 3 
TOTAL 300 100 0 300 100 0 300 100 0 
Source : Jayasena, W. G. and Herath, H. M. G. (1984) 
440 
It is seen that most holdings are concentrated in the different size groups below 10 
acres in all three districts. The 1—2, 2—4 and 4—10 acre size groups are the most 
predominant for Ratnapura, Kalutara and Kegalle Districts, comprising approximately 
80 percent of the holdings. However, in terms of area the 2—4 and 4—10 acre size groups 
are predominant in all three districts. 
The distribution of the holdings and the extent of land under each category attests 
to some degree of inequality in land distribution. For example, 91 • 7 percent of holdings 
in the Ratnapura District is below 10 acres but this group accounted for only 60 • 9 percent 
of total extent of land. In Kalutara and Kegalle, 98 *0 and 93 3 percent of the holdings 
are below 10 acres accounting for 90-2 and 71 • 1 percent of total land area, respectively. 
Structure of land ownership 
The pattern of ownership of rubber land indicates that there are five broad categories. 
Sole ownership is the predominant ownership category in all districts. Sole ownership 
was observed in 74 9, 78 3 and 90 0 percent of land in the Ratnapura, Kalutara and 
Kegalle Districts, respectively. The joint owner and operator group comprised 14 2, 5 2 
and 4 2 percent of the land the Ratnapura, Kalutara, and Kegalle Districts, respectively. 
It was also observed that the substantial extent of rubber lands amounting to 5 8 
percent to 10 9 percent are still under various tenure systems such as Viharagam, Devala-
gam, Nindagam, LDO and encroachments where operators cannot provide any clear 
titles. The impact of these joint and uncertain ownership on production has been amply 
demonstrated in other contexts. There is lack of incentive and initiative amongst the 
farmers to invest in production where clear titles are absent. This is particularly im­
portant in perennial crop production such as rubber which is a form of long term 
investment. The disincentive effects of uncertain tenure on rubber smallholdings 
should thus be carefully borne in mind. 
Type of crops grown 
Another important feature observed is that most smallholder rubber farmers sur­
veyed cultivated other crops such as paddy, coconuts and mixed crops. The average 
size of the rubber holding and that of other crops in the three districts is given in Table 5. 
Table 5. Average size of holdings allocated to different crops (acres) 
Crop Ratnapura Kalutara Kegalle 
Rubber 4 0 2-5 3 6 
Tea 1 1 — 1-6 
Coconut 1-3 1-2 2 1 
Paddy 1-5 1-5 1-2 
Mixed crops 1 9 1 0 1-3 
Source : Jayasena, W. G. and Herath, H. M. G. (1984) 
441 
The average size of rubber holding is 4 acres in the Ratnapura District, and 3 6 in 
the Kegalle District. In Kalutara it is 2-5 acres. The farm sizes of most other crops do 
not change when the size of the rubber holding increases. The average extent of land 
allocated to other crops is larger than the average rubber holding for farmers having less 
than one acre of rubber in all three districts. The paddy acreage varies from 1 1 to 2 3 
acres in Ratnapura. In Kalutara and Kegalle it varies from 0 8 — - 3 3 acres and 0 6—3 • 0 
acres, respectively. Tea did not account for any significant acreage in the Ratnapura 
and Kegalle Districts and was totally absent in the Kalutara District.'The proportion 
of other crops to rubber declines very rapidly with increase in the rubber holding size, 
Pattern of labour use 
Labour is also an indispensable input in rubber production. Tapping is the most 
labour intensive operation and an examination of labour use in tapping gives an insight 
into the labour use pattern amongst the smallholders. For this reason labour use for 
tapping is examined in Table 6. 
Table 6. Percent farmers reporting different types of labour for rubber tapping 
Type of labour Ratnapura Kalutara Kegalle 
Family labour 65 9 68-8 42-1 
Hired labour 17-5 2 8 9 56-5 
Family and hired labour 3 1 2 3 1-4 
Share basis 13-5 —. _ 
Source : Jayasena, W. G. and Herath, H. M. G. (1984) 
It is obvious that family labour forms the predominant source of labour. Approxi­
mately 65-9, 68 • 8 and 42 1 percent of farmers used family labour in the Ratnapura, 
Kalutara and Kegalle Districts, respectively. In Kegalle hired labour appears more im­
portant than family labour. Share labour although not present in the Kalutara and 
Kegalle Districts constituted 13-5 percent in the Ratnapura District. The number of 
farmers using hired labour over those using family labour increases when the holding size 
increased. In contrast most estates have a resident labour force under a centralised form 
of management. 
Income profile of rubber smallholders 
The type of activity and crop combinations determine the level and distribution of 
income. The average monthly household income derived from all sources, in the Ratna­
pura, Kalutara and Kegalle Districts are Rs. 1,562, Rs. 1,389 and Rs. 2,303 respectively. 
The average incomes increase with increase in the size of holding. The below one acre 
size group in the Ratnapura, Kalutara and Kegalle districts had an average household 
income of Rs. 1,021, Rs. 796 and 1,053, respectively. 
442 
It is necessary to point out that rubber is not the only source of income for most 
farmers. Most rubber farmers grow other crops such as paddy, coconuts, mixed crops 
etc., while some family members engage in other forms of employment. Thus most 
farmers obtain incomes from agricultural and non-agricultural sources, other than rubber. 
The percentage composition of income per household by source and holding size is given 
in Table 7. The overall average non-agricultural income accounts for 56 7,48 7 and 31 0 
percent for the Ratnapura, Kalutara and Kegalle Districts, respectively. On an average 
basis income from rubber comes first in the Kegalle District while it comes second both 
in the Ratnapura and Kalutara Districts. 
Implications for development 
The foregoing discussion highlights the most salient socio-economic characteristics 
of rubber smallholders in Sri Lanka which have important implications for their deve­
lopment. The basic approach for development of the rubber smallholder sector lies in 
the development of high yielding technologies which overcome the land constraint. Usually 
high yielding clones are associated with other improved management practices such as 
fertilizer use and pest and disease control. In addition to new technologies other institu­
tional and infrastructural facilities must be made available to facilitate the diffusion and 
adoption of such technologies. It is important although not realized too often that the 
technologies developed should be consistent with the resource endowment of the small­
holders (Ruttan and Binswanger, 1978). It was already pointed out that smallholders 
lack capital but have an abundance of labour. Thus technologies developed for them 
should be more labour intensive but less capital intensive. 
The low level of incomes of small farmers make them reticent to adopt new inno­
vations. This is particularly significant in the case of perennial crops such as rubber 
where funds once committed cannot be easily reversed. Also particularly in the case 
of perennials like rubber where the asset provides some income although due for replace­
ment the farmers show reluctance to replace their assets since they do not wish to forego 
their present incomes, however, little it may be. The examination of the age distribution 
of smallholder rubber confirmed this observation. In the Ratnapura District for example 
18 3 percent of the smallholder rubber area was above 30 years old, although the per­
centages are lower in the other two districts. Although subsidies are provided to cover 
costs in replanting the deferment of income during the immature phase has discouraged 
many farmers from undertaking replanting. Intercropping has now been encouraged 
during the immature phase in order to recoup some of the incomes foregone. Even this 
has not had its desired effect and a subsidy for intercropping itself may have a salutary 
effect on smallholder replanting. Even ownership problems which are more prevalent 
amongst the smallholders act as a disincentive in adopting new technologies as observed 
earlier. 
Where replanting is effected it is again observed that most smallholders have done 
so with the old clones available rather than the new ones. This is evident in Table 8 
where the proportion of RRIC varieties developed by the Rubber Research Institute of 
Sri Lanka (RRISL) is very low. However, these RRIC clones especially 100 series more 
diseases resistant and more vigorous compared to PB 86, an older clone introduced into 
Sri Lanka in the 1940s. 
443 
Table 7. Percentage composition of household income by sources and rubber holding size 
Kegalle 
Holding Size 
(acres) 
Rubber 
Ratnapura 
Paddy Other* 
agricultural 
incomes 
Non-
agriculture** 
Rubber 
Kalutara 
Paddy Other* 
agricultural 
incomes 
Non-
agricul­
ture** 
Rubber Paddy Other* 
agricul­
tural 
incomes 
Non-
agricul­
ture** 
Below 1 1 3 1 13-6 4-6 68-7 21-9 1 0 0 1 1 1 57-0 20-9 14-7 22-9 41-5 
1 to below 2 14-8 9-4 3-4 72-4 25-9 13-3 9-8 5 1 0 22-3 12-5 18-7 46-5 
2 to below 4 28-9 9-6 5-1 56-4 31-9 15-6 7-6 44-9 30-7 14-4 20-5 34-4 
4 to below 10 29-8 12-2 5-9 56-1 32-9 9-5 5-5 52-1 42-5 14-9 16-2 26-4 
10 to below 25 58-2 5-9 3-7 32-2 46-7 13-8 3-7 35-8 66-7 9-6 8-2 15-5 
25 to below 50 6 8 0 8-4 2-6 2 1 0 — — — — 90-8 9 1 — — 
Overall district: 28-4 10-2 4-7 56'7 30-8 12-8 7-7 48-7 39-2 13-3 16-5 3 1 0 
* Coconut, tea, arecanut, cinnamon, cloves and livestock 
'* White collar and other regular jobs, gem mining and off farm incomes 
Source : Jayasena, W. G. and Herath, H. M. G. (1984) 
Table 8. Percentage area of smallholdings under different rubber clones 
Clone Ratnapura Kalutara Kegalle 
PB 86 60-6 58-8 66 0 
RRIC 45 ^ 2-4 3 0 11-5 
RRIC 52 —. 0-9 1-3 
RRIC 37 — — 1 1 
RRIC 100 1 1 —. —. 
RRIC 101 — —. — 
RRIC 102 —. — — 
RRIC 103 — — —, 
Wagga 6278 0-4 — —, 
Unidentified Budded 6-2 8 1 6 3 
Clonal rubber 19-3 20-9 8-4 
Seedling Rubber 10 0 8-3 5-4 
Source : Jayasena, W. G., Herath, H. M. G. (1984 b) 
Several reasons may be adduced for this behaviour. It was noted that less than 4 
percent of the farmers are aware of the RRIC varieties (Jayasena and Herath, 1984 b). 
The farmers are risk averse due to low income and hence do not venture into accepting 
innovations that are untried. The importance of extension agents to create awareness 
and convince them is thus an imperative. The high level of literacy and education of 
smallholders may be a positive sign in this connection. It is also useful to have demonstra­
tion plots which will have a visual impact on them and they are more likely to adopt new 
innovations. 
Even in the adoption of management practices the specific characteristics of small­
holders result in certain patterns of adoption. For example, fertilizer use which is one 
important management practice but capital intensive, is still unsatisfactory. A survey 
in 1981 indicated that only 27-7 and 19 3 percent of the immature rubber extent and 
4 • 6 and 5 • 1 percent of the mature rubber extent farmers in the Ratnapura, and Kalutara 
Districts had been fertilized (Jayasena and Herath 1984). However, in 1983, in a repeti­
tion of the same survey revealed that more than 90 percent of the farmers applied fertilizer 
for immature rubber. However, the frequency of application is not adequate. Applica­
tion of fertilizer for mature rubber also very poor being 3 6, 6 2 and 15 4 percent of the 
extent in the Ratnapura, Kalutara and Kegalle Districts, respectively. (Jayasena and 
Herath 1984 b). This was found to be due to lack of awareness of the correct time and 
amount, inadequate incomes, non availability of fertilizer at required time and lack of 
interest. It was also noted that practices which require only labour such as weed control, 
cover cropping etc., appear to be more readily adopted by a wider spectrum of farmers. 
The implication of this is that when management practices are developed, it is useful 
to concentrate on new innovations that use more labour than capital. Thus research 
on chemical weed control or yield stimulants which are capital intensive appear a question­
able approach to development of the smallholder rubber sector, (Barlow and Peries, 1977). 
445 
The particular social circumstances of farmers have pre-disposed them to certain 
patterns of exploitation of rubber. For example, although half spiral alternate daily 
is the recommended system of tapping many smallholders adopt the half-spiral daily 
system due to their necessity to have daily incomes (Table 9). Amongst the estates usually 
the alternate daily system is followed. This more intensive tapping systems can lead to 
advancement of senescent and also a higher incidence of disease particularly panel dis­
eases. There is a need to educate the farmers on the need to tap less intensively. More 
importantly it is necessary to take account of the need of daily incomes for small farmers 
and develop varieties that would withstand more intensive tapping. At present, in most 
yield trials alternate daily system of tapping is used as a standard method in screening 
new clones which should be changed to more intensive tapping systems as well parti­
cularly for small holders. 
Table 9. Percent of farmers adopting different tapping systems (for all three districts) 
Holding size 
(acres) 
Half spiral Half spiral Half spiral 
daily alternate third daily 
daily 
7» % % 
2-half 
spiral 
Ldaily 
m% 
2-half 
spiral 
alternate 
daily . 
% 
Slaughter 
tapping 
% 
Mixed 
tapping 
systems 
% 
Below 1 35-7 — — 21-4 7-1 — 42-8 
1 to below 2 36-5 13-5 1-4 14-9 6-8 18-9 24-3 
2 to below 4 35-9 17-4 2-2 18-5 1 1 23-9 18-5 
4 to below 10 50-0 11-3 — 2 9 0 4-8 2 1 0 14-5 
10 to below 25 46-2 46-2 — 38-5 7-7 7-7 30-8 
25 to below 50 — 66-7 — — — — 66-7 
Total 39-5 15-9 1-2 20-9 4-3 19-4 21-3 
Source: Jayasena, W. G. and Herath, H. M. G. (1984b) 
The low level of producer incomes, lack of proper processing utencils and also lead 
to poor knowhow, unsatisfactory marketing system also lead to poor processing standards 
by smallholders. It was noted that farmers in most of the small sized holdings do not 
produce any RSS1 which is the best grade. Most of them produce RSS2 or RSS3 (Jaya­
sena and Herath, 1984). The low quality produce leads to low incomes which thus creates 
a vicious circle tending to perpetuate the present system. The small farmer is in a state 
of low income equilibrium. There is a need to improve infrastructure so that the nece­
ssary facilities are available to produce high grade rubber. 
The growing of several crops imply that their labour and capital resources are directed 
towards several crops. This tendency could be strongly felt amongst the smallest farmers. 
They have limited resources such as labour and capital to be used for all crops. Diver­
sion of these resources to other crops during critical periods could affect the operations 
of the rubber stand. Management operations and even tapping could be sacrificed for 
urgently needed cultural operations in other crops such as paddy. Several potential 
problems could arise from this scenario. The effectiveness of measures such as subsidies 
credit facilities could be jeopardised due to possible diversion of such facilities to other 
crops. Fungibility of agricultural credit has been observed to be a particular problem 
amongst peasant farmers in developing countries. Thus, unless measures to counteract 
446 
such tendencies are effectively enforced, underutilisation of resources for the target crop 
could occur. Also, when several crops are grown the need and the ability to specialize 
in any one crop which facilitates sharpening of technical skill are neither needed nor 
present. This could thwart introduction of new technology. Further, when the farmer 
has a crop portfolio it affords considerable flexibility for him under conditions of price 
variability. Thus, changes in the price of rubber can result in wide fluctuations in supply. 
The price elasticity of supply amongst the smallholders could be generally higher than 
what is expected for perennial crops. The resulting uneven supply can have serious reper­
cussions on productions and exports. If these fluctuations are significant, it may have 
implications for price policy. From the farmers point of view however, it is a risk averse 
strategy and farmers may show preference for such a system due to their low tolerance to 
risk. 
CONCLUSION 
The small sample of rubber smallholders investigated in this study limits the gene­
ralization of the research findings to the wide spectrum of producers. Undoubtedly, a 
large sample of farmers, is necessary to arrive at more firm conclusions. Generalizing 
from the findings of this study, however, leads to an important statement about develop­
ment of rubber smallholders. One of the major conclusions is that the type of technology 
developed should be consistent with the factor endowment of smallholders. They should 
particularly be labour intensive but less capital intensive. Some of the factors investigated 
such as low income, uncertain tenure, mixed cropping, small farm size all help explain 
why rubber smallholders are slow to change their production patterns while the estates 
are far superior in this respect. Even institutional patterns should be geared to the par­
ticular socio-economic features of the smallholders. If the above factors are considered 
in development policy, it will impinge in a substantial way on the well being of large number 
of rubber farmers and will have a crucial impact on the success or failure of important 
aspects of development planning. 
REFERENCES 
BARLOW, C . AND O. S. PERIES, (1977). On some Biases in the Generation of Technologies 
by Rubber Research Institutes, Journal of the Rubber Research Institute of Sri Lanka, 
5 4 , pp. 445 — 459. 
COMMONWEALTH DEVELOPMENT CORPORATION, (1979). Report on the Rubber Masterplan 
Study, V. 
JAYASENA, W. G. AND H . M. G. HERATH, (1984). Socio-Economic Conditions of Rubber 
Smallholders in Sri Lanka : A Pre-Project study of Ratnapura, Kalutara and Kegalle 
Districts. Agrarian Research and Training Institute, Colombo. 
JAYASENA, W. G . AND HERATH, H . M. G. (1984 b). Innovation Receptivity and Adoption 
in Rubber Smallholdings of Sri Lanka. Draft report submitted to the Agrarian Research 
and Training Institute, Colombo. 
RUTTAN, V. W. AND H. P. BINSWANGER, (1978). Induced Innovation and the Green Revo­
lution, in Induced Innovation, Technology, Institutions and Development. Edited 
by H P . Binswanger and V. W. Ruttan and others. Johns Hopkins Press, Baltimore. 
447 
P A T H O G E N I C I T Y V A R I A B I L I T Y 
O F RIG1DOPORVS LIGNOSUS A N D PHELL1NUS NOXIUS 
By 
GEIGER J. P., NANDRIS, D. , NICOLE, M. AND RIO, B . 
( ORSTOM, Ivory Coast) 
INTRODUCTION 
Researches devoted to the biology of root and butt rot fungi have shown a variability 
at the physiological and pathogenic levels within the populations of these microorganisms 
(Huttermann et al, 1979 ; Morrison, 1982 inter alia). In tropical climates, the white 
rots of Hevea brasiliensis (Wild, ex. Adr. de Juss) Mull. Arg. caused by Rigidoporus 
lignosus (Kl . ) Imazeki and Phellinus noxius (Corner) G. H. Cunn, have been comparably 
studied (John, 1966 ; Lim, 1970 ; Peries and Irugalbandara, 1973 ; Liyanage and Peries, 
1973 ; Geiger et al, 1976 ; Fox 1977 ; Nicole et al, 1982a, b ; Tran Van Canh, 1982 ; 
Nandris et al, 1984). 
In the context of the ORSTOM research program on these parasites in the Ivory 
Coast, epidemiological surveys in rubber plantations have shown the existence of differ­
ences in the course of the infectious cycle, both among different foci and among trees in the 
same focus (Nandris et al., 1983a). These differences can be attributed to either differ­
ences in host reaction capacities (Geiger et al. 1983) or to variations in pathogenicity 
of the infecting strains. In order to decide between these two hypotheses, a pathogeny 
study was carried out in a greenhouse by artificially infecting young rubber trees with 
different isolates of each of the above parasites. In parallel, the possibility of a physio­
logical heterogeneity in these strains was sought in vitro (manuscript in preparation). 
The results obtained showed pathogenic variations in the R. lignosus and P. noxius 
strains collected from various hosts in West Africa. The present review initially describes 
prior results acquired in this field and then expands on a pathogenicity analysis of a popu­
lation of isolates from the same rubber plantation. Finally, an attempt is made to corre­
late the parasitic behaviour of these strains with saprophytic characteristics observed in 
vitro. 
MATERIALS AND METHODS 
Artificial Infections 
Fungal strains 
The R. lignosus and P. noxius strains were isolated from the roots of trees parasitized 
by the two fungi. The nature of the host plantation or forest varieties and the geographic 
origins are given in Table 1. Strains were grown on 2 % malt medium (Difco) containing 
2 % agar, at 28°C in the dark. The stock cultures of isolates were kept in malt medium 
in test tubes at 16°C. 
449 
Table 1. Characteristics of the various tested isolates 
Rigidopcrus lignosus isolates collected in different areas from various hosts. 
Isolate Country origin Host Isolation 
(Stock date 
1* South East Ivory Coast Hevea brasiliensis 1978 
9 East Ivory Coast Primary forest tree 1978 
13* West Ivory Coast Primary forest tree 1978 
21* South East Liberia Hevea brasiliensis 1979 
37* , South East Ivory Coast Hevea brasiliensis 1981 
38* Cameroon Hevea brasiliensis 1981 
42 South West Ivory Coast Hevea brasiliensis 1981 
52* East Ivory Coast Tectona grandis 1981 
Rigidoporus lignosus isolates collected in the same rubber plantation 
Isolate Country origin Host Isolation 
(Stock n°) date 
9 Hevea brasiliensis 1978 
36 1981 
64A East Ivory Coast 1982 
64C 1982 
64D 1982 
64E 1982 
64F » 1982 
Phellinus noxius isolated in different areas from various hosts. 
Isolate Country origin Host Isolation 
(Stock n°) date 
7 West Ivory Coast Primary forest tree 1978 
45 East Ivory Coast Cedrela odorata 1981 
31 South West Ivory Coast Hevea brasiliensis 1980 
35 East Ivory Coast Hevea brasiliensis 1981 
39 Cameroon Hevea brasiliensis 1981 
2 South West Ivory Coast Hevea brasiliensis 1977 
32 South West Ivory Coast Cedrela odorata 1980 
* Isolates used as reference in pathogenicity test of R. lignosus isolates collected 
the same rubber plantation. 
450 
The inoculum 
Fragments of rubber tree branches (8 cm long by T 5 cm in diameter) were deposited 
in 1 liter Roux bottles containing 100 ml of water. Each bottle was then autoclaved for 
1 h at 100°C and then inoculated with 8 mycelian implants ( 5 mm in diameter) removed 
from the periphery of a 5 day-old preculture on malt agar. These infected sticks were 
used as inoculum 11 months after their inoculation with R. lignosus and 5 months after 
that by P. noxius. 
The experimental system of infecting rubber trees 
Artificial infections were performed in greenhouse under controlled conditions as 
described by Nandris et al, (1983b). Young plants (clone GT 1) obtained from seeds 
were transferred to a vegetation tub (I m 3 ) filled with forest soil. At the age of one month 
the plants were infected with five inoculum sticks deposited at a depth of 20 cm in the 
soil against the tap root. The moisture content of the soil (on a volume basis) was main­
tained at 21% for R. lignosus and at 19% for P. noxius, by controlled watering determi­
ned by a neutron moisture gauge. Thirty rubber plants in the same container were 
infected with each isolate. Non-infected plants, as well as those whose tap roots were 
surrounded with sterile stems, were used as controls. The experiment lasted about 
5 months, from October to March. 
Two scries of experiments were performed using this general outline : 
— one utilized strains isolated from various hosts and from diverse geographic 
origins. During the experiment, 10 randomly chosen plants were removed and 
examined 2 months after infestation by each of the isolates of the two parasites ; 
— the other series was carried out in order to determine the possible variability of 
isolates collected in the same plantation. " Reference " strains, whose previously 
tested aggressiveness was known, were used in parallel to these strains. 
In this context, it is to be noted that aggressiveness levels of these strains, measured . 
yearly for 2 years, varied only slightly (Table 2). 
Table 2. Severity indexes of some isolates of R. lignosus recorded during 2 successive 
experimentations 
Isolates 1983 1984 
(Stock n°) Severity Index Standard Error Severity Index Standard Error 
1 8,4 0,56 8,4 0,46 
21 8,1 0,76 8,23 0,55 
37 5,4 1,16 5,4 1,00 
38 8,3 0,62 8,3 0,39 
52 8,9 0,22 8,6 0,41 
451 
Evaluation of the pathogenicity 
The quantitative determination of the infestation level of each plant was evaluated 
on the basis of the scoring scale 6hown in Table 3. The mean of the severity index (S. I.) 
attributed to each of the 30 plants in the same container, is a reflection of the aggressiveness 
of the isolate tested. In addition to the presence of hyphae on and in the tap root, as 
well as the proportion of necrotic tissue, the following criteria were systematically taken 
into account: 
Table 3. Stages in thecle velopmen I of the disease 
Severity scale of the 
attack Visual rating 
RIGWOPORUS LIGNOSUS 
0 no mycelium on roots 
1 non-aggregate hyphae 
2 rhizomorphs 
3 rhizomorphs and punctual penetration 
4 rhizomorphs and localized necrosis 
5 rhizomorphs and partial tap root decay < 20% 
6 rhizomorphs and 20 to 50 % tap root decay 
7 rhizomorphs and tap root decay > 50 % 
8 foliar symptoms 
9 plant death 
PHELLINUS NOXIUS 
0 no mycelium on roots 
1 mycelial crust in formation 
2 well-formed mycelial crust 
3 mycelial crust and punctual penetration 
4 mycelial crust and localized necrosis 
5 mycelial crust and partial tap root decay < 20 % 
6 mycelial crust and 20 to 50 % tap root decay 
7 mycelial crust and tap root decay > 50 % 
8 foliar symptoms 
9 plant death. 
® growth of stem (measured weekly), 
0 the percentage of contaminated or penetrated or presenting foliar symptoms or 
dead plants, 
® the percentage of plants manifesting anatomic reactions (secondary rhizogenesis) 
which have been described elsewhere (Nicole et al, 1983), 
452 
$ the localization of necrosis into the tap root, 
® the period of time between the appearance of foliar symptoms and death, 
® the date of plant death, 
® enzymatie activities (laccases, hydrolases) of the fungus in the inoculum, assayed 
before, during and after the experiment as described by Geiger (1975). 
In vitro physiological variability of isolates 
Effect of pH and temperature on the growth fate 
The effect of pH on fungal growth rate was determined by growing the different strains 
at 30°C on malt medium where the pH was adjusted with 1 N HC 1. or NaOH, between 
values of 3 and 9. 
The effect of temperature on fungal growth was determined by depositing cultures on 
malt agar and growing them in incubators maintained at 20, 25, 30 and 35°C. 
In both cases, differences among strains were evaluated on the basis of their growth 
after 3 days of culture (mean of two perpendicular thallus diameters). 
Evaluation of wood degrading capacity 
Test blocks were formed from pieces of rubber tree wood (5 x 1 X 0 3 cm in size) 
taken from healthy tap roots in plantations. After drying, the test blocks were placed in 
constriction tubes containing 20 ml of water. Each tube was inoculated with a mycelian 
implant. After 7 months of incubation, the infested blocks were used for two series 
of tests. Wood degradation, i.e. the loss in dry weight (DWL) and lignin content (LIG) 
were measured. In addition, fungal enzymatic activities in the wood were measured : 
beta-glucosidase ((J-glu), alpha-galactosidasc ( « - gal), beta-galactosidase ((5 - gal), pectinase 
(pect), ccllulasc (cell.) and catalase (cat.). Conventional techniques were used for these 
assays (Geiger, 1975). 
Statistical interpretation of the results 
All the data of each experiment were subjected to two types of processing. A multi­
dimensional principal component analysis ( P C A ) was used to establish correlations 
existing among variables and to furnish a schematic representation of the data recorded, 
enabling individuals to be compared among themselves. Multiple regression with re­
gressive selection of variables was used to correlate a given variable of one strain with a 
set of variables, e.g. the S. I. of a strain established in vivo with its enzymatic characteris­
tics. The validity of the resulting correlation coefficient was determined with the F 
test at the 5 % confidence level. 
453 
fcESULfS 
Pathogenicity of isolates of Rigidoponis lignosus 
The study of the pathogenicity of strains collected from forest varieties in various 
regions of West Africa has shown the existence of significant difference in their infectious 
behaviour towards rubber plants (Nandris et al., 1984). This variability is expressed 
as early as the second month after infestation and was particularly pronounced after the 5 
months of the experiment (Table 4). 
Even though the totality of plants infected with each isolate was contaminated and 
penetrated — in the absence of qualitative differences among strains — the percentages 
of plant mortality in fact varied between 10 and 95%, depending on the isolate. In addi­
tion, differences ininfestations were also manifested by the various indicators characterizing 
parasitic attack (Table 4). 
Statistical processing of the data showed that the first two axes of the PCA accounted 
for 97% total variation. Fig. la , b thus gives a satisfactory representation of the 
structure of the parasite population and enables the isolates to be mutually discriminated. 
In order to further define this analysis, additional experimentation was carried to 
test the pathogenic capacity of R. lignosus strains isolated in the same rubber plantation 
and mutually separated by a distance of several hundred meters. The data gathered 
for each isolate (Table 5) show, as before, that there is a considerable variability among 
the seven isolates examined. This variability is most apparent at the level of the percen­
tage of necrotic tissue and of the number of dead plants. It was also noted that some 
strains preferentially attacked the lower part of the root system and then progressively 
worked upwards towards the collar, while others simultaneously attacked the collar and 
the tap root. 
Differences in aggressiveness among these strains were confirmed by the mortality 
rates (Fig. 2a, b). Thus, attack by the most pathogenic strains was brutal, with mortality 
occurring as soon as the sixth week. In the case of other isolates (N° 64 D , 9 and 64 A), 
on the other hand, the infestation remained quite moderate. The statistical positioning 
of the isolates on the basis of their characteristics is shown in Fig. 3a, b. For the sake 
of comparison, reference strains are also included in this representation (axes 1 and 2) 
but they were not considered directly in the analysis of the seven strains examined. It 
should be noted that the classification obtained for some strains on the basis of plant 
mortality level (Table 5) differs from that furnished by the PCA for all the variables 
considered. 
In vitro physiological variability of R. lignosus 
In parallel to the demonstration of in vivo pathogenic variability, we carried out a 
study of the growth and physiological characters of these strains. It was found first 
that the isolates could not be descriminated on the basis of growth rates on agar medium 
as a function of pH and of temperature. The analyses of test blocks degradation para­
meters (loss of weight and lignin contents) and of enzymatic characteristics, showed con­
sequent variations of data recorded for the isolates of each parasite. Nevertheless the 
existence of a correlation between enzyme activities and loss of weight was established. 
454 
38 
Axis 2 
13 
52 . 4 2 
37 
x i s 1 
I 9 
. 1. Pathogenicity variation among Rigidoporus lignosus isolates collected in various areas from 
various hosts: 
. la . position of the strains on the plan 1 — 2 of the principal component analysis (PCA). 
The most pathogenic strains lie on axis 1 negative. 
4 5 5 
a x i s 2 
a x i s 1 
Fig. tb. Position of the variables on the correlation circle. 
MORT : mortality ; NECR : root necrosis ; PENE : root penetration ; RHIZO : rcactional 
rhizogenesis ; ROOT : length of root ; SI : severity index ; STEM : length of stem ; 
variables MORT and SI have nearly the same coordinates. 
456 
457 
axis 2 
Ma 
64q 3 6 
5 2 3 8 
21 
6 4 . 
3Aoi 
9 
a x i s 1 
6 4 . 
37 
Fig. 3. Pathogenicity variation among Rigiilopurus lignosus isolates collected in the same rubber estate : 
Fig. 3a. position ,of the strains on the plan 1 — 2 of the PCA. The most aggressive strains He on 
axis 1 negative. 
4 5 8 
Ffg. 3 b. Position of the variables on the correlation circle. 
LOCA : location of the infection site ; MORT : mortality ; NECR : root necrosis; 
PENE : root penetration ; RHIZO : reactional rhizogenesis ; SI : severity index ; 
STEM : length of stem. 
459 
© 
Table 4. Incidence and severity on rubber seedlings of Rigidoponis lignosus isolates collected in different areas from various host 
RATING CRITERIA 
Isolates 
(Stock n°) 
Severity 
Index 
Standard Growth of Contamina-
error stem tion 
cm % 
Penetra­
tion 
% 
Foliar 
symptoms 
% 
Mortality 
"/ 
so 
Secondary Amount 
rhizogenesis of decay 
% % 
Location of 
infection site 
52 8,9 0,22 41 100 100 0 95 0 91 Tap root 
42b 8,7 0,5 47 100 100 0 90 5 78 Tap root 
+ collar 
13 8,7 0,37 44 100 100 0 85 5 86 Tap root 
1 8,4 0,56 47 160 100 0 75 10 72 Tap root 
+ collar 
38 8,3 0,62 48 100 100 5 70 0 85 Tap root 
21 8,1 0,76 54 100 100 0 70 15 60 Tap root 
4- collar 
9 7,15 1,08 55 100 100 5 45 10 45 Tap root 
+ collar 
37b 5,47 1,16 78 100 90 0 10 60 32 Tap root 
means of 20 inoculated plants 
Table 5. Incidence and severity on rubber seedlings o/Rigidoporus lignosus isolates, collected in the same rubber plantation 
Isolates 
(Stock n°) Severity 
Index 
Standard Length of 
error stem 
cm 
Contamina­
tion 
% 
RATING 
Penetra­
tion 
% 
CRITERIA 
Foliar Mortality 
symptoms 
% % 
Secondary 
rhizogenesis 
/o 
Amount 
of decav 
% 
Location 
infection 
site 
64C 8,8 0,23 59 100 100 6,6 86,6 13 73 Tap root 
+ collar 
64F 7,9 0.50 71 100 100 3,3 50 35 68 Tap root 
+ collar 
36 7,3 0.42 72 100 100 0 20 23 59 Tap root 
64E 7 0,93 84 100 93 0 36,6 32 55 Tap root 
4- collar 
64D 6,8 0,48 83 100 100 0 16,6 30 45 Tap root 
9 5,7 0,67 102 100 100 0 6,6 17 27 Tap root 
64A 5,3 0,54 97 100 100 0 6 43 17 Tap root 
means of 30 inoculated plants. 
Attempt to correlate pathogenicity and physiological characteristics of R. lignosus strains 
Table 6 contains the values characterizing the most representative variables (a priori) 
of the saprophytic behaviour of the isolates, and also the SI of the attacks. The multiple 
regression test enabled us to demonstrate the absence of significant correlations between 
pathogenicity and enzymatic activities assayed in vitro (Table 7). 
As a result of the lack of correlation between these two types of variation, it was of 
interest to determine the effect of the in vitro characteristics on the structure of the 
parasite population, previously determined on the basis of only pathogenic criteria (see 
Fig. 1). 
Four active non-correlated variables were used for this new PCA : the SI, the 
percentage of plants reacting to the attack, the weight loss of the test blocks, and the 
growth rate of the isolates in culture at 30°C and optimal pH. This analysis generated 
a new structuration of the population (Fig. 4a) in which close to 88 % (55 -f- 33) of the 
recorded variation can be accounted for by axes 1 and 2. The comparison of positions 
of individuals and of variables (Figs. 4a, 4b) stresses the preponderant role of the SI 
on the configuration of the population. Although the statistical representation of this 
set of isolates remains globally comparable to that seen in Fig. 1, it is nonetheless noted 
that there is a better discrimination among the five most aggressive isolates. 
Pathogenic and physiological variabilities of P. noxius isolates 
Experimentation identical to that carried out with R. lignosus was performed with 
P. noxius isolates. Table 8 represents the data obtained in vitro and in vivo, respectively ; 
Fig. 5 shows the structure of fungal population. Although there are obvious quanti­
tative differences in the parasitic and saprophytic behaviour of the two fungal species, 
the conclusions of the analysis performed with the P. noxius strains are qualitatively similar 
to those obtained with R. lignosus. 
DISCUSSION 
The analysis of the results shows that variations exist in the pathogenic behaviour of 
R. lignosus and P. noxius isolates collected from different hosts and from different regions 
in West Africa This variability must be considered when practical means for combatting 
these root diseases are put into practice. Henceforth, resistant tree selection and fungi­
cide screening will no longer involve a single strain for performing tests for extrapolating 
the results to all the strains comprising the parasite population of a given region. 
A comparable analysis has been carried out by Liyanage et al. (1977) with R. lignosus 
isolates collected in different Sri Lanka rubber plantations. Even though the general 
experimental design and the main conclusions are analogous in both studies, there arc 
nevertheless certain differences. 
First, the authors stated that the most aggressive strain differed from the others 
tested in this geographic region by the lack of rhizomorph differenciation, regardless of 
medium conditions. The authors also stated that the rhizomorph was initiated only 
i 462 
a x i s 2 
37 
9 
4. Correlation between in vitro and in vivo characteristics of Rigidoporus lignosus isolates. 
4a. position of the. strains on the plan 1 — 2 of the PCA. The most aggressive strains 
lie on axis 1 positive. 
4 6 3 
464 
"9; 
3 2 axis 2 
31 
3 5 
axis! 
4 5 
3 9 
Fig. 5. Correlation between in vitro and in vivo characteristics of PheU'mus noxius isolates. 
Fig. 5a position of the strains on the plan 1 — 2 of the PCA. The most aggressive strains lie 
on axis 1 negative. 
4 6 5 
Fig. 5b. Position of the variables on the correlation circle. 
DWL : difference in weight loss ; RHIZO : secondary rhizogenesis ; S. GRO : speed of 
Isolate growth ; S.I. : severity Index. 
466 
4, * « « < * 
Table 6. Comparison of in vitro a/a/ in vivo characteristics of R. lignosus 
Isolates 
(Stock n°) 
Severity 
Index 
(S.I.) 
Secondary 
Rhizogene-
sis 
(RHIZO) 
% 
Speed of 
growth 
(S. GRO) 
cm/day 
Optimum 
Tempera­
ture 
*C 
Optimum 
PH 
A 
Weight 
Loss 
(DWL) 
Lignin 
Degrada­
tion 
(LIG) 
a galacto-
sidase 
(<*GAL) 
P galacto- p glucosi-
sidase dase 
(pBGAL) ((JBGLU) 
Pectinase 
(PBCT) 
Cellulase 
(CELL) 
52 8.9 0 2,0 30 7,6 21,2 13 0 21,6 21,4 22,8 1 
42 8,7 5 1.7 30 7.0 19,5 7,3 11,8 95,4 57,7 3,2 105 
13 8,7 5 1,8 30 7,6 17,8 9,0 3,0 35,8 21,6 1.7 125 
1 8,4 10 2,0 30 7,6 17,2 16,4 11,2 94,1 39,5 3,5 505 
38 8,3 0 1.4 - 30 7.6 19,7 16,4 34.5 111,0 18,4 3,8 250 
21 8,1 15 2,0 30 7,6 18,7 6,2 10,6 54,0 7,0 1.7 185 
9 7,1 10 1,5 30 > 9 ( ? ) 35.7 23.7 30.0 95,5 12,0 3,5 570 
37 -5,5 60 1,8 30 > 9 ( 7 ) 16.0 15.3 6,8 31,1 19,2 3.7 95 
Table 7. Multiple regression of the severity indexes of R. lignosus isolates explained by 
enzymatic characteristics, weight loss and lignin degradation of the blocks-test 
Percentage of SI Correlation multiple 
Variables Studied variation coefficient F-value 
v explained by " r " x 
p Gal. Pect. Cell., 
and Cat. 34.1(a) 0,58 1,25 NS 
P glu. a gal. 
A W L , and 32,4 (b) 0,57 1,19 NS 
A LIGN. 
(a) S.I. ="A + B, pgal - C.CAT + D.PECT — E.CELL. 
(b) S.I. — A' + B'. ALIG + C \ AWL + D'. (JGLU + E'. a GAL. 
(n - 3)r2 
x Fcal = • with n — 8 and 3 degrees ox freedom 
2 ( 1 — r 2 ) 
F table — 8,35 at 5% confidence level 
when individual hyphae were incapable of penetrating host roots. These observations 
are in. contradiction with results obtained in our laboratory concerning the morphogenesis 
(Boisson, 1968) and the parasitic behaviour of the African isolates of this fungus. All 
the strains we studied in fact produce rhizomorphs, the contamination structure " par 
excellence ", from which the penetration hyphae are differenciated. The differences in 
aggressiveness noted among strains are only manifested after the phase of root penetra­
tion, in the course of tissues colonization. This shows the absence of a relationship bet­
ween rhizomorphogenesis and pathogenicity in the model in question. 
Secondly, Liyanage et al. (1977) observed that the same highly aggressive strain 
was obtained in a region of Sti Lanka which was devastated by root diseases. In the 
Ivory Coast, on the other hand, no direct relation between the origin of the strain and 
its pathogenic capacity could be established, since there are considerable differences in 
aggressiveness among strains obtained in the same area (Table 5). Similar variability 
has been reported in the case of a population of Artnillaria, whose individuals were isolated 
in a distance of less than 100 meters from each other (Redfern, 1975). 
The reproducibility of the results, reflected by the low variations of the SI. of each 
isolate with time (Table 3), shows that experimental conditions of storage during the 
saprophytic stage and of greenhouse inoculations of the plants did not affect the patho­
genic capacity of the strains. The pathogenic variations observed among these isolates 
may thus be considered as being constitutive of each. 
We may now question the basis of this variability. In this context, attempts to corre­
late pathogenicity with the capacity of different strains to degrade plant polymers in vitro 
468 
Table 8. Incidence and .severity on rubber seedlings iv/Phellinus noxius isolates, collected in different areas from various hosts 
Isolates RATING CRITERIA (in vivo) In vitro characteristics' 
(Stock n°) Severity Standard Length Contamina- Penetra- Foliar Mortality Secondary Amount of Location A Weight Speed of 
Index Error Stem tion tion Symptom Rhizogenesis Decay of Infection loss growth 
cm % % % % % % site cm/day 
45 6,45 1,76 49 90 80 15 45 5 53 Taproot 46,8 1,8 
32 6,2 1,3 59 95 95 5 35 25 24 Taproot 37,2 2,3 
35 5,05 1,56 60 85 80 15 15 30 30 Taproot 36,3 1,8 
S 2 4,54 1,9 62 65 65 5 20 25 16 Taproot 40,8 1,9 
7 2,6 1,21 74 60 60 5 0 7 5 Taproot 38,8 1,8 
39 1,5 1,3 74 35 30 0 5 10 5 Taproot 42,8 1,8 
31 0,35 0,56 78 10 10 0 0
 / 0 0 Taproot 35,9 1,9 
means of 20 inoculated seedlings. 
have not furnished positive results. This fact is also confirmed after assaying various 
enzymes (polyphenoloxidases and hydrolases) in the inoculum of each isolate before, 
during and after the experiment (unpublished data). Furthermore, several authors have 
also reported the absence of a correlation between saprophytic (in vitro) and parasitic 
characteristics (Raabe, 1967 ; James and Cobb, 1982 ; Worral, 1983). 
We may invoke two hypotheses to explain this result : 
— pathogeny does not depend only on the degradation of root structures. This 
would be suggested by some recent observations of young rubber plants killed by R. lignosus 
(artificial infection), whose tap root was only partially necrosed. We may thus reconsider 
the hypothesis of the participation of a fungal toxin in the infectious process already 
suggested by Peries, 1959 ; 
— the saprophytic behaviour of these fungi, I.e. their capacity to degrade inert 
substrates, would not characterize the actual mechanisms utilized by the parasite to colonize 
its host. Extrapolation of results acquired on parasite behaviour in vitro would thus 
not be valid. 
Other explanations have still been advanced to explain the variability existing among 
individuals (Prillinger and Molitoris, 1979) : 
— a variation related primally to the stage of development of each isolate, 
— a variation depending on the ecological conditions prevailing in the region in 
which the isolate originated, 
— a variation caused by genetic differences. 
The results described in the present article, in particular the stability of the pathogeny 
of a given strain with time and its variability recorded in a limited perimeter tend to prove 
that the third of the above hypotheses is the best adapted to explain the observed pheno­
menon. Thus, as reported for the genus Armillarla (Redfern, 1975), the African popula­
tions of Rigidoporus lignosus and Phellinus noxius would thus each constitute a mosaic 
of " c l o n e s " with different pathogenic potentials. 
REFERENCES 
BOISSON, C , ( 1 9 6 8 ) . Mise en evidence de deux phases myceliennes successives au cours 
du developpement du Leptoporus lignosus (Kl . ) Heim. C. R. Acad. Sci. serie D , 116, 
1 1 1 2 — 1 1 1 5 . 
Fox, R. A., ( 1 9 7 7 ) . The impact of ecological, cultural and biological factors on the 
strategy and costs of controlling root diseases in tropical plantation crops as exempli­
fied by Hevea brasiliensis. J. Rubb. Res. Inst. Sri Lanka, 54, 3 2 9 — 3 6 2 . 
470 
(jBIOBR, J. P . , (1975). Aspects physiologiques et biochimiques de la specialisation para-
sitaire. Cas particulier des Corticium rolfsit et Leptoporus lignosus. Etude in vitro. 
Physiol, veg. 1 3 (2), 307 — 330. 
GEIGER, J. P . , N A N D R I S , D . , GOUJON, M., (1976). Activites deslaccases et des peroxydases 
au sein des racines d'Hevea attaquees par le paurridie blanc (Leptoporus lignosus (Kl . ) 
, Heim). Physiol, veg. 1 4 , 2 7 1 — 2 8 2 . 
GBIGBR, J . P., NANDRIS, D . , NICOLE, M., (1983). Comparative studies of rubber root 
rot caused by Rigidoporus lignosus and Phellinus noxius in the Ivory Coast. Physio­
logical, biochemical and cytological aspects of host-parasite interactions. Poster, 
6th international conference of IUFRO on " Root and Butt rots ", Melbourne, Aus­
tralia. 
HUTTERMANN,A. ,VOLLBR,Ch. ,SCHORN, R . , A H N E R T , G . , G A N S S E R , H . G. , (1979) . Studies 
on isoenzyme polymorphism in Fomes annosus. Eur. J. For, Pathology 9 (5), 265—274. 
JAMES, R. L., COBB, J. R. (1982). Variability in virulence of Heterobasidion annosum 
isolates from ponderosa and Jeffrey pine in areas of high and low photochemical air 
pollution. Plant Disease 66, 835 — 837. 
JOHN, K., (1966). Effect of inoculum size and age of trees on root disease infection of 
Hevea brasiliensis. J. Rubb. Res. Inst. Malaya 1 9 , 226 — 230. 
LIM, T . M., (1970). Stem rot of Hevea caused by Phellinus noxius. Crop protection in 
Malaya p. 221. (Wastie R.L. and Wood B. K. ed.) Kuala Lumpur. 
LIYANAGE, G . W., PERIES, O. S., (1973). The control of white root disease in Sri Lanka. 
/. Rubb. Res. Inst. Sri Lanka 50, 201 — 207. 
LIYANAGE, G. W., LIYANAGE, A. DE S., PERIES, O. S., HALANGODA, L., (1977). Studies 
of the variability and pathogenicity of Rigidoporus lignosus. J. Rubb. Res. Inst. 
Sri Lanka 54, 362 — 372. 
MORRISON, D . J . , (1982). Variation among British isolates of Armillaria mellea. Trans. 
Br. mycol. Soc. 78 (3), 459 — 464. 
N A N D R I S , D . , NICOLE, M., GEIGER, J . P., MALLET, B., (1983a). Root rot diseases in the 
forests and plantation of the Ivory Coast. Communication, 6th international con­
ference of IUFRO on " Root and Butt rots ", Melbourne, Australia. 
NANDRIS , D . , NICOLE, M., GEIGER, J. P., (1983b). Infections artificielles de jeunes plants 
d'Hevea brasiliensis par Rigidoporus lignosus (Kl ) . Imaz. et Phellinus noxius (Corner) 
G . H. Cumm. Eur. J. For. Pathology 1 3 (2), 65 — 76. 
NANDRIS, D . , NICOLE, M., GEIGER, J. P. (1984). Pathogenicity of some tropical root 
rot fungi 11. Evidence of pathotypes among Rigidoporus lignosus and Phellinus noxius 
populations in West Africa. Proposed for publication in Phytopathology. 
471 
Nicole, M . , Geiger, J. P., Nandris, D . (1982a). Interactions h6te-parasite entre Hevea 
brasiliensis et les agents de pourriture racinaire Phellinus noxius et Rigidoporus lignosus. 
Etude physiopathologiqtie comparee. Phytopai/i. Z. 105, 311 — 326. 
Nicole, M . , Geiger, J . P . , Nandris, D . (1982b). Aspects ultrastructuraux de la degra­
dation du phloeme des racines d' Hevea brasiliensis parasite par Rigidoporus lignosus. 
C. R. Acad. Sc., Paris 295, 471 - 476. 
Nicole, M . , Geloer, J . P., (1983). Cinetique de iinfection de plants d'Hevea brasiliensis 
par Rigidoporus lignosus (Kl.) Tmazeki. Can. J. For Res. 13 (3), 359 — 364. 
Peries, O. S . ( 1 9 5 9 ) . Studies on the production of toxin by Fomes lignosus. I. Preliminary 
investigations. J. Rubb. Res. Inst. Sri Lanka 35, 38 — 40. 
Peries, O. S., Jrugalbandara, Z. E . , (1973). Histology of Hevea roots infected by Fomes 
lignosus. Ann. appl. Biol. 73, I — 7 . 
Prillinger, H . , Molitoris, P . H. (1979). Genetic analysis in wood decaying Fungi. 
I. Genetic variation and evidence for Allopatric Speciation in Pleurotus ostreatus 
using Phenoloxydase Zymograms and Morphological Criteria. Physiol. Plant. 46, 
2 6 5 — 2 7 7 . 
Raabe, R . D . ( 1 9 6 7 ) . Variation in pathogenicity and virulence in Armillaria mellea. 
Phytopathology 57, 7 3 — 7 5 . 
Redfern, D. B. ( 1 9 7 5 ) . Influence of food base on rhizomorph growth and pathogenicity 
of Armillaria mellea isolates, in Biology and Control of Soil borne Plant pathogens 
(ed. G . W. Bruehl). pp. 69 — 7 3 . American Phytopathological Society, St. Paul, 
Minnesota, U.S.A. 
T Ran Van Canh, ( 1 9 8 2 ) . Lutle contre lc Fomes : nouvelle methode d'etude. Caout. 
Plas't. 6 1 7 — 6 1 8 . 
Worrall, J . J . , Parmeter, J . R., Cobb, F. W. (1983). Host specialization of Hetero-
baridion annosum: Phytopathology 73. 304 — 307. 
472 
A NEW METHOD OF DIRECT CONTROL OF FOMES 
LIGNOSUS* CAUSING AGENT OF WHITE ROOT 
ROT OF HEVEA 
By 
TRAN V A N CANH 
( I R C A COTE D ' IVOIRE Ivory Coast) 
SUMMARY 
White root rot of Hevea, caused by the fungus Fomes lignosus presents a problem of 
particular severity in Lower Ivory Coast. 
The author has presented first the particular aspects of the disease in Ivory Coast: the 
environment, the pathogen, the damage. Then he has shown a new method of artificial 
inoculation which allowed to demonstrate good efficacy of Calixin against the parasite. 
Studies on the rate and frequency of application of the product have shown that by applying 
10 g ofTridemorph, the a.i. of Calixin, per tree every 6 months, one can protect Hevea trees 
vei'y well from contamination with Fomes. 
Field trials, running since 3 years, have confirmed the efficacy of Calixin for direct 
control of the disease. Following these results, the plantations of Ivory Coast have applied 
this new method, consisting of detection of diseased trees (once per year) followed by 2 Calixin 
treatments per year of the diseased trees and their two direct neighbours in the line. 
INTRODUCTION 
Among wood destroying rots which attack cultivated plants and arc prevailing in 
Ivory Coast, white root rot caused by Fomes lignosus is an important root disease of 
Hevea in the forest zone. This disease is a problem for Hevea cultivation in the country 
since 1953, when the first large scale plantations were established. It seems that the 
parasite has found nearly optimum conditions on the sandy soils of tertiary origin of 
Lower Ivory Coast. This applies both to its development and virulence : Crops culti­
vated on new cleared forest areas, rich in sources of primary infection (Chevaugeon, 1956; 
Delabarre, 1970) ; light top soil, medium high temperature, favourable rainfall and soil 
moisture mild dry season, host plant receptible and of little resistance. 
Due to the position of the affected plant parts in the soil and the very solid structure 
of the rhizomorphs which propagate easily at greater depth, little was available until 
recently to control the disease effectively. 
Since 1978 the " Service de Phytopathologic " of IRCA has concentrated its activities 
on research concerning this disease. The combined outcome of this research allowed to 
put forward a control method against Fomes which until now seems to show satisfactory 
results. 
At first, peculiarities of the disease in Ivory Coast arc described, in the following parts 
the different procedures are presented which allowed to recommend Calixin for the control 
of Fomes and its large scale application. 
"The old name of Fomes lignosus is used for Rigidoponis lignosus throughout this paper. 
473 
The special conditions of Ivory Coast 
The behaviour of the pathogen (growth and virulence) depend on the environmental 
conditions and on the reactions of the host plant which are again influenced by the envi­
ronment. 
The ecological conditions of Ivory Coast are given for comparison with other sites. 
The environment 
Root diseases are common in forest lands. The young Hevea trees are contaminated 
by diseased stumps of certain species which are found in dense forests of the "rain forest" 
type. 
With regards to soil, all operating plantations are established in lower Ivory Coast, 
belonging schematically to two regions (Fig. 1) : 
— In the southeastern region on tertiary sands with fairly uniform physical proper­
ties. These soils are distinguished by 1 0 — 15 % clay content in the topsoil, 
and 15 — 20 % clay content in 1 meter depth. The average pH value is about 
4-7 to 5 0, with minima of 4«0 and maxima of 6-0. 
— Southwestern Region : The soils are slightly heavier with 25 — 30 % clay in the 
topsoil and over 30% in 1 meter depth. The average pH is about 4*6. 
The climatical data for the IRCA plantation, situated in the southeastern zone is 
summarized in Table 1. It shows two rainy seasons : The big rainy season in May - June 
and the small one in October and November. The average temperature is 26*3°C and 
the relative humidity is 86 6%. 
The soil is saturated with water for the biggest part of May, June and July. It dries 
up during the small dry season in August and becomes wet again in October - November. 
From January to March the soil moisture content is progressively reduced. 
In general, the ecological conditions are favourable for the' spread of the disease, 
particularly the soil conditions appear to be almost optimal. 
The Pathogen 
Fomes lignosus is a fungus parasitic to the roots of a great number of woody plants, 
both growing wild and cultivated. In the forest, its parasitic role is limited. After role 
clearing, conditions become favourable for its development, and it attacks quickly the 
stumps of the felled trees. Each colonized stump forms a potential source of infection 
that threatens the planted Hevea in a circle of up to 40 — 50 m radius. 
The disease spreads normally by means of mycelical filaments called rhizomorphs 
which develop on the roots of colonized host plants or simply by spreading through the 
soil. 
474 
475 
Contamination of Hevea roots takes place by contact with roots of infected forest 
stumps. The extension of the source of infestation (foyer) and the duration of its activity 
depend on the size of the stump and its contents of nutritive elements which can be utilized 
by the parasite. 
Symptoms : The symptoms of the disease develop with time : when the disease has reached 
an advanced stage, normally an off season flowering is observed. The leaves turn first 
yellow, then reddish brown and eventually fall off. Sometimes the tree refoliates, but 
the new leaves are smaller and lighter in colour. 
At the same time of or following the wilting of the foliage, progressive die back of 
branches, forming of carpophores at the collar and eventually the death of the tree 
is observed. Often one can see also striate grooves (cannelures) on the base of the trunk 
of infected trees. This is due to the destruction of the main root, the function of which 
is replaced by lateral roots. These become very strong and lead to an asymmetrical form 
of the trunk base. In this case, the foliage normally remains healthy. 
The symptoms of the disease have been precisely described by several authors and 
in different countrys where Hevea is grown. 
In Ivory Coast, this has been done by Martin et al, (1968). According to them, 
it seems that grooves at the trunk base are the first visible symptoms. Foliar symptoms 
(discoloration and leaf fall) are only visible at an ultimate stage. 
The damage caused by Fomes is conspicuous after each cyclone. One observes that 
trees with obviously normal leaves are uprooted by the storm. An examination of the 
root system shows that the main root is rotten and broken 25 — 30 cm below the collar. 
The Damage : The damage caused by Fomes can be seen early. From the end of the year 
of planting the first symptoms can be actually seen, but from the third year after planting 
the level of infection increases rapidly. 
This is partly due to the extension of the root system of the Hevea trees, which multi­
plies the chances of contact with infected stumps, and partly due to the fungus which, 
after the forest had been cleared, has colonized a certain number of stumps and spreads 
by growing along their roots. Mortality increases from year to year until it reaches the 
maximum in the 6th year (6 % annual mortality). After the 6th year, a decrease of viru­
lence of the root rot can be observed, partly due to the disappearance of forest stumps, 
partly because of a large increase of root mass of the Hevea trees. Annual mortality dimi­
nishes and levels out at 1 — 2 % (Fig, 2). 
In later stages, the extension of the disease spreads out from spots which develop 
into clearing inside the plantation and which favour losses due to windbreak. Because 
of this, one can see in certain areas several clearings of more than 1 ha within a block 
of 25 ha (20 year old trees). 
476 
Table 1. Weather data — Meteorological Station ofIRCA ANCUEDEDOU 
Averages 1957 —1983 
Month Jan. Feb. March April May June July Aug. Sept. Oct. Nov. Dec. Total 
or meat 
Rainfall in mm 36,3 74,5 103 150,4 245,1 571,5 241,3 57,7 91,5 156,3 136,9 70 1936,5 
Mean temperature 26,8 27,4 27,8 27,6 26,9 25,1 24,3 24,5 25,3 26,5 26,8 27,2 26,3 
Mean rel. humidity 82,1 82 81,6 83,1 86,5 91,1 89,9 87,5 88,3 86,4 87 85 85,6 
Mean duration of sunshine 
in hours 74-83 180 162 171 179 150 90 78 54 72 117 174 162 1599 
Fig. 2 : Development of the disease in relation to the age of Hevea trees (Martin et Col. 1968) 
MATERIALS ANTJ METHODS 
Fungicide efficacy test 
The technique of artificial inoculation, the different treatments and the evaluation 
of the intensity of the disease are described below. 
Selection and preparation of host plants 
The Hevea plants are grown in an open air nursery. They are lifted at the age of 
6 months. Only plants with a diameter between 1 0 and 1 • 5 cm are used. 
They are trimmed to 35 cm length of the main root (pivot) and 30 cm of the stem. 
They are washed in water, and, after all rootlets were removed, planted into a container 
(A) filled with washed sea sand (surface : 1 m a , depth : 50 cm). 
The stumps so prepared are arranged around vertically inserted plastic tubes of 20 cm 
diameter, 10 stumps surrounding one tube and four tubes per container, representing four 
replicates (Fig. 3). 
The plants are watered every 2 days : the first five times with pure water (100 cc / 
stump), the following with a nutrient solution (100 cc/stump) — for composition see 
Table 2. 
Debudding is carried out regularly to allow only 1 stem to develop. 
Table 2. Composition of the nutrient solution 
Macroelements Trace elements 
Salts g/1 Salts mg/1 
Ca ( N O a ) 2 4 H 2 0 0,590 Mn CI, 4 H 2 0 1,80 
K N 0 2 0,505 Cu S 0 4 5 H 2 0 0,18 
K H g P 0 4 0,136 Zn S 0 4 7H a O 0,22 
Mg S 0 4 7 H 2 0 0,492 H 3 B O a 2,86 
N H 4 N 0 3 0,200 M 0 7 0 2 4 ( N H 4 ) 4H a O 0,26 
Choice and preparation of the inoculum 
Main roots (pivots) from 4 year old Hevea trees which have recently died ftomFomes 
infection provide the natural inoculum. 
After taking them out of the plantations they are washed, cut to size and then incu­
bated in an other container filled with sand (B). 
They are watered every 2 days for 3 weeks. 
478 
' 0 . 
11 
e 1? 
t • 19 o 
• 4 18 « 
eluamcnt 
lortttj 
infected raoti 
plastk tubs 
479 
inoculation 
Inoculation takes place when all young plants have at least one set of branches, 
approx. 6 weeks after transplanting into box A. 
The plastic tubes are removed and replaced by the infected " pivots ", coming out 
of container B (Fig. 4). 
Treatments 
Two weeks after inoculation, the plants are drenched once with solutions of different 
fungicides. All plants of the same container receive 100 cc of same fungicide solution, 
poured around each stem base, 18 fungicidial formulations (Table 3) were tested at the 
same dose of 10 g active ingredient per liter solution (g.a.i./l). They were compared with 
an untreated control, giving a total of 19 treatments corresponding to 19 containers. 
Table 3. Fungicides tested. Concentration : 10 g. a. i. per liter water 
Commercial name Active ingredient 
Aliette (S) 
Arbotec (S) 
Benlate (S) 
Benodanil (S) 
Calixine (S) 
Cryptonol (S) 
Derosal (S) 
Dowco 444 (S) 
BAS 38901 F (S) 
Chevron 20615 (S) 
Pelt (S) 
Ridomil (S) 
Rovral (C) 
Thiovit (C) 
SICAROL (S) 
GALBEN (S) 
SERINAL (S) 
TILT (S) 
Ethylphosphite d'aluminium 
2 - (4 - thiazolyl) benzimidazole * 5 g.m.a/1 
Benomyl 
2 - iodobenzanilide 
Tridemorphe 
pentachloronitrobenzene 
Carbendazine 
Furmecyclox 
Milfuram 
methyl - thiophanate 
N° CGA 48988 
Iprodione 
micronized sulphur 
Pyracarbolide 
Benalaxyl 
Dichlozalinate 
Propicanazole * g.m.a./l 
(S) product with systemic action (C) product with contact action 
* Concentration of 10 g a.i. per 1 caused the death of all plants due to phytotoxicity 3 
weeks after treatment. 
Evaluation of efficacy 
Efficacy evaluation of the products is carried out when practically all plants of the 
untreated control are killed, approx. 6 months after inoculation. 
480 
481 
Scoring of intensity of the disease is being done by means of a simple assessment 
system based on the state of the infection of the main root of the test plants. After 
up-rooting and splitting of the main root, each plant is classified ranging from 1—5. 
Healthy plant: N o trace of rhizomorphs on the bark of main root. 
Contaminated plant : Rhizomorphs visible on the main root, tissue of the bark and wood 
healthy. 
Infected plant: Rhizomorphs visible on the main root, with a quarter of the tissue necrotic 
(beginning of infection). 
Heavily infected plant : Rhizomorphs abundant with half the tissue of the main root 
necrotic. 
Dead plant : Rhizomorphs, tissues entirely necrotic. 
An average.score (N) is calculated for each treatment following the formula : 
N — average score of 1 treatment (1 container) 
n — score of each stump of the same replicate 
p — number of plants of each replicate (10) 
r — number of replicates per container (4) 
Efficacy of a product is estimated partly by the average score (N) giving the incidence 
of disease development in one container and partly by the percentage of killed and infected 
plants. 
Efficacy of a treatment is considered as good, medium or bad according to the average 
score (N) and the percentage of dead and infected plants (Table 4). 
Table 4. Determination on efficacy 
% of the plants 
Average score Infected Dead 
Efficacy (N) (dead or living) 
Good < 2 - 5 v < 3 0 % < 3 0 % 
Average between 2 • 5 and 3 5 30 to 50 % 30 to 50 % 
Bad > 3 - 5 > 5 0 % > 5 0 % 
Studies on rate and frequency of Calixin application against Fomes. 
482 
The same method of artificial inoculation was used with the following modifications: 
— host plant : grafted stumps GT1, 20 months old 
— substrate and experimental lay out : The stumps are planted into a concrete 
container open to the ground and filled with soil (dimensions : 1 - 2 m x 0 9 m). 
They are planted in groups of two. There are five groups per container (Fig. 5). 
Treatments 
For the trial " rates " : One week after inoculation, all stumps of the same container 
receive the same fungicide emulsion (Calixin) i.e. 1 litre around their stem base. 
The rates are 4, 10, 20, 30 and 40 g active ingredient per litre emulsion, including an 
untreated control altogether six treatments with two replicates. 
6 months after the first application, all plants are again treated as before. 
For the trial " frequency " : The first treatment is made one week after inoculation. 
2-1 of a 0 • 5 % a.i. emulsion of Calixin are applied per tree. The different frequencies of 
application are : one single treatment, treatment every 6 months (twice per year), every 
4 months (three times per year), every 3 months (four times per year), every 2 months 
(six times per year) and every month (12 times per year). 
Evaluation of efficacy 
The efficacy was evaluated after 12 months for " r a t e s " and after 15 months for 
" frequences ". 
Field trials 
Objective of these trials is to limit the extension of sources of infection with Fomes. 
A plot of 25 ha of 4 year old Hevea, showing heavy Fomes damage, was selected. Detec­
tion was carried out by doing a very precise survey, tree by tree, with removal of the soil 
up to 15 cm depth around the collar and the lateral roots. Dead trees and those identi­
fied as infected (mycelium of Fomes visible at the collar and necrosis 10 cm below) were 
eliminated by extraction. Then the trees were classified into 3 groups. 
— 1st group : trees known as contaminated with Fomes : mycelium is visible 
at the collar, but the main root at 15 cm depth is still healthy. 
— 2nd group : trees with direct contact with a source of infection : healthy trees 
neighbouring trees contaminated with or killed by Fomes (1 st tree on each side). 
— 3rd group : trees with indirect contact with a source of infection (2nd tree on 
each side). 
483 
484 
The treatment consisted of carefully loosening the soil and forming a funnel like 
furrow (cuvette) around the main root, then pouring into this furrow 2-1 of a 0 5% 
Calixin emulsion. Treatments were first carried out at the commencement of the trial, 
then in the 6th and 12th month. 
RESULTS 
Efficacy test of fungicides 
In the tests some cases of mortality were observed from the 3rd month after inocula­
tion onwards: towards the 6th month more than 90 % of the untreated plants (control) 
were dead. 
The results are shown in Fig. 6 which indicates the " average score " (N) as well 
as the percentage of dead and infected plants for the different treatments. 
After 6 months, practically all treatments showed 100% of the plants infected and 
more than 89% dead. 
Only two products produced evidently lower percentages of infected plants than 
the other treatments : Rovral with 60% and Calixin with 10%. 
— Rovral: With 60 % of the plants infected, the mortality reached 35 % ; its average 
score is 3-2. Its efficacy can be classified as medium to insufficient. 
— Calixin : With 10% of the plants infected, and an average score of 1-3, this 
product can be considered as effective. In this treatment 12 • 5 % of the plants 
were dead one month after application of the product, but no trace of rhizo­
morphs could be seen on the main roots. As the surviving plants showed a re­
tardation of growth as compared with untreated controls, one is allowed to 
assume phytotoxicity. 
Studies on rate and frequency 
The results are summarized in Figs. 7 and 8. These show the average scores and 
for each treatment the percentage of dead, infected, contaminated and healthy stumps. 
One can see that the threshold of efficacy lies at 10 g a.i. per tree. At this concen­
tration, the average score is 2 3 against 4 8 of the control, and the percentage of healthy 
trees is in the order of 50 % which corresponds to " good " efficacy according to our system 
of evaluation. 
Three months after the last application it could be stated that it needs at least 3 
applications with 6 months interval to protect the stumps against Fomes attack. 
Application every 4 months, 3 months and 2 months gives much better control, and 
only a monthly application can assure 100% protection. But at this frequency phyto­
toxicity is observed : The stumps are much less developed than those of the untreated 
and non-inoculated controls. 
485 
average score 
plants 
aerloucly dieeased 4 J 
diseased 5 
lontaolnoted 2 
of 
tfFicacy of 
product % 
li 
Crypto. O«no-
dontl 
Jl 
pALUCHpeROSAU 
ih rli 
F u m a -
J l 
BAD 
Hi 
MEDIUM 
s.s 
GOOD 
products 
toatsd 
*•) Percentage of total number of plants of each treatment, 
* •) plants dead due to phytotoxicity 
Fig. 6 . Efficacy of fungicides against fomes lignosus mat artificially inoculated in vitro 6 months after 
inoculation. 
Average scor*. 
P ' « n t i dead 
plants seriously 
diseased 
5-1 
•H 
ptoriti diseased 3 J' 
Plgntj contaminated j J 
Piont i healthy-
Efficacy, o f 
product 
12 month? ji.f.ter inoculation 
n 
( H ) 
SAD 
3 . 1 
MCDlUrj 
a-i. rate 
">. a/ tree 
'Control 
0 * 9 .10 9 ro 9 3 0' g 40" g 
•n.of treatments 0 2 U m e 3 . ! . 1 woeVand 6 months after inoculation 
»* 
of 
'plants 
dead (50 + 0 2V~ 20 20 10 
infected 4 0 3 0- *0 10 10 -
contaminated 20 2 0 10 10 20 
heal thy 
-
10 50 60 do 20 
Fig. 7. Trial rates of Calixio, 
487 
average score 
p l a n t s dead 
planto seriously 
diseased 
•PJONTS 
diseased 
* l a ' i t o c o n t a m i n a t e d j 
plants healthy 
Rate: 10 g a . i i / t r e e 
T i m e : 15 months after 
inoculation 
Efficacy of product 
BAD 
U s 
1 
c o n t r o l 
1 s i n g l e 
t i n e 
e v e r y 
6 months 
e v e r y 
ii months 
e v e r y 
3 months. e v e r y 
2 months 
r—1 1 1 
e v e r y 
•month 
' n o , o r t r e a t m e n t s 
0 1 3 4 S 7 1) 
' S 
d e a d 00 4 0 1 0 10 
- -
. o f 
i n f e c t e d 4 0 1 0 10 1 0 
-
-
P l a n t s 
c o n t a m i n a t e d 
-
1 0 10 j o JO 3 0 -
h e a l t h y 
-
1 0 10. 6 0 7 0 7 0 1 0 0 
MEDIU) 
GOOD 
Fig. 8. Trial frequency of application. 
488 
Field trial : Limitation of sources of infection 
Six months after the third treatment, all trees of the experiment were examined for 
Fomes. The results are summarized in Table 5. According to these data, it can be stated : 
— in the group of contaminated trees 43 • 5 % of the trees are dead in the control 
against 12% of those treated with Calixin. 6 4 3 % of the trees treated with 
Calixin have no rhizomorphs against 3 5 3 % in the control. 
— in the group of trees with direct contact with the source of infection, only 7 2 % 
of the trees treated with Calixin became diseased against 32-8% in the control. 
— in the group of trees with indirect contact with the source of infection, only 4-6% 
of the trees treated with Calixin became diseased against 20-5% in the control. 
The results of this trial are remarkably similar to those of trials which were carried 
out at other places. They allow us to believe that by applying 2 1 of a 0 • 5 % Calixin 
emulsion every 6 months, Calixin protects the trees very well against the contamination 
with Fomes. 
DISCUSSION AND CONCLUSIONS 
The aim of the method of artificial inoculation is to cause in relatively short time the 
highest possible mortality in the infected plants. To reach this, a strong inoculum (20 cm 
diameter) was used as compared with the size of the test plants (1 5 cm). Moreover, the 
distance between parasite and host plant is very narrow. 
Death of plants was, therefore, apparent from the 3rd month after inoculation; it 
reached over 90% towards the 6th month. 
These results are in accordance with those of John who had shown that the 
capacity of the parasite to infect and to kill plants depends on the volume of the inoculum. 
The bigger the inoculum which serves as nutritive base of the parasite, the more is the 
incubation time reduced and mortality increased. This method permits thus to carry out 
rapid preliminary tests of the efficacy of a large number of fungicides against Fomes. 
The use of sand as substrate assures good penetration into greater depth, on the other 
hand it eliminates the influence of different soil factors like micro-organisms and physical 
and chemical properties. 
Among the products tested in our study, Calixin appears to be the most effective. 
We draw attention to Lim who has already shown its efficacy against Fomes 
by applying a mixture of Calixin and bitumen on the collar (" collar protectant "). 
The results of the second series of trials gave us more precise information on the rates 
and frequencies of application to be used. The study has shown that by applying 10 g 
a.i. of Calixin per tree, 3 times with 6 month interval, one can protect the stumps against 
Fomes attack. This concentration and this frequency arc, therefore, used in the field 
trials. 
489 
Table 5. Results offield experiment Fomes : Limitation of sources of infection 
(a) 1st group : Diseased trees : mycelium visible at the collar 
diseased trees 
treatment in Jan. 82 
dead trees 
in June 83 
trees remaining 
diseased 
in June 83 
trees without 
trace of mycelium 
in June 83 
number (N) N % N % N % 
control 85 37 43,5 18 21,2 30 .35,3 
treated with Calixin 
5 cc/1 and 2 1/tree 84 10 11,9 20 23,8 54 64,3 
(b) 2nd group : Healthy trees in direct contact with source of infection 
treatment 
diseased trees 
in Jan. 82 
dead trees 
in June 83 
trees infected 
in June 83 
trees remaining 
healthy in June 83 
number (N) N % N % N % 
control 131 6 4,6 43 32,8 82 62,6 
Calixin 125 1 0,8 9 7,2 115 92 
(c) 3rd group : Healthy trees in indirect contact with source of infection 
treatment 
diseased trees 
in Jan. 82 
dead trees 
in June 83 
trees infected 
in June 83 
trees remaining 
healthy in June 83 
number (N) N % N % N % 
control 88 3 3,4 18 20,5 67 76,1 
Calixin 87 — — 4 4,6 83 95,4 
The results of the trials for limiting the sources of infection have confirmed that, at 
an interval of application of 6 months. Calixin has a good preventive action in the field. 
Based on these results, all Hevea plantations in Ivory Coast have been treated with 
Calixin, using the following procedure. 
— Detection and Marking (once per year, May - June - July) 
A detection team of 5, led by a foreman, is responsible for 800 - I 000 ha. Each 
man examines systematically each single tree by removing the soil around the 
collar and the lateral roots in order to search for the presence of Fomes. If he 
finds signs of the parasite, he marks the tree as follows : a red plastic band around 
the trunk of the diseased tree and a white plastic band on the two trees in 
direct contact, provided these are healthy. 
490 
At the same time, he prepares a funnel like furrow (cuvette) to accommodate 
the fungicide. Dead trees are marked with black plastic bands, they are elimi­
nated by extraction. 
The foreman follows the five men of the team and takes notes. For each line he 
records the number of trees which are dead, diseased or neighbours and have 
to be protected. Then he calculates the total number of trees to be treated in 
each block. 
Treatment (twice per year, July - August and December - January 
After detection, another team carries out the treatment of the trees marked with 
plastic bands. The treatment consists of pouring 2 • 1 of 0 • 5 % Calixin emulsion 
into the cuvette. Twice at 6 months interval. 
Thi6 new method which is applied in Ivory Coast and Camerun since 1982 seems 
to give encouraging results : 
In actual fact, in June 1982, the diseased trees of one plot were treated with Calixin 
after detection was carried out. In June 1983, we had among 76 diseased trees 
8 dead ones (10-5%), 16 trees were still diseased (21 • 1 %) and 39 trees which 
were free from mycelium (68 4%). From the 84 trees in direct contact, we 
observed three trees which became diseased (3 6%) and 81 which remained 
healthy (96 4%). From the 53 trees in indirect contact, only one single tree 
became diseased. 
As compared with the traditional method of control (excavation + Fomac), 
the amount of hand labour required is considerably reduced. Actually, 80 — 100 
trees can be treated per man-day, depending on soil conditions. 
All the information accumulated in this experimental programme permitted to 
work out a control method which appear to be effective while its application 
is relatively easy. Although good and very encouraging, these results require 
long term confirmation : but considering the actual situation, one can hope 
to reduce effectively the losses due to Fomes by treating from the 2nd year until 
the 6th year after planting. 
491 
RtJBBER ROOT ROTS: CELLULAR AND MOLECULAR 
ASPECTS OF HOST-PARASITE INTERACTIONS 
By 
NICOLE, M . , GEIGER, J. P. AND N A N D R I S , D . 
(Laboratoire de Phytopathologie 0 RSTO M, Ivory Coast) 
Host-parasite relations between. Hevea brasiliensis and two African strains of wood 
rot fungi, Rigidoporus lignosus and Phellinus noxius, have been studied at both the cellular 
and biochemical levels. Aggression of the plant by each parasite, as well as the reactions 
elaborated against the diseases, are explained successively in detail in this publication. 
Host aggression 
Microscopic study 
Observations were performed on young plants, 6 months old and infected artificially. 
Penetration of hyphae into the tap root occurs via two distinct mechanisms : 
— passive penetration : this occurs through natural openings and wounds. In 
the case of R. lignosus, the rhizomorphs exert pressure on the underlying tissues 
until they are dislodged, thus facilitating the introduction of the fungal filaments 
into the cortical tissues ; 
— active penetration : each fungus causes the degradation of the periderm walls, 
after first perforating them. Intercellular spaces are also digested: in this case, 
host structures take on an alveolar appearance. 
During this stage, R. lignosus exhibits the feature of differentiating haustorium-like 
structures which have never before been described for this species of fungus. 
Colonization of cortical tissues is intimately related to the capacity of each fungus 
to degrade the walls of the host. Fungal filaments have thus been observed in intercellular, 
intracellular and interparietal positions. Regardless of the nature of the host walls (suberin, 
cellulose), the hyphae cause considerable damage. Some observations suggest the par­
ticipation of fungal enzymes in this process : 
— increased osmiophilia of the walls upon contact (or not) of the fungus ; 
— a disorganization of the fibrillar structure of cellulose and suberin ; 
— dislocation of the host cells. 
A similar increase of the osmiophilic properties of the middle lamella precedes its 
perforation and complete digestion. This is particularly evident in the phloem, which 
493 
is also the site of a particular disturbance to the rubber tree. Latice vessels undergo mod.-
fications under the effect of the infection, leading to the internal coagulation of the latexi 
Under the influence of metabolites, probably resulting from the degradation of the walls, 
the lutoid membranes burst. The resulting release of coagulating factors favours the 
gelling of the rubber into a mass. 
Even though the degradation of non-lignified root tissues of the rubber tree is similar 
in the cases of attack by R. lignosus and P. noxius, it is nonetheless more rapid with P. 
noxius, as a result of the higher activities of its hydrolytic enzymes. The colonization 
and degradation of the wood, on the other hand, is greater with R. lignosus. 
The hyphae of both R. lignosus and P. noxius penetrate the xylem via the parenchy­
matous rays. The vessels are invaded by either pits or after perforaticn of the walls. 
In this case, the filaments are inserted in the middle lamella by the initiation of micro-
hyphae. 
The degradation of vessel walls by R. lignosus generally begins at the level of the highly 
lignified fractions : the middle lamella and the primary wall. Under the action of the 
hyphae, the native structure of these cell components is disorganized, resulting in the 
appearance of a granular matrix, which is progressively perforated to form spaces. This 
erosion mechanism then spreads towards the secondary wall, whose S 2 layer is particularly 
deteriorated, in contrast to the S 3 layer. 
The mode of xylem degradation by P. noxius, on the other hand, is not selective. 
There is a simultaneous attack on walls rich in either lignin or in polysaccharides. Appa­
rently, no wall offers resistance to these fungal enzymes. 
Biochemical studies 
This line of research has been followed in order to determine the mechanisms of 
degradation of root tissues of rubber trees, as well as the differences in the enzymatic 
equipment which could explain the differences in behaviour between R. lignosus and P. 
noxius. 
From a technical point of view, the analyses have been carried out on extracts from 
healthy arid parasitized tissues, taken from young plants infected artificially or from tap 
roots of adult rubber trees infected naturally in a plantation. These studies have been 
extended by enzyme assays on culture filtrates from the two fungi. This procedure 
has enabled comparisons to be performed at several levels: 
@ Healthy tissues/parasitized tissues : leads to an inventory of the metabolic changes 
which can be correlated with the parasitic attack. 
@ Parasitized tissues/culture filtrates : comparison between enzymes detected in 
vivo and in vitro. 
@ R. UgnosuslP. noxius (extracts of parasitized tissues and/or culture filtrates) : 
specificity of the enzymatic equipment secreted by each parasite. 
494 
f he main results are as follows i 
(1) Compared to the enzyme activities which can be extracted from healthy tissues 
extracts from parasitized tissues exhibit a number of enzymatic modifications' 
both quantitative and qualitative. 
(a) Quantitative : an increase — occasionally considerable — of the activity 
of certain glycosidases (beta-glucosidase, alpha - and beta - galactosidase,) 
esterases (acid phosphatase) and proteases (leucine aminopeptidase). 
(b) Qualitative : These are enzymes activities not detected in healthy tissues : 
polyglycosidases (cellulase, xylanase, pectinolytic enzymes) and oxidases 
(laccase — p-diphenol oxidase). 
(2) R. lignosus and P. noxius both excrete glycosidases, polyglycosidases and oxidases 
(laccases) in vitro, but at activity levels which are specific to each fungus. Thus 
there is a predominance of laccases in the case of R. lignosus and of hydrolases 
in the case of P. noxius. 
These same quantative variations exist in parasitized tissues and are directly related 
to the nature of the pathogen responsible for the parasitic attack. 
It is thus probable that the abnormally high level of certain enzyme activities in para­
sitized tissues, as well as a fortiori the presence of enzymes specific to them (polygly­
cosidases and laccases) are to be attributed to the pathogen colonizing the tissues. This 
hypothesis has been verified for certain glycosidases (alpha-galactosidase, beta-glucosi­
dase) and for laccase and phosphatases : the enzymes (and isoenzymes) extracted from 
tissues parasitized by R. lignosus and P. noxius are in fact identical (starch gel electro­
phoresis) to the homologous enzymes excreted in vitro by each fungus, respectively. 
(3) The enzymes secreted by the pathogens are well adapted to the degradation of 
plant tissues. The action of enzymes contained in culture filtrates causes the 
release of various saccharides (monomers and oligomers) from insoluble chips 
of Hevea. 
Similarly, laccases must play a role in lignin metabolism. We have in fact been 
able to show that one of these enzymes (laccase Lj) produced by R. lignosus causes both 
the condensation and the depolymerization (in vitro) of lignin in the thioglycolic form* 
What conclusions or hypotheses can we draw from these, results ? 
(a) Both parasites have enzymatic equipment enabling them to degrade both poly­
saccharide structures and lignin. All these tissues should thus be colonized 
by each fungus. 
(b) Quantitative differences and especially the ratio of oxidative enzymes (laccase) 
and polysaccharidases should have an effect on the relative capacity (the rate) 
of the two fungi to degrade the different polymers of the walls, and thus the 
different root tissues. In practice, it is expected that P. noxius is more active 
495 
than R. lignosus on polysaccharide structures and thus in tissues winch ate 
richest in cellulose, xylan and pectinic materials, and the inverse should be true 
of highly lignified tissues. 
The comparison of lignin contents between healthy and parasitized tissues confirms 
that — in a general manner — both fungi can degrade both the polysaccharide fraction 
of these tissues and lignin. These results also agree with data obtained in vitro in trials 
of degradation of previously sterilized sticks of rubber tree wood. 
Histological studies have contributed more detail to this result, by the direct obser­
vation of the tissue degradations described above. In agreement with biochemical data, 
it would seem that P. noxius degrades cortical tissues more rapidly and more completely 
than R. lignosus, as a result of its very active hydrolases which are excreted. The latter 
fungus similarly degrades the woody lignin-rich structures more rapidly (primary wall, 
Sx and S 2 layers of the secondary wall) than the structures composed primarily of poly­
saccharides, i.e. the S 3 sublayer. P. noxius, on the other hand, degrades xylem substruc­
tures in an equivalent manner, regardless of their chemical composition. 
This latter observation is in apparent disagreement with biochemical data. It is 
nevertheless probable. 
— that the enzymatic equipment leading to the degradation of lignin is not a limi­
ting factor in the invasion of the xylem by P. noxius ; 
— that — on the contrary — the battery of hydrolases which are deficient in R. 
lignosus, limits the potentials of this parasite. 
This hypothesis presents the interests of accounting for both the histological and bio­
chemical data. It nontheless remains to be verified. 
Reactions of the host 
Numerous reactions elaborated by rubber trees have been characterized at all the 
stages of the infectious process of each parasite. Our own work has been carried out 
at the anatomical, cellular and molecular levels. The reactions are described in the order 
of their appearance during the infestation. 
Early reactions. They occur at the initial stages of penetration. The most characte­
ristic of these reactions is a hypertrophy and hyperplasia of young differentiated cells of 
the phellogen. This mechanism is often initiated around one or two necrotic cells, raising 
the possibility of hypersensitivity. Cytological tests carried out in parallel have shown a 
stimulation of the pool of phenolics localized under the points of penetration. 
Post-initial reactions. These reactions are elaborated during the penetration and 
nvasion of the first cell layers. The most consistently occurring of these reactions in 
Hevea is most certainly the increase of cell layers under the points of penetration. It 
results from a stimulation of the activity of the cork cambium. The walls of certain cells 
thicken concomitantly. The identification of the nature of this thickening showed that 
it was composed of suberin or lignin. 
496 
Reactions occurring during the colonization of tissues. All tissues, whether or not 
they arc ligaified, are involved by these reactions. In the young cork, for example, some 
cell walls are deformed, taking on a notched appearance. Microscopic observation has 
revealed wall appositions in the periplasmic space of the cells. When examined under 
ultraviolet light, these deposits fluoresce and are thus probably composed of callose. 
Similarly in the phloem, clumps of callose obstruct the pores of sieve tubes. 
In the region of the xylem, the lignification process becomes heterogeneous. Cell 
islets are differentiated before the cambium in the young phloem probably in order to slow 
down the progression of invading hyphae. This is also the role of the tyloses which 
obstruct the water bearing vessels of the wood. In addition, some of these vessels present 
a " new wall", laid down in the lumen in contact with the S 3 layer of the secondary wall. 
Late reactions 
The anatomical organization of the Hevea tap root is occasionally disorganized by 
the differentiation of new tissues. We can thus distinguish : 
— meristematic - like tissues initiated from the parenchymatous rays of the xylem. 
The organization of the cell foci is reminiscent of that of a meristem ; 
— phloem and ligneous tissues, set in place by a neocambium. Native diseased 
tissues are rejected towards the exterior or are isolated from the rest of the root. 
In greenhouse conditions, 10% of the plants infected with R. lignosus exhibit 
this phenomenon. It is probable that it is identical to the mechanism initiating 
healing callus formation in adult trees naturally infected. 
At the biochemical level, there is a considerable change in the spectrum of isoperoxi-
dases which can be extracted from different types of tissues. Healthy tissues are charac­
terized by the presence of numerous isoenzymes, while reacting tissues and especially 
parasitized tissues from adult tap roots (the phenomenon seems to be less specific in young 
subjects) are characterized by the presence of one major isoenzyme. In the most typical 
cases, its activity accounts for almost 100% of total peroxidase activity and corresponds 
to the stimulation of the homologous isoenzyme of healthy tissues by a factor of 100. 
This enzyme polymerizes coniferyl and p - coumarilic alcohols into insoluble pro­
ducts (DHP — dehydrogenase polymerisates) whose chemical structure is analogous to 
that of lignin. It is thus possible that this peroxidase contributes to an abnormally ele­
vated lignification of certain tissues, leading to the construction of a barrier, whose pur­
pose is to prevent the intertissue progression of the fungus. This type of wall thickening 
has in fact been observed. 
Is this an early or late process ? Our results are consistent with it probably being 
an early reaction : healthy tissues close to the front of parasite progression already seem 
to be the site of an analogous disturbance, although of lesser amplitude, 
497 
The lignin contents in different tissues studied would tend to confirm this hypothesis. 
Healthy tissues close to the front of parasite progression and above all the infected tissues 
taken from this front of progression have an abnormally high lignin content. It may be 
supposed that parasite invasion causes the host to react by lignifying. This reaction is, 
however, ineffective, since these tissues are subsequently degraded. 
In the case of infection by P. noxius, the host reaction is reduced, both in terms of 
peroxidase synthesis and lignification. This absence of reaction may be compared to 
the rapidity of the intertissue progression of the fungus. 
The global efficacy of all these reactions does not enable Hevea (GT1 clone) to resist 
attack by R. lignosus and P. noxius. When considered individually, however, certain of 
these reactions prevent the spread of hyphae in the tissues, which is the case for the cellular 
hypertrophy and hyperplasia and the supernumerary cortical cell layers under the points 
of penetration. Other reactions, on the other hand, are apparently much less effective, 
e.g. lignification and suberin synthesis of walls or callose deposits (wall apposition). 
Each of the parasites possesses enzymes which can degrade them. 
On the other hand, tissue neogenesis eliminates diseased tissues. N o contamination 
of new root structures has been observed after this process. This mechanism assures the 
survival of the plant, which assumes normal growth. 
The comparison of these defense mechanisms with the CODIT (Compartmentaliza-
tion of Decay in Trees) model shows, among other thing, that the process of tissue neo­
genesis is equivalent to wall 4 of this model. The purpose of this barrier is to prevent 
the spread of decay to differentiated tissues after the parasitic attack. Finally, the com-
partmentalization mechanism of wood rots, already described in the case of other ligneous 
plants parasitized by other wood rot agents, is also applicable to the host parasite couple 
H- brasiliensis j R. lignosus «nd P. noxius. 
498 
AN INTEGRATED APPROACH TO CONTROL OF WHITE 
ROOT DISEASE IN SRI LANKA 
By 
A . de S. LIYANAGE, O . S. PERIES, S. S. WARNAPURA, E . A. T. SENADEERA AND 
W, AMARATUNGA 
ABSTRACT 
White root disease caused by Rigidoporus lignosus can be controlled at three stages 
of cultivation of rubber viz. pre-planting, planting and post-planting. In the first stage, 
it is essential to demarcate infected patches encompassing three to four rows of trees 
bordering infected patches and three to four rows of stumps buried closest to the perimeter 
of infected patches. Removal of most of the infected roots as practicable and burning 
them in situ is advocated. Changing the planting distance between replanting cycles has 
helped to reduce the incidence of white root disease in the new clearing. Creeping legu­
minous covers enhanced the decay of root debris, encouraged the growth of antagonistic 
flora and dissipated the highly inocuous infected roots into ineffective ones. At planting, 
sprinkle 114 g of sulphur around each plant. Application of sulphur should be confined to 
plants in demarcated patches where infected trees stood in the old stand. Amendment of 
soil with sulphur helped to reduce the pH for periods up to 18 months and encouraged 
the growth of Penicillium and Aspergillus species which were antagonistic to R. lignosus. 
Sprinkling sulphur on the surface was as effective as mixing it with the soil but the latter 
was more expensive. At post-planting, detection of infected trees by foliar symptoms 
was confined to demarcated patches. This helped in the early detection of infected trees. 
The application of Collar Protectant (Pentachloronitrobenzene) on the roots of neigh­
bouring trees prevented further spread of the disease. Preliminary experiments have 
indicated that drenching the soil with Calixin (tridemorph) has helped to kill the super­
ficial strands of the mycelium on roots of trees in early stages of infection. The possibi­
lity of using some selfed clonal seedlings as resistant root stocks is also discussed. 
INTRODUCTION 
White root disease (WRD) of the rubber tree, Hevea brasiliensis Muell. Arg., caused 
by the fungus Rigidoporus lignosus (Klotzsch) Imazeki is the most destructive root 
disease in Sri Lanka. Rubber trees of any age or vigour succumb to the disease, killing 
them directly, bringing about the largest number of tree losses in the rubber plantations 
(Liyanage, 1977). Several reasons could be attributed to the widespread occurrence of 
WRD in Sri Lanka. The insidious nature of the disease is often not realised until after 
it has become well established in the plantation. Also, there is a tendency to neglect 
the treatment of root diseases because the benefits of control are not immediately felt. 
This is further aggravated by the suspension of control measures during periods of financial 
difficulties (Fox, 1977). The damage caused during the immature period results in the 
reduction of the stand leading to a lowering of the production and therefore, the export 
earnings from rubber which accounts for nearly 11 % of the gross national income. The 
advances in research made over the last two decades have led to the development of 
more efficient and economic methods of control. Yet, the disease has not been satisfac­
torily controlled mainly due to the negligence of growers. 
499 
This paper examines the present status of W R D in Sri Lanka and discusses how 
cultural, biological and chemical methods are used singly and in combination to combat 
its spread. 
EXPERIMENTAL 
Extent of damage 
Disease incidence : Several rubber estates covering an extent of 2128 ha were surveyed 
in 1977 — 78, to assess the extent of damage caused by R. lignosus. The damage was 
higher in Ratnapura, Kelani Valley, Kalutara and Galle regions. The infection was 
lower in the drier areas of Kurunegala and Matale (Table 1). 
Table 1. Damage caused by white root disease in different rubber growing areas 
Areas Extent of damage 
(%) 
Ratnapura 8-9 
Kelani Valley 8 4 
Kalutara 7 0 
Galle 6 6 
Kegalle 4 -8 
Kurunegala 1 5 
Matale 0 3 
When the clearings of all areas surveyed were grouped according to their age, it was 
shown that the incidence was low in the first 5 years of growth but steadily increased to 
reach a peak in the 20 — 25th year of planting or just before uprooting. A recent esti­
mate of W R D incidence was not made in the smallholdings, which occupies nearly 70 
percent of the total area under rubber. However, Wijewantha (1964) showed that the 
highest infection occurred during the first 3 years in a replanting, after which there was 
a drop in the rate of infection. 
Pre-planting methods 
Disease demarcation : New centres of infection in the replantings arise from the infected 
root debris left in the soil. These usually occur at the border of large open confluent 
infected patches of the old stand, or close to infected trees of the original stand if the 
disease occurs in small or isolated patches (Fig. 1). 
Infections in the replanting rarely originate from the centre of infected patches, as 
most of the stumps have decayed beyond the stage which can support the growth of the pa­
thogen. However, upto four rows of stumps usually buried in the soil, closest to the patch 
were shown to harbour the pathogen, roots of trees bordering the periphery of patches 
showed the least amount of decay and contained highly viable inoculum. The remaining 
stumps, lying towards the centre of patches, showed a progressive increase in the decay 
and a decline in the viability of the inoculum. Similarly, the roots of up to four rows 
of apparently healthy standing trees round infected patches were infected with R. lignosus. 
500 
X . STANDING TREES IK THE OTtlGKTAL STAND. 
m INFECTED TRESS Itf 1rt£ ORIGINAL STAMfl. 
e HEW P1ASTINS POINTS. 
© NEW 'INFECTION. 
Fig. 1. Sites of new infections in the replanting in relation to the infected debris in the old 
The actual number of infected rows of trees and stumps varied depending on the planting 
distance. The spread was greater if the inter row space was closer and less if it was wider 
as in the avenue system of planting. The clearings in experiments done during the period 
1975 — 78 (Liyanage, 1976, 1977 and 1978) did not show any new infection sites in areas 
which had healthy trees in the old stand, except when infected stumps were introduced 
to these areas by rolling them down the slopes during the process of land clearing. 
Therefore, the area to be replanted must be carefully inspected to demarcate all 
infected patches prior to uprooting the old stand, encompassing 3 — 4 rows of apparently 
healthy trees and 3 — 4 rows of stumps buried in the soil towards the centre of the patches. 
This helped to identify the areas containing active sources of inoculum, so that special 
attention can be paid to eradicate them. 
Uprooting the old stand: This entails removal of rubber trees in the old stand, to ensure 
that little or no infected roots are left in the soil as it was recognised that infection can 
arise only by root contact (Petch, 1921). The mycelium of R. lignosus did not survive 
in the soil, when removed from its food base (Bancroft, 1912) and it did not cause infection 
except from a large piece of infected root, as it is a weak parasite (De Jong, 1933). The 
results of a number of experiments reported by Fox (1961b), indicated that complete 
eradication of roots was uneconomic when considering the number of trees saved. How­
ever, under Sri Lankan conditions Peries (1963 ; 1974), showed that removal of all the 
potential sources of infection was essential when the incidence of W R D was high in the 
old stand. More recently, Liyanage (1978) laid out several field experiments to examine 
the methods of clearing land prior to replanting. 
The distribution of roots in trees at different soil depths was examined. It was 
observed that 90% of the lateral roots of rubber trees were present within the first 60 cm 
soil depth (Fig. 2). 
The roots within this region were usually larger than those below this region. A 
high percentage of the infected roots were also present in the upper 60 cm layer of soil. 
The tap root length varied (range 107 — 262 cm) according to the soil condition, type 
of clone and other factors but the mean length of those measured was 158 cm. The 
mean length up to which the infection was detected ranged from 91 — 186 cm with a mean 
of 125 cm. 
Table 2. Distribution of lateral roots at different soil depths 
t 
0—30 cm 31-1-60 cm 61—90 cm 
7 0 1 9-8 
65 9 24 4 9-7 
Type of root 
Total (%) 
Infected (%) 
These observations have helped to identify the area in which active sources of inoculum 
were present, so that proper attention can be paid to remove as much of the large pieces 
of inoculum as practicable, within demarcated patches. All infected roots must be collected 
and burnt preferably in situ, to prevent accidental introduction into areas hitherto free 
of the disease. 
502 
4 * 
When differ ent sizes of roots of a standard length were buried in planting holes, the 
small roots ( < 0 • 6 cm diameter) lost their viability in 6 months and failed to cause infection 
of rubber plants. However, larger laterals viz. 1-2, 2-5 and 5 0 cm diameter, caused 
20% , 40% and 80 % infection, respectively, within 1 year of their placement in the planting 
hole (Table 3). 
Table 3. Effect of size of inoculum in the planting hole in infection of rubber plants 
Inoculum diameter Infection 
(cm) (%) 
0 6 0 
1-2 20 
2-5 40 
5 0 80 
At this stage, most of the 2 5 and 5 0 cm diameter inocula remained viable. When 
infected stumps or large laterals were placed at a distance of 180 and 60 cm respectively, 
away from the plants in the row, infection occurred in the plants adjacent to inocula in 
18 months. 
The superiority of uprooting the infected sources, as compared to leaving unpoisoned 
stumps cut at ground level, was evident from results shown in Table 4. 
Table 4. Effect of methods of clearing on tree losses under different covers, 5 years after 
planting 
Tree losses (%) 
Method of clearing Naturals Legumes 
Stumps uprooted and burnt 2 7 2 5 
Stumps uprooted and stacked 2 5 2 9 
Stumps cut at ground level (not poisoned) 6 3 7-8 
The process of decay in untreated stumps was slow. When diseased trees were cut 
at ground level, the parasitic activity was increased making the stumps active sites for 
transmitting the disease to trees in the new stand, until they have become moribund. 
The steep terrain in Sri Lanka precludes the use of expensive mechanical clearing 
methods, because of the high risk involved in exposure of soil to erosion, although it 
is a recommended practice in Malaysia (Newsam, 1964). 
Poisoning methods : Sodium arsenite was used by Napper (1939) to poison the old stand 
of rubber at the time of clearing land. This was replaced with the 5% butyl ester of 
2, 4, 5 - T in diesolene, and it was significantly better than sodium arsenite in enhancing 
505 
tlic decay of rubber trees or stumps (Newsam, John and Rao, 1961).Tree poisoning was 
superior to stump poisoning (Hutchinson, 1961 ; Fox, 1961a). The freshly cut surfaces 
of stumps were prone to infection by the wind-borne basidiospores (John, 1964 ; Lim, 
1976), and the application of creosote, a wood preservative, on the surface of stumps 
inhibited spore colonisation. The use of this clearing procedure was discouraged in 
Sri Lanka (Young, 1964). Later, a series of experiments were conducted by Liyanage 
(1975), in highly infected and uninfected clearings, located in wet rubber growing districts 
and in slightly infected areas in the dry districts. These trials were conducted to test the 
efficacy of some chemicals on stump decay viz. urea, borax, 2, 4 ,5, - T in diesolene, with 
uprooting and burning stumps and leaving untreated stumps as controls. It was observed 
that 2, 4 , 5 - T induced decay markedly faster than the rest of the chemicals and the un­
treated controls. Although application of creosote to the cut surfaces was aimed at 
preventing infection through basidiospores, none of the untreated stumps were infected 
in this manner. Poisoning the stumps could not eliminate the disease from infected 
trees. When the old stand was free of W R D , new infection did not occur in the replantings 
located in the wet districts. Stumps of trees already infected, transmitted the disease 
to new plants, irrespective of the treatment to which they were subjected. Infected stumps 
left in the soil of the old stand gave rise to new infections in replantings of both wet and 
dry areas. 
Poisoning trees of the old stand was not included as a treatment as the falling branches 
or trees could damage the immature trees and also it would incur an additional cost, to 
remove the debris. 
Altering the planting distance : Changing the planting distance between replanting cycles, 
a method hitherto not attempted, as a means to control WRD, has resulted in an early 
reduction of the incidence of the disease in the new clearing (Table 5). Preliminary ob­
servations recorded an year after planting indicated that losses were higher when the 
planting distance was the same or similar to that of the original stand, even when the 
infected sites in the old stand were relatively low (viz. 31 %). There was a reduction in 
the infection when the interrow space was increased, although the number of potential 
sites of infection was high (viz. 43 4%). 
Table 5. Effect of planting distances on tree losses after one year 
Planting distance Infected trees* in Losses in the 
the old stand new stand 
0 (%) 
2 4 X 9 1 
* Includes infected trees and vacancies due to death of trees 
** Planting distance of the original stand 
(m) 
4 6 X 4 9** 46 9 2 6 
3 6 X 6 1 32 3 1 6 
3 3 X 7-3 31 0 2-5 
43-4 0-3 
506 
Cover crops : The beneficial effects of creeping leguminous covers and the undesirable 
effects of woody shrubs as indicators of potential foci of infection have been reviewed 
(Hutchinson, 1961 ; Fox, 1961 b, 1965a, 1971, 1977 ; Liyanage, 1980). The fungus 
freely infects the roots of covers encouraging its spread but not necessarily the disease. 
In the process of dissipation of inoculum they get diffused into small ones which are less 
dangerous and ineffective. Covers also provide an environment which encourage rapid 
decay of roots and timber by saprophytic organisms, causing them to exhaust their food 
reserves (Fox, 1961a, 1956a, 1977). 
Evidence that creeping leguminous covers reduce the incidence of WRD in 
replantings, compared to the naturals (a mixture of grasses and legumes) could not be 
established. This was because legumes were more abundant than grasses in the 
mixture. However, the decay of roots was accelerated beneath the pure legume covers. 
The activity of R. lignosus was considerably reduced when it was grown on soils taken 
from under a cover of Pueraria phaseoloides (Table 6). 
Table 6. Growth of R. lignosus on soils obtained beneath Pueraria and clean weeded soil 
under laboratory conditions 
Type of cover Mean rhizomorphs Mean length of 
N o . distinct 
rhizomorphs (cm) 
Pueraria phaseoloides 1 2 2 2 
Clean weeded soil 4 6 5 6 
However, the number of rhizomorphs and their length showed different patterns 
when grown on different soil types, collected under a cover of Pueraria (Table 7). 
Table 7. Growth ofR. lignosus on four soil types collected beneath a cover of Pueraria, 
under laboratory conditions 
Soil series Mean rhizomorphs 
No . 
Mean length of 
rhizomorphs (cm) 
Boraiu 3-3 2 0 
Ratnapura 4 4 2-2 
Agalawatta 5-3 2 9 
Matale 5-5 1-7 
Planting methods 
Soil amendment : The use of sulphur (114 g) as a soil amendment in the planting hole, to 
reduce the incidence of WRD was first demonstrated by Peries (1965, 1966,1972). Later, 
Satchuthananthavale (1971), Liyanage (1980), confirmed the earlier findings of Peries. 
507 
Liyanage (1980) showed that the application of 114 g of sulphur at the time of planting 
was effective in reducing new infections in the replanting, especially when most of the 
large infected roots of the old stand were removed. The number of infections increased 
in the absence of sulphur and inadequate removal of inocula, but in its presence losses 
were markedly reduced (Table 8). 
Table 8. Effect of removal of infected roots and application of sulphur or tree losses after 
18 months 
Treatment Cumulative losses (%) 
+ G + S 0 
+ G — S 1-2 
— G + S 0 5 
— G — S 4 7 
+ G = roots removed 
— G = roots left in the soil 
+ S =» 114 g sulphur applied 
— S = sulphur not applied 
A sharp decrease in soil pH was recorded 3 months after the application of 114 g 
and 228 g of sulphur, the latter being slightly more effective than the former (Fig. 3). 
The increase in acidity was accompanied by a concomitant increase in the species 
of Penicillium and Trichoderma. The population of the latter nearly doubled at the soil 
surface while that of Penicillium declined but recorded a sharp increase at 15 and 30 cm 
depths. There was no change in the fungal population of Aspergillus and several uniden­
tified species (Liyanage, 1980 ; Liyanage, Peries and Liyanage, 1983). They also observed 
that some of these fungal species can cause lysis of the mycelium of R. lignosus, showing 
their antagonistic activity. 
Sprinkling of sulphur and forking it into the soil after planting was cheaper than 
mixing it with the soil before planting or mixing 57 g of sulphur with the soil of the lower 
half of the planting hole and sprinkling the balance on the surface, after planting (Liyanage, 
1980). Most of the roots were confined to the planting hole up to about 6 months after 
planting. In 12 months most laterals grew out of the planting hole. Therefore, one 
application of sulphur was sufficient and should be confined to planting holes within the 
demarcated patches. 
Post-planting methods 
Disease detection : The detection of infected trees periodically by the time consuming 
collar inspection method was widely adopted in Malaysia (Napper, 1932 a, b ; 1938), 
until Alston (1950 c) and Fox (1961 a, b) found that wounds caused during collar inspection 
increased the disease susceptibility. The recognition of infected trees by foliar symptoms 
is less expensive, fast and can be done more frequently (Newsam, 1964). Further investi­
gations showed that it was not worthwhile to cure infected trees showing foliar symptoms 
508 
228 9 OF SULPHUR 0 - 10 cm 
UK g OF SULPHUR 0 - 10 cm 
a~© 228 g Of SULPHUR 30 • 40 
t-A 111 9 OF SULPHUR 50 - *»0 
f • r ' • \ •« 1 
3 6 9 12 15 18 21 24 
MONTHS 
Fig. 3 . Change to soft pH after amendment with sulphur. 
509 
but they served a useful purpose in indicating foci of infection so that the neighbouring 
trees could be protected (Fox, 1961 a, b). The use of this method was also recommended 
in Sri Lanka (Peries, 1970). In Ivory Coast, detection by baiting (Declert, 1964 ; Martin 
and de Plessix, 1969) was recognised as a cheap, effective and less injurious method of 
identifying the infected trees. However, Fox (1977) observed that apart from obviating 
wounding it was less advantageous in that it resulted in misidentification with P.. zonalis 
and further, there was no indication of the activity and size of the inoculum in the soil. 
The use of shrubs to indicate the centres of infection was advocated by Bertrand and 
Minor (1937) but Newsam (1963) and Fox (1961a) pointed out that its economic disad­
vantages far outweighted their value as a root disease indicator, as it would perpetuate 
the infection and transfer it to immature rubber trees. 
Treatment of immature trees : The earliest method of treatment of white root disease was 
to uproot the infected trees (Petch, 1921). Later, Napper (1932 a, b ; 1938) recommended 
excavation and removal of all infected roots after tracing them. Further, an application 
of 2 percent solution of copper sulphate was also recommended after scraping the super­
ficial mycelium from healthy roots. As an alternative to copper sulphate, Riggenbach 
(1958) recommended the use of 1 percent Tillex, an organomercurial fungicide as a soil 
drench. This method gave disappointing results as a high percentage of treated trees 
later succumbed to the disease (Young, 1954). Later, Peries, Fernando and Samaraweera 
(1963, 1965) proved that it was not essential to use fungicides to control white root disease 
but mere removal of the food bases was sufficient. A grease-based formulation containing 
20 % pentachloronitrobenzene was developed to be applied on the collar of trees, adjacent 
to infected trees to prevent the epiphytic growth of the fungus (Fox, 1965 b, 1966). This 
method of treatment was very effective and helped to prevent the spread of the fungus 
along the rows. It later became a standard practice in the procedure to control white 
root disease in Sri Lanka (Peries, 1970). In Ivory Coast, the manual eradication process 
was replaced with the use of Calixin (Tridemorph), which gave satisfactory results when 
used as a soil drench (Tran Van Canh, 1984). In preliminary trials conducted in Sri 
Lanka, the same fungicide when used as a soil drench, killed the superficial mycelial growth 
of trees in the early stages of infection (Liyanage, 1984). This method has to be fully 
evaluated under field conditions before a firm recommendation can be made. 
Treatment of mature trees : If the recommendations are properly understood and promptly 
implemented, the mature clearings should be free of the disease (Fox, 1964 a). However, 
in Sri Lanka the incidence of WRD increased with age (Liyanage, 1977), mainly due to 
the failure to detect and control the disease during the early stages of the growth of the 
rubber plant. In addition, careless disposal of infected boles and roots, especially during 
the period of clearing land, allowing unprotected stumps, branches and wounds to be 
exposed to spore colonisation have also led to the increased incidence of W R D in mature 
stands. Therefore, prompt attention to seal-off the wounds with a suitable wound dressing 
such as creosote should help to eliminate the risk of spore colonisation. 
Usually in this method of control the trees that can be saved are those adjacent to 
infected trees. In the course of treatment of infected trees in trials conducted by Liyanage, 
Wettasinghe and Dharmaratne (1977) it was necessary to reduce the canopy weight by 
cutting some branches prior to treatment. All infected roots were dug out. The re­
maining healthy roots were served to leave a gap of about 30 cm to prevent the fungus 
510 
advancing towards the bole of the tree. All the cut ends were protected with tar. Before 
refilling the soil, a fungicide containing PCNB as the active ingredient was applied round 
the collar and the upper end of the tap roots and laterals to cover a distance of about 
30 cm. In the control treatments, excision of infected roots and scraping the superficial 
mycelial growth were sufficient to help the plant to recover by producing new roots. Two 
men can only treat four 1 2 — 15 year old trees in this manner and would require one 
litre of the Collar Protectant. 
Cutting trenches in mature plantations to check the spread of the disease has met 
with little success (Anon, 1974), The difficulty in the proper siting of the trenches 
and inability to maintain them by regular rounds for desilting have been the contributory 
factors for using this method only as an emergency measure in the control of WRD. 
Clonal susceptibility 
A technique for screening rootstocks resistant to R. lignosus was developed by 
Liyanage (1978). He showed that LCB 1320 was consistently tolerant to this fungus. 
Number of other clones showed varying degree of susceptibility with RRIC 36 being 
highly susceptible to the disease. Absence of absolute resistance is due to the fact that 
some seeds in the population that were used for screening could have been cross pollina­
ted, although the seeds were collected from the centre of a large plot. 
Exploitation of bare patches 
Utilization of bare land: A suggestion was made by Liyanage (1977) to utilize the bare 
patches created by loss of trees due to W R D , by growing annual and perennial crops. 
One of the draw backs for proper implementation of this scheme is that the vacant patches 
of various sizes are scattered throughout the estate and protecting any crop grown in 
them especially at harvest time against thieving was difficult. It was also suggested to 
plant quick maturing trees such as Alstonia, but the main disadvantage in that was the 
roots of these trees can perpetuate the fungus and could pass it on to immature trees at 
replanting. As an alternative, establishment of a good cover will help to enhance decay 
of infected roots left in diseased patches. It would be prudent to sacrifice 3 — 4 rows of 
trees round each patch after exploiting them to get the maximum amount of latex. This 
should be done when the patches are small so that only a few trees are uprooted. This 
hypothesis has not been tested under experimental conditions yet. 
DISCUSSION 
The demarcation of diseased patches has helped to eliminate much of the sources 
of infection and to economise on the use of sulphur by confirming its application to the 
planting holes within patches. Further, post-planting detection of infected trees by 
foliar symptoms has been made more effective due to careful inspection of plants within 
patches of the original stand. Infections in the new stand always occur near the border 
of infected patches, where the carry over of the infective material is in the form of large 
stumps that are buried in the soil. Also, felling of trees increases the activity of the fungus 
of infected roots, which results in the increase of the size of the food base by colonising 
the remaining roots thus increasing the risk of infection in the clearing. 
511 
The number of losses in the new stand is low and remains within manageable limits, 
although the potential foci for infection, in the old stand was very high. Further, results 
of the survey also revealed that the incidence was low within the first five years after planting. 
In the process of uprooting most infected laterals are removed as they are confined to 
the upper layer of the soil, unless they are cut where they join the bole of the tree to faci­
litate easy removal of trees or decayed due to the disease. Therefore, most of the sources 
of infection are removed during land preparation. However, stumps bordering infected 
patches are potential sources of infection in the new stand. Removal of these sources 
which remain active for 2 — 3 years could help to eliminate the new infections in the 
clearing. A thorough cleaning up, however, is precluded on economic grounds and 
practical difficulties. Substantial evidence is present to show that pre-planting eradica­
tion of the infected material is adequate. Therefore, these results are in agreement with 
those in Malaysia (Fox, 1965a, 1977) and which showed that a thorough pre-planting 
eradication gives only a marginal benefit and is uneconomic. 
Poisoning of stumps with 2, 4, 5 — T helped to kill roots quickly and allowed them 
to be rapidly invaded by saprophytes which have a high competitive ability. Poisoning 
of stumps could be done in dry rubber growing areas and also in the wet areas provided 
infection in the old stand is absent. But stump poisoning is not suitable for heavily 
infected areas as poisoning cannot eliminate the disease from trees already infected. 
Creosoting the cut surface of stumps has been recommended in Malaysia to prevent 
colonisation by basidiospores (Lim, 1976) but under conditions prevalent in Sri Lanka 
none of the untreated stumps was colonized by spores (Liyanage, 1980). 
It has also been shown that change of planting distances between replanting cycles 
help in reducing the incidence of disease in the new clearing. This is possibly due to 
locating the new planting points away from the sources of infection there by increasing 
what Fox (1977) called the " grace period ". When roots of new plants come into con­
tact with the old roots most of them have lost their ability to transmit the disease. 
Beneficial effects of creeping leguminous cover plants have been reviewed by Fox 
(1961a, 1965a and 1977). In the present study, no differences were detected between the 
legumes and naturals in reducing the tree losses due to WRD, possibly because more le­
gumes were present in the natural plots masking the less beneficial effects of grasses. How­
ever, the surface soil containing the leaf litter showed an antagonistic effect on rhizomorpb 
production and growth, compared to soils from a clean weeded area. These findings 
agree with that of Fox (1965a and 1977). Our preliminary study however, indicated 
that there was no difference in the rate of decay of infected roots when placed beneath 
several creeping leguminous covers. Some differences in the growth of R. lignosus was 
noted in soils obtained beneath a cover of Pueraria grown on different soil types, indicating 
that some other factors are involved. These aspects require further investigation. Cover 
plants encourage the spread of the pathogen amongst their roots which helps in reducing 
the effectiveness of the inoculum. However, when the cover was not kept away from the 
base of the collar, infection of the collar region resulted in death of plants. Regular 
weeding to keep the cover away from the trees is essential. 
The application of sulphur to soil decreased the acidity and increased the antagonistic 
mycoflora (Liyanage, 1980). Sulphur was particularly effective when most of the large 
512 
laterals were removed. Mixing sulphur with the top soil after planting is better than 
other methods, as it helps to effectively reduce the pH and increase the antagonistic flora 
in the soil of the planting hole. 
The increase in the incidence of W R D indicates that either control measures are 
neglected or inadequately carried out. The spread of the fungus within the row can be 
stopped if the collar of trees neighbouring infected trees are treated during the early stages 
of the immaturity period. Failure to do this results in the easy transmission of the disease 
due to over lapping of the roots between plants in the row, eventually leading to death 
of plants and forming large open confluent patches, which are left fallow until the 
next replanting cycle. 
The possibility of using selfed clonal seedlings which are resistant to W R D opens 
up exciting prospects to control this disease (Liyanage, 1978). However, further work 
has to be carried out before this can be done in field scale. 
SUMMARY OF CURRENT RECOMMENDATIONS 
Pre-planting treatments (cultural methods) 
1. Do not establish nurseries in patches affected by white root disease in mature clearings. 
2. Demarcation of diseased patches to include 3 — 4 rows of infected trees and stumps. 
3. Uproot and burn all infected roots and boles as practicable paying special attention 
to eradicate roots in the first 30 cm depth, around periphery of infected patches. 
4. Change the planting distance between replanting cycles, where ever possible. Locate 
the new planting row inbetween two rows in the old stand. 
5. Establish a creeping leguminous cover crop. 
6. Keep the leguminous cover away from the base of the tree by circle weeding. 
Planting treatment (Biological methods) 
1. Amend soil with 114 g of sulphur 
(i) Limit the application of sulphur to the plants within the demarcated patches, 
which were known to be heavily infected in the old stand. 
(ii) Sprinkle sulphur on the surface to cover about one square meter leaving a IS cm 
diameter circle round the base of each plant. 
(iii) Take special care to prevent direct contact of roots with sulphur as it scorches roots. 
(iv) Mixing sulphur with the soil iu the planting hole and sprinkling sulphur on the 
surface of soil, gave similar results but the latter was cheaper and helps in checking 
whether or not sulphur was applied. 
513 
Post-planting treatments (Chemical methods) 
1. Carryout foliar inspections of trees, every 3 months, commencing 6 — 9 months 
after planting, up to the time of tapping trees. 
2. Confine the foliar inspection to the trees in and around the demarcated patches. 
3. Uproot the trees showing advanced stages of foliar symptoms as they are beyond 
treatment. It may be possible to save such trees, if the infection is detected early, 
but it is not essential. 
4. Trace the sources of infection in the planting row and remove them. 
5. Inspect the roots of trees on either side of infected trees. 
6. Excise infected roots (latex will not exude when pricking with a pen knife) and scrape-
off superficial mycelium from healthy roots. 
7. Apply a fungicide containing pentachloronitrobenzene (PCNB) as an active ingredient 
(Collar Protectant or Fomac 2), to cover about 30 cm of the tap and lateral roots 
after carefully removing the soil adhering to the roots. 
, 8. Collect all infected roots and burn them at the site. D o not introduce infected roots 
into areas hitherto free of the disease. 
9. Refill soil taking care not to incorporate infected roots. 
10. Identify treated trees with a distinctive paint band and also indicate the date of treat­
ment. 
11. Supply vacancies with polybag plants in the first year, for which establish about 40 
plants per hectare, at the time of planting the clearing. 
12. To supply vacancies after the 2nd year, as stump buddings raise 40 plants per hectare, 
in trenches, these should be located in an area which was free of white root disease 
in the old stand. 
13. Reinspect treated trees after two years and retreat them, if necessary. 
14. Just before the final thinning and tapping, carryout a tree-to-tree survey to detect 
trees which are low to show symptoms of the disease. 
REFERENCES 
ALSTON, R. A . (1950). c. Path. Div. Report for the 1948, Rept. Rubb. Res. fust. Malaysia, 
1945-1948, p. 4 1 — 68. 
A N O N , ( 1 9 7 4 ) . Root diseases control. Firs' Bull. Rubb. Res. Inst. Malaysia. No. 1 3 4 , 
p. 157— 164. 
BANCROFT, K . (1912). A root disease of the para rubber tree. Dept. Agr. Federated 
Malay. States, Bull. 1 3 , p. 30. 
514 
CANM, T R A N V A N , ( 1 9 8 4 ) . A new method of direct control of Rigidoporus lignosus, 
causal agent of white root disease of Hevea, Proc- Int. Rubb. Conf. Colombo 1 9 8 4 . 
(in press). 
DECLBRT, C. ( 1 9 6 1 ) . Une technique nouvelle de detection des agents de pourridie la 
buchette —piege. Son application a l'elude du Leptoporus lignosus (KT) Heim. Revue 
Mycol. 26 p. 1 1 9 — 1 2 7 . 
D E J O N I G , W. H. ( 1 9 3 3 ) . Hot parasitisme Van Rigidoporus microporus (Swartz) van 
Overeem, Syn. Fomes lignosus. Klotzsch. bij Hevea brasiliensis. Archf. Rubber cult. 
Ned. Indie. 17, p. 8 3 — 1 0 0 . 
Fox, R. A. ( 1 9 6 1 a). White root disease of Hevea brasiliensis recent developments in 
control techniques. Com. Mycol. Conf. 6th Rept. p. 4 1 — 4 8 . 
Fox , R. A. ( 1 9 6 1 b). White root disease of Hevea brasiliensis the role of fungicides in 
control techniques, Com. Mycol. Conf. 6th Rept. p. 9 7 — 1 0 0 . 
Fox, R. A. ( 1 9 5 7 ) . Decaying timber spreads root diseases. Plrs' Bull. Rubb. Res. Inst. 
Malaysia, 29 p . 3 1 — 3 5 . 
Fox , R. A. ( 1 9 6 5 a). The role of biological eradication in root disease control in re-
plantings of Hevea brasiliensis, Ecology of soil borne plant pathogens, p. 3 4 8 — 3 6 2 . 
Eds. K. F. Baker and W. C. Snyder. Berkeley : Univ. of Calif. Press. 
Fox, R. A. ( 1 9 7 1 ) . A comparison of methods of dispersal survival and parasitism in 
some fungi causing root diseases of tropical plantation crops. Root diseases on 
soil borne pathogens, p. 1 7 9 — 1 8 7 . Eds. T. A. Toussoun, R. V. Bega and P. H . 
Nelson. Berkeley : Univ. of Calif. Press. 
Fox, R. A. ( 1 9 7 7 ) . The impact of ecological cultural and biological factors on the strategy 
and cost of controlling root diseases in tropical plantation crops as exemplified by 
Hevea brasiliensis. Conf. Proc. Rubb. Res. Inst. Sri Lanka. 5 4 , 1 ( 2 ) p. 3 2 9 — 3 6 2 . 
Fox, R. A. ( 1 9 6 4 a). A report on a visit to Nigeria to make a preliminary study of root 
diseases of rubber. Research Archive Rubb. Res. Inst. Malaya, 21, p. 3 4 . 
H U T C H I S O N , F. W. ( 1 9 6 1 ) . Factors affecting root disease incidence and control in re-
plantings. Proc. Nat. Rubb. Res. Conf. Rubb. Res. Inst. Malaya, 1 9 6 0 , p. 4 8 3 — 4 9 5 . 
J O H N , K. P. ( 1 9 6 4 ) . Spore dissemination of root diseases. Plrs' Bull. Rubb. Res. Inst. 
Malaya. No. 75 , p. 2 3 3 — 2 3 7 . 
L I M , T. M . ( 1 9 7 6 ) . Spore production germination infection in Rigidoporus lignosus on 
Hevea. Journal Sains Pusat Penyelidikan Getah Malaysia, Jilid 1 , Bilanganl, 15 . 
L I Y A N A G E , A. de S. ( 1 9 7 5 ) . Review of the plant pathology Dept. Ann. Rev. Rubb. 
Res. Inst. Sri Lanka. 
5 1 5 
LIYANAGE, A . de S. (1976). Review of the plant pathology Dept. Ann. Rev. Rubb. Res. 
Inst. Sri Lanka. 
LIYANAGE, A. de S. ( 1 9 7 7 ) . 1 . Studies on the variability and pathogenicity of Rigidoporus 
lignosus. Jl. Rubb. Res. Inst. Sri Lanka, 54 , 3 6 3 - 3 7 2 . 
LIYANAGE, A. de S. ( 1 9 7 7 ) . 2 . The distribution spread and control o f black root disease 
in Sri Lanka. Abstract 33rd Ann. Ses. Sri Lanka Association for the Advancement 
of Science, 1 9 7 7 . 
LIYANAGE, A. de S. ( 1 9 7 8 ) . Review of the plant pathology Dept. Ann. Rev. Rubb. 
Res. Inst. Sri Lanka. 
LIYANAGE, A. de S. ( 1 9 8 0 ) . Review of the plant pathology Dept. Ann. Rev. Rubb. Res. 
Inst. Sri Lanka. 
LIYANAGE, A. d e S . , PERIES, O . S . and LIYANAGE, N . I. S. ( 1 9 8 3 ) . Towards biocontrol 
of root disease caused by Rigidoporus lignosus 4th Int. Cong. Plant pathology 1 7 — 2 4 . 
August 1 9 8 3 . 
LIYANAGE, A. de S. ( 1 9 8 4 ) . Review of the plant pathology Dept. Ann. Rev. Rubb. Res. 
Inst. Sri Lanka. 
M A R T I N , R. AND D U PLESSIX, C. J . ( 1 9 6 9 ) . White root rot (Leptoporus lignosus) of rubber 
on Lower Ivory Coast. Jl. Rubb. Res. Inst. Sri Lanka 21, p. 9 6 — 1 0 6 . 
NAPPER, R. P. N . ( 1 9 3 2 a). Observations on the root diseases of rubber trees caused by 
Fomes lignosus. Jl. Rubb. Res. Inst. Malaya 4 p. 5 — 3 3 . 
NAPPER, R. P. N . ( 1 9 3 2 b). A scheme of treatment for the control of Fomes lignosus 
in young rubber areas. Jl. Rubb. Res. Inst. Malaya 4 p. 3 4 — 3 8 . 
N A P P E R , R. P. N . ( 1 9 3 8 ) . Root disease and underground pests in new plantings. Planter, 
Kuala Lumpur, 19 p. 4 5 3 — 4 5 5 . 
NAPPER, R. P. N . ( 1 9 3 9 ) . Root disease investigations. Path. Div. Rept. for the year 
1 9 3 8 . Rubb. Res. Inst. Malaya. Rept. 1 9 3 8 . 1 1 6 — 1 2 3 . 
NEWSAM, A. ( 1 9 6 3 ) . Covers and root diseases. Plrs' Bull. Rubb. Res. Inst. Malaya, 
N o . 6 8 , p. 1 7 7 — 1 8 1 . 
NEWSAM, A. ( 1 9 6 4 ) . Root diseases of Hevea. Plrs' Bull. Rubb. Res. Inst. Malaya No . 7 5 , 
p. 2 0 7 — 2 0 9 . 
NEWSAM, A. , JOHN, K . P. AND R A O , SIRIPATHY, B . ( 1 9 6 1 ) . Decay of rubber wood. Proc. 
Nat. Rubb. Res. Conf. Kuala Lumpur 1 9 6 0 , p. 5 0 3 — 5 0 9 . 
5 1 6 
PERIES, O . S . , FERNANDO, T. M. AND SAMARAWEERA, S. K . ( 1 9 6 3 ) . Field evaluation of 
methods for control of white root disease of Hevea. Jl. Rubb. Res. Inst. Sri Lanka, 
39, p. 9 — 1 5 . 
PERIES, O . S., FERNANDO, T. M. A N D SAMARAWEERA, S . K . ( 1 9 6 5 ) . Control of white root 
disease of Hevea brasiliensis. Jl. Rubb. Res. Inst. Sri Lanka. 41 , p. 81 — 89. 
PERIES, O . S . ( 1 9 6 6 ) . The economics of disease control in Hevea. Bull. Rubb. Res. Inst. 
Sri Lanka, 1, p. 2 7 — 3 3 . 
PERIES, O . S . ( 1 9 7 0 ) . Economics of control if white root disease Fomes lignosus of Hevea 
brasiliensis in Ceylon. Root diseases and soil borne pathogens, p. 2 5 2 , Eds. T. A. 
Toussoun, R . B. Bega and P . E. Nelson : Uni. of Calif. Press. 
PERIES, O . S . ( 1 9 7 2 ) . Review of the plant pathology Dept. Ann. Rev. Rubb. Res. Inst. 
Sri Lanka. 
PERIES, O . S. ( 1 9 7 4 ) . An assessment of the significance of the food base in relation to 
infection of Hevea roots by Fomes lignosus. Plant. Dis. Reptr. 5 8 . 2 9 5 — 2 9 7 . 
PETCH , T. ( 1 9 2 1 ) . The diseases and pests of the rubber tree. London : Macmillan and 
Co. , p. 2 7 8 . 
RJGGENBACH, A. ( 1 9 5 8 ) . A note on the chemical control of white root disease of rubber. 
Quart. Circ. Rubb. Res. Inst. Sri Lanka. 34, p. 8 — 1 0 . 
SATCHUTHANANTHAVALE, V . AND HALANGODA, L. ( 1 9 7 1 ) . Sulphur in the control of white 
root disease. Q. Jl. Rubb. Res. Inst. Sri Lanka, 48, p. 8 2 — 9 1 . 
WIJEWANTHA, R. T. ( 1 9 6 4 ) . Influence of environment on incidence of Fomes lignosus 
in rubber replantations in Ceylon. Q. Jl. Rubb. Res. Inst. Ceylon. 4 0 , p. 5 6 — 6 3 . 
Y O U N G , H . E. ( 1 9 5 4 ) . White root disease of Hevea, Leptoporus lignosus (Fomes lignosus) 
Adv. Cir. Rubb. Res. Inst. Ceylon. 46, p. 1 — 8 . 
DISCUSSION 
Q — D. K. WEERASINGHE (TRI) : The clone, LCB 1320 shows resistance to white root disease, why 
don't you compare chemical constituents of RRIC 100 and LCB 1320, to ascertain the fungitoxic 
compounds. Would'nt it be better than Calixin. 
A — A . de S. LIYANAGE (RRISL) : We are in the process of examining the basis of resistance. It would 
be possible to do it only if these clones contain inherent inhibitory substances or develop phy-
toalexin like compounds after infection. However, it would be difficult to say, at this stage, 
whether these would be better than Calixin. 
Q — ANON : What is the effect of low soil pH of 3 — 4 on the growth of rubber plants ? 
A — A . de S. LIYANAGE (RRISL) : We have observed that low soil pH caused by the application of 
sulphur can cause a slight retardation of growth initially compared to control plants, but this 
effects disappears in about one year. 
517 
-ANON : The effect of sulphur persists only for one year. D o you recommend continuous applica­
tion of sulphur in the same area as a prophylactic treatment ? 
- A . de S. LIYANAGE (RRISL) : Continuous application at the end of the first year could be useful 
as a prophylactic measure, especially in areas where the infection was heavy in the old stand. 
In areas where the incidence was mild this too can be eliminated taking into consideration the 
present cost of sulphur and labour. Experiments on root development have shown that in the 
second year the lateral roots grow well beyond the area on which sulphur is applied. 
- ANON : Would you please explain whether Trichoderma or Penicillium is more antagonistic to 
R. lignosus under field conditions ? 
• A . de S. LIYANAGE (RRISL) : Quantitative studies to determine whether Trichoderma is more anta­
gonistic than Penicillia have not been done under field conditions. This is rather difficult to 
assess as Trichoderma population is .more on the surface layers and Penicillia are more in the 
deeper layers. 
• ANON : What are your recommendation for sulphur applications ? 
(i) Which kind of sulphur 1 
(ii) Periods of application and frequency ? 
(iii) Method of application ? 
(iv) Quantity per tree ? 
• A . de S. LIYANAGE (RRISL): 
(i) Elemental sulphur in powder form of 99 % purity 
(ii) At the time of planting and sometimes a second application at the end of the first year of 
planting, where the infection had been heavy in the old stand 
(iii) Sprinkle sulphur on the surface of soil and gently fork it in. Mixing with soil is not necessary. 
(iv) 113 g ( i lb of sulphur) 
DISCUSSION 
SESSION V LEAF AND BARK DISEASES 
C . RODRIGO,( JEDB I) : You mentioned that the younger rubber tree is more susceptible to bzrk 
rot infection. Could you clarify on that as to how is it that the younger rubber tree is more 
susceptible than more mature trees. 
A . de S. LIYANAGE (RRISL) : The virgin bark is highly susceptible to infection compared to the 
renewed bark. That is why immature trees are more susceptible than mature trees. 
GNANARATNAM, (JEDB I) : There is an apprehension or a fear that when cacao is interplanted with 
rubber there is possibility of the Phytophthora fungus affecting rubber, infecting cacao and causing 
black pod disease or the cankei of cacao. Have you come across any positive evidence to justify 
this fear ? Are the strains of Phytophthora different from the one that cause the black pod and 
one that causes bark rot ? 
A . de S. LIYANAGE (RRISL): Under Sri Lankan conditions the predominant species that affect 
rubber is P. meadii and the only species that affects cacao P. palmivora. With detached plants 
parts of cacao, it was possible to cause infection with a heavy dose of inoculum of the rubber 
isolate. P. meadii. However, when cross inoculations were done under field conditions the in­
fection was restricted. 
M. NICOLE, Ivory Coast : Do you know if resistant Hevea clones become susceptible to some strains 
of Microcyclus and if they are resistant do they show a particular reaction ? 
DARMONO : There are a number of strains of M. ulei and clones show differential susceptibility 
depending on the strain. In some regions like in Bahia FX 25 is susceptible but in Experito 
Santo it is free of severe infection. It appears that climatic conditions can also affect the disease 
severity. 
P. de A. GURUSTOGHE : Department of Minor Export Crops, Matale : Has the genetics of the re­
sistant cultivars to South American leaf blight been worked out, at least to some races ? 
DARMONO : The genetic aspects has an effect in the disease susceptibility. For instance FX 710 
was originally resistant to races 1, 2 and 3 but now it is susceptible to races,4, 5 and 6. Genetic 
aspects need more detailed investigations. 
My question is, is it due to major gene resistance or polygenic resistance ? 
May I suggest that we discuss this question later together. 
518 
Q — DAHANAYAKE, Mahawela S .P. Ratnapura : Is there any way of preventing bark rot by way of panel 
dressings ? That is by way of any chemical treatment ? 
A — A . de S. LIYANAGE (RRISL) : Yes, you can, if you protect the tapping panels with fungicides when 
rubber pods are actively producing inoculum during the South West monsoon season. Several 
effective fungicides are available. Antimucin (0-83 %), Brunolinum plantarium (15 %), Difolatan 
(0-8 — 1 6 % ) , Fylomac 90' (1 %) and Ridomil (0-2 — 0-4%). Others like Aliette is being 
tried. Most water soluble fungicides run the risk of getting washed off, during the wet months 
when the disease is most active. The use of systemic or water repellant fungicide formulations 
can give better results. 
Q — RANJIT BOPEARACHCHT, Chemical Industries, (Colombo) Ltd.,: Dr. Liyanage has already answered 
part of my question. I would require further clarification from Dr. Liyanage. I know from 
what he said already that Phytophthora problems occur mostly during rainy months. Sometime 
back RRISL did not recommend tapping during wet weather. I would like to know his views 
regarding chemical control of Phytophthora versus tapping during wet weather ? 
A — A. de S. LIYANAGE (RRISL) : It is always better to prevent than cure. However, during the wet 
months the estimated crop cannot be harvested. Tapping is done on any intervening dry period. 
We have seen from our studies that an injury to the tapping panel can predispose it to infection, 
provided there is inoculum and therefore the tapping cut remains susceptible up to 72 h after 
injury. This means that there is a risk of panels getting infected even if one avoids tapping 
during wet days. This is why we have been looking for fungicides which are systemic or more 
persistent. We reported on some work that we have carried out on these lines in 1976. 
Q — M. NICOLE, Ivory Coast: You suggested that there was a restricted reaction in tolerant trees. 
Can you be precise on the nature of some of these reactions. 
A — A. de S. LIYANAGE (RRISL) : When the artificial inoculation technique was used to determine 
clonal susceptibility, two distinct types of reactions were observed. The spreading lesions were 
always brownish in colony but in the lesions formed in tolerant clones were blackish suggesting 
that some biochemical changes have occurred in these clones. 
Q 1 — DE ALWIS, Mahaoya Estate : I would like to know the preference of the skirt-type rain guards as 
against the gutter-type and which adhesive you used for the polythene rainguard because when 
we used this about 20 years ago we found that water filtering through after about 4 weeks, and 
also much infection of bark rot under the skirt. 
A — A. de S. LIYANAGE (RRISL) : From the point of view of bark rot control the skirt type of rainguard 
is preferred as it will prevent the seepage of inoculum into the cut provided it is properly sealed 
to the tree. The adhesive we used is Shellkote. We had no seepage upto 20 weeks. 
Q 2 — There was another problem that of damage by crickets, quite regularly and so that the skirt had 
to be replaced ever so often. 
I think this is one of the reasons for abandoning the skirt. In those days crickets were attracted 
to the skirt. But today it is reckoned as a minor problem. 
TRAN VAN CANH, Ivory Coast : We know that Microcyclus has many spore forms ; conidia pycno-
spores and ascospores. What type of spore did you use for testing the clones ? 
DORMONO : We used leaf disks and inoculation was done with a conidial suspension, which has 
been standardized. The lesion size was measured after incubating the leaf disks for a known 
period at low temperature. 
— You showed inoculation of leaves of rubber with conidia. 
We did inoculations of clones including, South American clones and some FX clones. There were 
differences in lesion size between these clones. 
TRAN VAN CANH, Ivory Coast : What is the make of the fogging machine, the fungicides used, 
rates of application, the quantity of the product for one hectare and frequency of application. 
Is the efficacy of the fungicide affected at very high temperature ? 
The make of the machine is Leco Model 120 D . We used three fungicides viz. Benlate, Chlorothalo­
nil, Trifornil. At first we used 100 g of active ingredient per hectare with Benlate, but later we 
used Trifornil ? Volume sprayed was 7 1 for ha. We took uninfected rubber leaves of clones 
FX 4163 and FX 3864 and hung them on nylon thread in the field. Then we fogged the area, 
leaves were collected and brought back to the laboratory for inoculation after the smoke dis­
appeared completely. 
A — 
Q -
A — 
Q 2 -
A — 
Q -
519 
We have done some experiments on the speed of fogging and the formulation of the fung-'cide. 
We monitored the temperature and found that there is a difference between what is shown in 
the pyrometer and at the point of emission. 
Q 2 — If you have an hour of a day for treatment and if the direction of the wind changes one way, what 
would you do ? 
A — We really use this experiment in flat and square land with roads on the 4 sides to take up any variation 
of the wind. 
Q — TRAN VAN CANH, Ivory Coast : What is the method of control of black stripe and abnormal leaf 
f o i l ..caA :« o-: T * 
A — A. de S. LIYANAGE ( R R I S L ) : We no longer control abnormal leaf fall caused by Phvtophthora 
spp. as the incidence is mild. This is largely due to the lack of routine control of Oidium leaf 
disease. Consequently the pod formation is reduced resulting in lowering the foci of infection 
This integrated approach has been unique and has been in practice for the last 20 years or so 
ith regard to bark rot control the fungicides we recommend are Antimucin, Brunolinum 
tarium, Difolatan, Fylomac and Ridomil. Experiments are in progress to determine the 
icy of Aliette and Sandarfan. 
# 
520 
SESSION 9. S O I L M A N A G E M E N T 
MANAGEMENT OF GROUND COVERS FOR OPTIMUM 
PRODUCTION 
By 
N . Y O G A R A T N A M , A . M . A . P E R E R A A N D G . J. D E M E L 
(Rubber Research Institute of Sri Lanka, Agalawatta) 
A B S T R A C T 
The increase in efficiency of the appropriate ground covers by proper management 
and the effect of such practices on the NPK nutrition of Hevea are discussed. 
Investigations in use of mixed creeping and bush legumes indicate that several species 
can be grown successfully and that the adverse effect of Stylosanthus guyanensis was seen 
only in areas with annual rainfall in the region of 3500 mm, possibly due to the absence of 
competition for moisture. The use of legumes to supply the considerable amounts of nitrogen 
needed and thus enhance the vigour and performance of Hevea have been recognised and the 
beneficial effects of legumes persisted for 4 years during the early mature period. Although 
compensatory N applied to trees in naturals improved their growth ; but from an economic 
point of view, legumes give the highest discounted returns. 
Phosphate application to covers led to better tree girths than where the phosphate was 
applied to the trees, irrespective of the type of cover grown and this beneficial effect persisted 
for 8 years after the treatments were discontinued. On the other hand, with K, application 
direct to the rubber tree appears to be as efficient as its application to the ground covers 
in particular on K deficient soils. 
Guide lines offered for practical applications are : 
* Should endeavour to establish a leguminous cover at replanting. 
* The most appropriate species of legume should be chosen. 
* Compensatory nitrogen should be applied where legume establishment and main­
tenance become difficult. 
* From an economic and agronomic point of view, P and K fertilizers can be applied 
to the ground covers during the immature period instead of their application direct 
to the rubber tree. 
INTRODUCTION 
The influence of ground covers on early growth of the rubber trees is believed to be 
mainly the result of moisture and nutritional advantage. However, the effects have been 
variable and contradictory. This may be attributed largely to variation in soil and climatic 
factors. This paper discusses the findings on the use of appropriate covers and their 
management to increase their efficiency under Sri Lankan conditions. 
521 
MATERIALS AND METHODS 
Details of five experiments on types and nutrition of covers are discussed in relation 
to growth and yield of Hevea. 
Experiment 1 
Effects of six leguminous cover species viz. Pueraria phaseoloides, Calapagonium 
muconoides, Desmodium ovalifolium, Mimosa invisa, Centrosema pubscens and Stylosanthus 
gayanesis (gracillis) on growth of clone PB 86 were studied on two sites in 1977 replantings. 
The control plots in these experiments were in naturals. Site 1 was located in the Kalutara 
District receiving on annual rainfall of 3125 to 3750 mm and site 2 in the Kurunegala 
District with an annual rainfall of 2250 and 2500 mm. Soil in sites 1 and 2 had been of 
the Boralu (laterization) and Parambe (biotite gneiss) type respectively (Silva, 1969). 
Experimental treatments were assigned in both sites according to randomized block de­
signs and replicated five times in each sites. 
Experiment 2 
This experiment was started in 1972, on a sandy loam of the Boralu series using clone 
PB 28/59. The treatments that were studied in a randomized block design replicated 
four times are : (1) Sown mixed legumes (mainly Pueraria phaseoloides and Desmodium 
ovalifolium, L) (2) Selective elimination of non-legume cover constituents from the natural 
cover (L : S) (3) Non-selective control of natural cover ( N : NS) (4) Non-selective control 
of natural cover with supplementary nitrogen equivalent to thrice the normal recommen­
dation (N : NSn x ) (5) Non-selective control of natural cover with manuring of rubber 
controlled by leaf analysis to the levels with sown mixed cover legume (N : NSn 2 ) . 
Experiment 3 
This experiment was started in 1970 on a sandy clay loam of the Agalawatta series 
using clone RRIM 623. The design was a split plot, with treatments being replicated 
twice. Legumes (mainly Mimosa invisa) and naturals were in the main plots. Four 
levels of nitrogen (level two corresponds to the normally recommended rate and level 
four twice the normal recommendation) and four methods of application of phosphate 
viz, no phosphorus, phosphorus to cover, phosphorus to rubber and phosphorus to cover 
and rubber, were in the sub-plots (Table 1). 
Experiment 4 
The influence of potassium application on growth of Clone PB 68 in a 1977 replanting 
was studied in this experiment on Boralu series soil. Three methods of application of 
potassium at two levels of K were tested in a randomized block design with four repli­
cates. The experimental treatments were ; Nil potassium, potassium to rubber only, 
potassium of cover only at two levels of K and potassium rubber and cover at two levels 
of K (Table 1). In all experiments rubber trees received uniform applications of non-
experimental fertilizers according to the Rubber Research Institute of Sri Lanka (RRISL) 
recommendation (Anon, 1980). 
322 
Table 1. Experimental fertilizer schedule 
Experiment 3 
Rock, phosphate (28 % P a 0 6 ) 
(oz./tree) 
Period Pr Pc 
Y e a r l 6 2 
Year 2 12 2 
Year 3 18 2 
Year 4 18 1 
Year 5 18 1 
Year 6 —. 
Experiment 4 
Muriate of Potash (60% K 8 0 ) 
(oz./tree) 
Kr KC-1 K C - 2 
2 2 4 
4 2 4 
6 2 4 
6 1 2 
6 1 2 
6 _ _ 
RESULTS AND DISCUSSION 
Types of covers 
The effect of various leguminous covers on growth of immature Hevea that were 
assessed in Experiment 1 at two sites by measuring final girth prior to commencement 
of tapping (Table 2) indicates that in general, growth of trees in legumes was superior to 
the trees in naturals except Stylosanthus guyanensis which continued to show a depressive 
effect on growth of immature PB 86 in site 2 located in the Kurunegala District, where the 
annual rainfall is much lower than at site 1. 
Table 2. Effect of ground covers on final girth of immature Hevea (PB 86)* 
Covers Boralu (laterization) Parambe (biotite gneiss) 
3125 — 3750mm 2250 — 2500mm 
Naturals (control) 48 • 12 48-63 
Pueraria 52 83** 54-40** 
Calopogonium 50-58* 52 88** 
Desmodium 50 70* 52-25** 
Mimosa 49-22 53-63** 
Stylosanthus 50-20* 50*04 
The differences in performance of rubber under Stylosanthus appear to be due to 
the differences in soil moisture levels between the two sites. Table 3 shews the rates of 
transpiration as measured by a parometer on leaf samples from rubber and cover plants 
growing in site 1. Although the period January to April had been fairly dry in 1983 
and there had been a steady decline in the rate of transpiration over this period, yet there 
appears to have been sufficient moisture in the soil. This suggests that in the wet districts 
e.g. Kalutara, competition for moisture between Stylosanthus and immature rubber may 
not exist to the extent of retarding growth of immature rubber in comparison with condi­
tions in districts with low rainfall i.e. Kurunegala. Another aspect that emerges from 
this data is that it would be possible to apply fertilizers during the dry periods of January 
and February if there had been approximately 1000mm of rain in November/December 
of the previous year. 
» In this and all other tables *, *• and *** indicate those treatments differing from the control at 
P => 0 05,0 01 and 0-001 respectively. 
523 
Table 3. Rate of transpiration (microgram cm-1 sec-2) 
Sample January February March April 
Rubber 6-588 6 030 3-544 3-464 
Pueraria phaseoloides . 15 720 11-632 13-452 7 191 
Desmodium ovalifolium 5-62 1*38 1-34 1 06. 
In experiment 2 , results at the end of 10 years from planting (Table 4) showed that 
the trees under legumes were superior to the trees in naturals with regard to both girth 
and early yield. Other treatments with extra nitrogen to naturals have also shown effects 
similar to the legume treatment suggesting that in the absence of legumes extra nitrogen 
will have to be applied, in some cases as high as three times the normal recommendations. 
Table 4. Effects of ground covers on growth and yield of immature Hevea (PB 86) at the 
end of 10 years 
Covers Mean girth Yield 
(cm) (kg/ha) 
Naturals (control) 5 6 1 1800 
Legumes 61 5* 2506* 
Elimination of non-legumes 58 9* 2440* 
Naturals with extra N 60-5* 2578* 
Naturals with extra N based on leaf analysis 58-8* 2402* 
Some beneficial effects of legumes are presented in Table 5. Contributions by cover 
leaves, green matter, litter and litter C/N ratio, were all in favour of legumes. Other 
benefits of legumes include total dry matter production and the total N content in leaves 
per tree. To illustrate this point, few trees were chosen in the experimental area at ran­
dom, they were completely defoliated and total dry weight and N content were recorded 
(Table 6). More dry matter production obviously indicates higher physiological activity 
in the tree and higher total N content in leaves would result in higher nitrogen reserves 
in the tree. In deciduous trees like rubber it is known that prior to wintering i.e. defolia­
tion, nutrients in the leaves move back into the tree (Yogaratnam et al, 1976) and are 
retained in the trees as reserves. This " nutrient bank " within the tree provides the 
new shoots with nutrients at the time of refoliation. 
Table 5. N. concentrations (%) and CjN ratio, 42 months after planting 
Cover leaves Green matter Litter Litter C/N 
Covers 
Weeds 1 60 1 03 1 65 22 26 
Legumes 2 22*** 1 46*** 2 26*** 12 09*** 
L; Selective 1 98* 1 14 2 15*** 13-38*** 
N - n ! 1-72 1 01 1-68 18 55 
N ; n a 1 6 4 1 07 1-57 19 18 
524 
Table 6. Total dry weight and N content of leaves per tree (g) 
Covers Dry wt. N content 
of leaves of leaves 
Weeds (control) 455 14 78 
Legumes 1856*** 61-76*** 
L ; Selective 896* 29 09* 
N ; D l 794* 25 16* 
N ; n a 435 13 40 
Nutrition 
Phosphate. Due to the restricted movement of phosphate in the soil, phosphate availability 
from applied fertilizers is rather low. Different methods of application of phosphate 
indicated that P application direct to ground cover is an efficient method of increasing 
phosphate uptake by Hevea trees (Table 7). P to covers showed higher leaf P values 
than P direct to rubber at the end of 6 years from planting. This increase in leaf P 
concentration obviously increased girthing and percentage tappability of trees at the end 
of 6 years (Table 8). 
Table 7. Effect of phosphate application on P concentration in leaves (%) 
Phosphate End of 6 years 
N i l P 0 254 
P to rubber 0-277** 
P to cover 0 289*** 
P to rubber and cover 0-291*** 
Table 8. Effect of P application on final girth and percentage tappability (6 yrs) 
Treatments Final girth (cm) Tappability (%) 
N i l P 45 51 28 
P to rubber 46-71 39* 
P to cover 49-76*** 55*** 
P to rubber + cover 50-14*** 
525 
The effect of applying phosphate to covers also showed long residual effects on pro­
duction. Mean girth and yield data at the end of 12 years from planting (Table 9) 
suggests that the beneficial effect of phosphate applied to covers persisted for a period 
of approximately 8 year6 from the time P application was discontinued. Some of 
the possible reasons for this effect had been discussed by Yogaratnam et al (1976). 
Table 9. Effect of Phosphate application on its residual effect at the end of 12 years 
Treatments Mean girth 
(cm) 
Mean yield 
(kg/ha) 
Nil P 64-28 1418 
P to rubber 65-98* 1565* 
P to cover 68-33** 1753** 
P to rubber + cover 68-24** 1660** 
Potassium 
With regard to potassium, mobility of K ions in the soil in not a problem, but K 
application was reported to have increased nitrogen fixation in some legumes. 
An experiment similar to the above but with the application of K to legumes indicated 
(Table 10) that unlike phosphate, the application of K direct to the rubber tree appeared 
to be as efficient as its application to the ground covers in particular on K deficient soils. 
Increasing the level of K beyond the current recommendation may not be beneficial. 
Table 10. The effect of potassium application on growth and tappability of PB 86 
Mean girth Final girth Percentage 
tappability 
Potassium (5 yrs.) (year 6) (year 7) 
cm cm (%) 
K to rubber : Nil 39 64 45 01 31 
Level 43 06** 48-85** 52** 
K to cover : Nil 37 04 42 31 29 
Level 1 43 23** 48 95*** 54** 
Level 2 43-78** 49-54*** 51** 
526 
P R A C T I C A L A P P L I C A T I O N S 
@ Establish leguminous covers at planting. 
© In high rainfall areas any legume species is suitable. 
© In low rainfall areas Stylosanthus should be avoided. 
@ Some effects of legumes persist for about 4 years during the early mature phase. 
© In areas under non-legume covers, application of extra nitrogen is important. 
© Application of P to cover is move beneficial than application direct to rubber. 
@ Residual effects of P applied to immature plantations would persist for about 8 
years. 
© In K deficient soils, potassium can be applied either to rubber or to covers. 
A C K N O W L E D G E M E N T S 
The authors wish to thank the experimental staff of the Soils and Plant Nutrition 
Department for technical assistance. 
R E F E R E N C E S 
A N O N , ( 1 9 8 0 ) . Advisory circular N o . 8 5 , Rubber Research Institute of Sri Lanka. 
S I L V A , C. G . ( 1 9 6 9 ) . Provisional classification of rubber soils of Ceylon with their rela­
tionship to Malayan soils. / . Rubb. Res. Inst. Malaya, 24, 1 4 5 . 
Y O G A R A T N A M , N . , S U L A I M A N , H . , K A R U N A R A T N A , A. D . M. A N D P E R I E S , K . S. A. C. ( 1 9 7 6 ) . 
Jl. Rubb. Res. Inst. Sri Lanka, 54, 2 9 1 — 2 9 8 . 
527 
SOME ASPECTS OF MOISTURE AND SOIL CONSERVATION 
IN RUBBER PLANTATIONS 
By 
L A L A N I S A M A R A P P U L I A N D N. Y O G A R A T N A M 
(Rubber Research Institute of Sri Lanka, Agalawatta) 
S U M M A R Y 
Results ofsome investigations on soil and plant water status and the effects of leguminous 
ground covers, mulch and naturals (weeds) in minimising losses of moisture and soil in rubber 
replantings during the early stages of growth of plants are discussed. Soil water in the 
root zone was estimated by the neutron moisture meter whilst plant physiological and micro-
meteorological measurements were maintained on the above ground portion of the system. 
In preliminary investigations, the location of access tubes for the use of neutron probe, 
a calibration curve for the neutron moisture probe and sampling techniques for measuring 
of leaf water potential in Hevea using a Scholander pressure bomb have been finalised. A 
close relationship appeared to exist between available soil water and rainfall pattern for a 
90 cm profile. 
Among the different soil management practices, dead mulch exhibited the highest soil 
moisture storage capacity in comparison with other practices such as growing leguminous 
covers or naturals or leaving the land bare (clean weeded). Similar results were observed 
with regard to other parameters such as leaf water potential, relative water content of leaf 
and leaf water deficit. Girthing of plants also showed similar effect with plants in mulch 
growing more vigorously than the plants under other practices. Investigations on the in­
fluence of ground covers including mulch on run-off and soil losses indicated that mulching 
with Guatamala grass loppings can reduce run-off and soil erosion markedly as compared 
to clean weeded and bare land. Plots In legumes and naturals also showed effective control 
of run-off and soil erosion losses, the beneficial effect of legumes was seen from the second 
year. 
I N T R O D U C T I O N 
Rubber plantations in Sri Lanka are generally situated on sloping land with varying 
degrees of steepness and the intensity of rainfall recorded in these areas had also been 
variable. The extent of moisture and soil losses by competition, surface runoff and ero­
sion varies widely depending on the type of agro-management practices adopted (Ellison, 
1952; Jackson, 1964; Mannering and Meyer, 1963 and Yogaratnam et al, 1977). 
Numerous factors are involved covering a wide range of biological cultural and cli-
matological parameters that must be known or measured in order to understand the 
effects of moisture and soil losses on the performance of rubber plants in the field. Among 
those are soil moisture (Laycock, 1960; Haridas, 1980) leaf water potential (Scholander, 
1965) relative water content (Weatherley, 1950) specific leaf weight (Pearcc, Carlson, Barnes, 
Hanson and Hart, 1969) soil erosion (Manipura, 1972) surface runoff (Hudson, 1957; 
Adams 1966) and loss of nutrients (Lai et al, 1980). 
529 
This paper discusses the results of some investigations on soil moisture, erosion and 
surface runoff and the effects of leguminous covers, mulch and naturals (weeds) in mini­
mising losses of moisture and soil in rubber replantings during the early stages of growth 
pf plants, 
M A T E R I A L S A N D M E T H O D S 
In soil-plant-water relation studies the neutron meter is used extensively to monitor 
soil water in field studies concerning the response of the plant to water content of the soil, 
Preliminary Investigations 
Calibration of neutron meter: Establishment of the basic calibration curve for 
the particular neutron meter, type of installation and for the general class of soil 
in the experimental areas is an important aspect. Field calibrations were therefore 
made by obtaining moisture meter readings in duplicate at depths of 10 cm intervals from 
1 0 — 160 cm. For gravimetric determination of moisture content and for bulk density 
determination, soil samples were collected from each depth at which moisture meter read­
ings were recorded by digging pits around each access tube. 
Some of the objectives of a measurement programme based on the neutron meter is 
the estimation of some complex variables of the profile water balance such as run-off 
and change in the soil water store. This profile water store is the difference between 
the upper and lower limits of water content at some specified depth, usually the root zone. 
Location of access tubes: In order to minimise sampling errors due to location 
and mode of installation of access tubes in the field, a pilot study was conducted 
to determine the most suitable position in relation to the rubber plant for the installation 
of access tubes. Trees were selected at random in a 3 years old replanting and access 
tubes were installed at distances 60,120 and 180 cm ( 2 , 4 and 6 ft) from the trunk of each 
plant in four directions, north, south, east and west, making in all 12 positions. 
Moisture readings were recorded weekly at depths of 10 cm intervals from 10 to 90 cm, 
which generally covers the rooting zone of immature rubber plants. 
Sampling for leaf water potential (LWP) measurements : Determination of a suitable 
leaf sampling technique for the measurement of LWP is important in order to 
minimise variations due to factors such as age (maturity) and position of leaves 
(effect of light) and the time of the day (differential rates of transpiration). Leaves 
of 3 stages of maturity, tender, intermediate and mature were collected from two 
positions; exposed to sunlight and shaded, at three times of the day ; 0900 to 1000, 
1300 to 1400 and 1700to 1800 hours. LWP measurements were made with a Scholander 
Pressure Bomb by inserting the leaf with petiole, after making an incision around the 
petiole and allowing one minute for the latex to ooze out from the leaf. 
Experiment 1 
This experiment was started in a 1981 replanting and was designed to study the effects 
of different soil management practices on moisture stress and their influence on growth 
530 
of immature PB 86 plants. Four ground cover management practices that were studl 
in a randomized block design with three replicates are ; 
(a) Bare (clean weeded) 
(b) Mixed legumes (Pueraria phaseoloides and Desmodium ovalifolium) 
(c) Natural cover 
(d) Dead mulch 
The soil classified as from the Boralu series is shallow, gravelly loam, brown to 
reddish yellow in colour and overlying cabook. It is characterised by the presence of 
iron concretions which occur throughout the soil mass (Silva 1964). Soil analysis indicated 
the following mean values: pH 4 7 CEC 5 9 ; organic C% 1 -7 total N 0-14% and total 
P 5 52 ppm in the planting strips. The water holding capacity of the soil was 40—60% 
and bulk density was 1 29. 
Access tubes were installed in experimental plots and weekly counts for soil Water 
content were made using a neutron meter at depths of 10 cm interval from 10 to 90 cm-
Soil samples were collected at 6 monthly intervals for determining some physical and 
chemical characteristics. (Mechanical analysis by pipette method and bulk density by 
using core samplers, Organic C by Walkly and Black's method exchangeable K and Mg 
and CEC by ammonium acetate—pH 7). A Scholander Pressure Bomb was used to assess 
leaf water potential fortnightly. Relative water content of leaf, leaf water deficit (Weather-
ley, 1950) and specific leaf weight (Pearce, Carlson, Barnes, Hart and Hanson, 1969) 
assessments were also made at 3 monthly intervals. Leaf samples were collected at 
6 monthly intervals for chemical analysis and girth measurements were recorded at 3 
monthly intervals. 
Experiment 2 
In this experiment the effect of leguminous covers* naturals (weeds) and mulch dti 
runoff and soil erosion losses during the early stages of growth of rubber in replantings 
were studied using soil erosion and runoff plots. Plots were constructed at slopes of 35 
to 40%. Each run off plot (Fig. 1) contained four planting points (PB plants) surrounded 
by plot bounderies made of brick walls. At the bottom end of the slope, a conveyance 
channel helped the water to flow through a " H-flume " device into a collection tank 
via a multislot divisor. The collection tank at the end measured the total volume of water 
and soil collected. 
The experimental treatments consisted of (a) Leguminous covers (mixed growth of 
Pueraria phaseoloids and Desmodium ovalifolium (b) Naturals (weeds) (c) Mulch (Guata-
mala loppings) and (d) Bare (clean weeded). 
RESULTS 
Preliminary Investigations 
Calibration curves 
Calibration curves obtained from 29 readings taken over a period of 12 months (Fig. 2) 
for the surface (0 — 10) is , moisture content (© = 0 41 Count Ratio (CR) — 0 18 and 
(Fig. 3) for the sub surface (10 — 160 cm) is © — 0 - » 0 - 5 2 CR — 0 39. 
531 
1st 
PLOT BOUNDARY 
H-FLUME MULTISLOT DIVISOR 
COLLECTION AREA CONVEYANCE CHANNEL ;,. 
COLLECTION TANK" 
Fig. 1 . Soil erosion and run-off plot. 

Q.40 
0.35 
0.30 
S 
I & 0.25 
to 
O 
g: 0.20 
—I 
0.15 
0.10 
1.35 1,00 1.05 1.10 1.15 1.20 1.25 1.30 
COUNT RATIO 
Fig. 3. FieU calibration of neutron moisture meter for the sob-surface layer, 10—160 cm. 
1.40 I.«j5 
Location of access tubes 
Statistical calculations (co-efficient of variation) of count ratios obtained from 12 
positions of access tubes in relation to planting points over a period of 12 months from 
September 1982 indicate that during wet period the count ratio may not be influenced by 
the location of the access tubes. But on dry days in general, distances 1 - 2 m to 1 - 8 m 
(4 to 6 ft.) from the tree would appear to give the most reliable information with increas­
ing accuracy at lower depths. N o significant difference was seen in relation to the direc­
tion of the location of access tubes (Table 1). 
Table 1. Co-efficient of variation (%) different at depths and distances of access tubes 
Soil moisture status 
Fig. 4 illustrates the available water storage capacity (AWSC) of the experimental 
area over a period of 12 months, which has an upper limit of 235-5 mm (field capacity) 
and a lower limit of 131 3 mm thus yielding an AWSC of 104 2 mm. Fig. 5 illustrates the 
available soil water in relation to the total rainfall for a 90 cm profile, which is generally 
the rooting depth of immature rubber, over a period of 12 months. A clear relationship 
appears to exist between available soil water and total rainfall during this period. 
Leaf water potential 
A study on the leaf sampling technique for the assessment of LWP indicated that 
sampling between 1300h t o l 4 0 0 h w a s the most suitable time for LWP measurements. 
It also appears that sampling of exposed leaves in the intermediate and mature stage is 
likely to give the most consistent results (Table 2). 
Table 2. Co-efficient of variation (%) of factors influencing leaf water potential 
Distance (cm) 
Depth (cm) 
10 
20 
30 
40 
90 
120 180 
30-6 28-6 
66-4 56-3 
49-2 45-3 
40-7 33-1 
12-1 15-3 
Time (h) 
Position of leaves 0900 to 1000 1300 to 1400 1700 to 1800 
Shaded, mature 
Shaded, intermediate 
Shaded,tender 
7 5 
7-5 
10 5 
10 5 
9-8 
15 5 
9 8 
5 2 
7-6 
19 0 
19 8 
17-2 
13 8 
14 0 
12 5 
Exposed, mature | 
| Exposed, intermediate | 
Exposed, tender 10-6 
535 


Experiment 1 
Among the different soil management practices that were tested for their effects on 
moisture conservation in immature rubber plantations, plots in dead mulch exhibited 
the highest soil moisture storage capacity of 27 64 cm in comparison with other practices 
such as growing leguminous covers or naturals or leaving the land bare (clean weeded) 
(Table 3). Similar results were observed with regard to other parameters such as LWP 
(Table 4) relative water content of leaf (Table 5) and leaf water deficit (Table 6). Girthing 
of plants (Table 7) also showed similar effects with plants in mulch growing more vigorously 
with a girth of 19 49 cm at the end of 24 months as compared to a value of 17 25 cm in 
the plots in naturals. The cation exchange capacity of the soil was also increased by 
mulching (Table 8). 
Table 3. Effect of different soil management practices on soil moisture storage capacity 
Treatment Soil moisture storage capacity (cm) 
Bare (clean weeded) 18 3 
Naturals 24 '3 
Legumes 21-7 
| Dead mulch 27 6 
Table 4. Effect of different soil management practices on leaf water potential ofrubber (L WP) 
Treatment LWP (bars) 
Bare (clean weeded) 26 • 7 
Naturals 26-5 
Legumes 27-4 
| Dead mulch 22 • 1 | 
Table 5. Effect of different soil management practices on relative water content of leaves 
(RWC) 
Treatment RWC 
Bare (clean weeded) 0-814 
Naturals 0 • 849 
Legumes 0-820 
Dead mulch 0 • 867 
Table 6. Effect of different soil management practices on leaf water deficit in rubber (LWD) 
Treatment L W D ( % ) 
Bare (clean weeded) 18-6 
Naturals 15 1 
Legumes 18 1 
Dead mulch 13-3 
538 
Table 7. Effect of different soil management practices on the girth {cm) of rubber trees 
Treatment March June Dec. June 
1983 1983 1983 1984 
Bare (clean weeded) 10-7 12-2 16 4 20-1 
Naturals 10-2 1 1 9 15 3 18-6 
Legumes 10 7 12 5 15-9 18-9 
Dead mulch 1 1 1 1 2 1 17-1 21-5 
Table 8. Effect of different soil management practices on CEC and organic carbon of the soil 
Treatment CEC (meq/lOOg) Organic carbon (%) 
Bare (clean weeded) 3 3 0 9 
Naturals 3 . 9 1 . 4 
| Legumes 5-3 | 1-3 
Dead Mulch 5-5 | 1-4 
Experiment 2 
Data on the influence of ground covers including mulch on soil erosion losses (Fig. 6) 
indicated that the plots in Guatamala grass loppings had the least soil erosion loss as 
compared to other treatments. Plots in bare land recorded a total soil loss of 60 91 
tons/ha over the 3 year period in comparison with a loss of only 3 55 tons/ha in the plots 
in mulch. In general, the plots in legumes and naturals also showed effective control 
of soil erosion loss. The beneficial effect of legumes was however seen from the second 
year only. Similar results (Table 9) were obtained with regard to run off losses as well as 
with the maximum runoff values in the bare plots ie plots that were clean weeded. In 
terms of nutrients all these amount to a loss of approximately 0-027 and 0 003 tonnes 
of N and K respectively (Table 10). 
Table 9. Effect of different soil management practices on total volume of water (litres) 
collected during the 3 years period 
Treatment Year 
1 2 3 
Bare (clean weeded) 3701 2805 481 
Naturals 1973 1259 124 
Legumes 4492 2163 66 
Dead mulch 2468 2089 277 
Table 10. Effect of different soil management practices on loss of nutrients (tonnes/ha) 
from soil (mean of 3 years) 
Treatment N K 
{Bare (clean weeded) 0 027 0 003 ~\ 
Naturals 0 001 0 0001 
Legumes 0 009 0 001 
Dead mulch 0 002 0 0002 
539 

D I S C U S S I O N 
Iastudieson soit-plant-water relations theNeutron Mctcrisuscd extensively to moni­
tor soil water balance in the root zone whilst plant physiological and micrometeorological 
measurements are maintained on the above ground parts of the system. This study 
clearly establishes the effectiveness of these measurements in immature Hevea plantations. 
It had been possible to establish fairly reliable calibration curves for the neutron probe 
for the surface, 0 — 10 cm, and subsurface soils 1 0 — 1 6 0 cm, possibly due to the absence 
of soil horizons with contrasting physical properties within the experimental area. This 
calibration would in any case be sufficient for comparison of agronomic treatment effects 
within the experimental site. 
It appears that during dry periods, distances varying from 1-2 to 1 8 m from the 
trunk of the tree are suitable for assessing the water content of the soil profile. This is 
in line with the previous reports based on the use of radiotracers that root activity in 
immature plantatinos is more towards the tree trunk than away from it (Liu Chong-Qun, 
1984). In our study, it is possible that on wet days when the soil was saturated with 
water, daily changes due to transpiration losses may not have caused much fluctuation 
in the soil water content. But on dry days, the process of depletion and recharging from 
the moist layers may be continuous in the profile close to the trunk. An incidental but 
important observation that emerges from this is that fertilizer applications to immature 
rubber should be confined to areas approximately 60 cm from the base of the tree for 
efficient uptake. 
The available soil water storage capacity of 104- 2 mm with the upper and lower limits 
of235-5and 131 3 mm, respectively, is likely to be the amount of water that was available 
to the plants, both rubber and covers over a period of 12 months in 1982. Clearly these 
upper and lower limits are somewhat arbitrary, yet in practical terms it has been possible 
to obtain a useful estimate of the profile water store and the available water storage capacity 
for Boralu type soils receiving a total rainfall of 2-80 mm. This is further elaborated 
by the data on available soil water in relation to precipitation where a clear relationship 
exists between these two variables. It is evident that even during the most dry periods 
in 1983 ie in January, February and March the available soil water has been in the region 
of 140 to 175 mm, with the lowest value of 140 mm in February. Total precipitation in 
the preceding 3 months ie in October, November and December 1982 had been around 
1000 mm. It therefore seems possible to apply fertilizers, at least to mature areas even 
during the dry periods of January to April if there had been approximately 1000 ram of 
rain during the north east monsoon in the previous year. 
However, it is known that during periods of water stress, growth of immature plants 
may be retarded to the extent of delaying the immaturity by even over 12 months (Siva­
nadyan, Hadrias and Pushparajah, 1973, Pushparajah and Haridas, 1976). It appears 
possible to eliminate or at least minimise the adverse effects of moisture stress by mulching. 
Soil moisture storage capacity in plots that were mulched was much higher than in the 
other plots. Moreover, some plant physiological measurements made, for example leaf 
water potential, relative water content and leaf water deficit, all indicate that the water 
status of plants were improved in plots that were in mulch in comparison with the plants 
in naturals or bare or in legumes. Whether it is legumes or naturals they are equally 
541 
deterimental in effecting evapotranspiration losses. However mulching and legume treat­
ments improved some important soil characteristics for example the cation exchange 
capacity. 
Mulching had been found to be very effective in not only avoiding evapotranspiration 
losses, but also in preventing run-off and soil erosion losses. Marked reduction in run-off 
losses with mulching indicates that the soils have inherently high infiltration capacity. 
Moreover, the surface soils in the mulch, legumes and natural plots were in a more friable 
condition than the bare plots, which possibly helped in greater absorption of water. 
Soil losses that occurred due to erosion from bare lands which were in the region of 
61 9 tons/ha could be easily reduced to 3-6 tons/ha by mulching. This would also help 
in reducing nutrient losses considerably. It is also pertinent to note that soil loss during 
the first year from plots in naturals is less than that from clean weeded and legume plots, 
thus emphasizing the need for selective weeding during the early stages of the planting. 
Any cover treatment other than mulching, takes 6 to 12 months before providing suffi­
cient protection to the soil and therefore substantial soil loss occurs during the first year. 
It may, therefore, be necessary to mulch the soil immediately after planting and then 
establish a cover crop or practise selective weeding which would help reduce run-off 
and soil erosion losses considerably. The importance of legumes in the nitrogen nutrition 
of Hevea has been amply demonstrated by other workers (Mainstone, 1969, Pushpa­
rajah and Chellapah, 1969; Yogaratnam et al, 1977). 
ACKNOWLEDGEMENTS 
The authors thank the Director, Dr O. S. Peries for his interest in this work and 
Messrs W. C. Dayaratna, P. Karunadasa and Indranee Denawaka for technical assistance. 
We are also indebted to Mr S. Dimantha of the Land Use Division of the Irrigation 
Dept. for his help. 
R E F E R E N C E S 
A D A M S , J. E. (1966). Influence of mulches on runoff, erosion and soil moisture depletion. 
Soil Sci. Soc. Amer. Proc. 30, 110—114. 
E L L I S O N , W. D. (1952). Raindrop energy and soil erosion. / / . of Experimental Agricul­
ture 20,81 — 9 7 . 
H A R I D A S , G. (1980). Soil moisture use and growth of young Hevea brasiliensis as deter­
mined from lysimeter studies. / . Rubb. Res. Inst. Malaysia 2 8 , 40 — 60. 
H A R I D A $ , G. (1984). The influence of irrigation on latex flow properties and yield of 
different Hevea cultivars. Proceeding of the ICOSANP (M.S.S.S.), Kuala Lumpur, 
13 — 1 5 August 1984. 
H U D S O N , N . W. (1957). The design of field experiments on soil erosion. / . Agric. Engng. 
Res, 2, 56 — 65. 
542 
JACKSON, D . C . (1964). Techniques for recording soil and water losses from soil erosion 
experiments. Research Bulletin, Nos. 8 and 9 , Dept. of conservation and Extension 
S. Rhodesia. 
L A L , R., D E V L E E S . C H A U W E R , D . A N D M A L A F A N G A N J E , R. (1980). Changes in properties 
of a newly cleared tropical alfisol as affected by mulching. Soil. Sci. Amcr. J. 4 4 , 
827 — 832. 
L A Y C O C K , D . H . (1960). Soil moisture studies. Report of the Tea Research Station, 
Nyasland for the years 1958 — 1959. 
Liu C H O N G — Q U N (1984). Radiotracer studies on root activity and foliar spray of rubber 
trees. (Hevea brasiliensis). Proceeding of the ICOSANP ( M . S . S . S . ) , Kuala Lumpur, 
13 — 1 5 August 1984. 
M A I N S T O N E , B. J. (1961). Effects of ground cover and continuity of nitrogenous fertilizer 
treatment on growth to tappable maturity of Hevea brasiliensis Proc. Nat. Rubb. 
Res. Conf. Kuala Lumpur, 1960, 362 — 375. 
M A N I P U R A , W . B . (1972). Influence of mulch and cover crops on surface runoff and soil 
erosion on tea land during the early growth o f replanted tea. Tea Q. 4 3 , 95 — 102. 
M A N N E R I N G , J. V . A N D M A Y E R , D. L. (1963). The effects of various rates of surface mulch 
on infiltration and erosion. Soil Sci. Soc. Amer. Proc. 2 7 , 84 — 86. 
P E A R C E , R. B., C A R L S O N , G. E., B A R N E S , P . K . , H A R T , R. H . A N D H A N S O N , C . H . (1969). 
Specific leaf weight and photosynthesis in alfalfa. Crop. Sci. 9, 423 — 426. 
P U S H P A R A J A H , E. A N D C H E L L A P A H , K . (1969). Manuring of rubber in relation to covers. 
/. Rubb. Res. Inst. Malaya, 21, 2, 126. 
P U S H P A R A J A H , E. A N D H A R I D A S , G., (1976). Developments in reduction in immaturity 
period of Hevea in Peninsular Malaysia. Rubb. Res. Inst. Sri Lanka centenary Conf 
1976. 
S C H O L A N D E R , P. F. , H A M M E L , H . T., H E M M I N G S F . N , E. A . A N D B R O A D S T R H E T , E. D. (1965). 
Sap pressure in vascular plants. Science N. Y. 1 4 8 , 339 — 46. 
S I L V A , C . G. (1964). Provisional classification of rubber soils of Ceylon and their rela­
tionship to Malayan soils. J. Rubb. Res. Inst. Malaya, 2 4 , 145. 
S I V A N A D Y A N , K . , H A R I D A S , G. A N D P U S H P A R A J A H , E. (1975). Reduced immaturity of 
Hevea brasiliensis. Proc. Int. Rubb. Conf. 1975 Kuala Lumpur, 3, 147. 
W E A T H E R L E Y , P. E. (1950). Studies on the water relations of the cotton plant. 1. The 
field measurement of water deficits in leaves New Phytol. 4 9 , 81 — 97. 
Y O G A R A T N A M , N . , S U L A I M A N , H . , K A R U N A R A T N A , A . D . M. A N D P I E R I S , K . S. A . C . (1977). 
Management of covers under Hevea in Sri Lanka. / . Rubb. Res. Inst. Sri Lanka, 
54, 291 — 298. 
543 
APPROACHES TOWARDS LAND EVALUATION SYSTEMS 
FOR HEVEA BRASILIENSIS CULTIVATION IN PENINSULAR 
MALAYSIA 
By 
H . Y . C H A N , F . K . Y E W A N D E . P U S H P A R A J A H 
(Rubber Research Institute of Malaysia, Kuala Lumpur, Malaysia) . 
A B S T R A C T 
In the early stages of development in the country, allocation of land for various uses 
had been on a laissez faire system. No problems on land use occurred then because land 
was plentiful. However, with the tempo of agricultural activity increasing rapidly in the 
past three decades, scarcity of good land and conflicting interests arising from competition 
for prime agricultural land between perennial crops have led to the need to evaluate land 
according to its intrinsic qualities and merits. This has resulted in the development of various 
approaches on land evaluation systems. 
For rubber, the first system proposed was the Soil Suitability Classification System in 
1972; the classification of which was based on the number and types of soil limitations. 
The system was non-parametric in approach. 
Later systems developed were of a parametric nature; the first being the Soil Suitability 
Technical Grouping System for Hevea, proposed in 1975 and was based on an approach 
whereby scores were accorded to the evaluation of the diagnostic criteria used. 
In 1982, a system to evaluate suitability for rubber cultivation has been developed based 
on the multiplicative approach. It differs from the other systems by considering the evalua­
tion of climate in addition to the evaluation of the land units. 
All three systems developed have been shown to have agronomic logic; higher yields 
being obtained on the better soils. The paper attempts to account the characteristic features 
of these approaches and recommends, on the basis of current evidence, a land suitability 
system for rubber cultivation. To facilitate use of this system, modern soil classification 
systems e.g. the USD A Soil Taxonomy have been used to describe the local soil units so 
that this technology can be transferred at a global level. 
P R E A M B L E A N D I N T R O D U C T I O N 
In the early years of active Hevea planting in the East, planters were in the habit of 
selecting, wherever possible, the banks of rivers or areas liable to flooding for the cultiva­
tion of the rubber trees (W. Wright, 1912). This was due to erroneous advice originally 
obtained from Brazil and traced to the custom of native collectors to tap areas nearest 
the rivers. 
545 
The true forests of Hevea were on the highlands and those scattered along the river 
sides did not reflect the actual conditions under which rubber could be grown. In fact, 
in the natural habitat, Hevea brasiliensis was found mainly on drained sites and occasionally 
in sites subjected to brief or slight inundations. One species, Hevea spruceana, occurred 
on the muddy soils of the islands and river banks which were subjected to periodic heavy 
inundation while another species, Hevea comporum, occurred in the dry savannah land. 
Hevea brasiliensis was introduced into Malaya in 1877, but it was not until the turn 
of the century that rubber was planted on a large scale (Wycherly, 1963). Choice of 
land for rubber cultivation then was determined essentially by road accessibility and ease 
of working conditions. Initially, therefore, the choice of land was restricted to the zone 
bordered by the steep hills of the Main Range in the interior and by mangrove, peat or 
padi areas on the coastal areas. Because of difficult terrain, plantings in the early part 
of this century for rubber hardly exceeded the 225 m (750 ft.) contour (Wycherly, 1963). 
The land use pattern for rubber has, therefore, been laid down on the easier terrain 
following somewhat, the rail and road network systems of the country. Following this, 
and with the encouraging steady rubber prices, rubber cultivation mushroomed over the 
country and was grown on a wider range of soils. 
The advent of soil survey in Peninsular Malaysia began in the early 1900's. The 
experiences and contributions from information gathered by soil surveyors culminated 
in a first schematic classification system of Peninsular Malaysian soils as suggested by 
Owen (1951), whereby it became evident to land users that soil was a heterogenous unit 
and land use needs to consider soil properties and quality. 
Experience with fertiliser use in rubber culture in Peninsular Malaysia which dates 
back to 1903 as reviewed by Pushparajah (1977), has shown that response to fertilisers 
can vary according to soils ; higher responses being obtained on some soils when 
compared to others. This further reflected on the variable fertility of the different soils. 
Recognising, therefore, the heterogeneous nature of soils as reflected by the differing 
agro-management responses experienced on them, the need to evaluate soil and land units 
according to their intrinsic qualities and merits gathered momentum in the past two de­
cades. Additionally, the rapidly increasing tempo of agricultural activity in the country, 
the scarcity of good land and the conflicting interests arising from competition for prime 
agricultural land between perennial crops emphasised further the importance of this con­
ceptual development on land use. This has resulted in the development of various ap­
proaches on land evaluations systems and the purpose of this paper is to account the 
characteristic features of these approaches. The merits and demerits of the each approach 
is discussed and suitable land evaluation systems which can be used for rubber cultivation 
are described. 
Soil suitability classification systems for rubber cultivation 
Early attempts 
A soil suitability classification system for rubber cultivation has been proposed by 
Chan and Pushparajah in 1972. It is made up of a framework of soil and physio­
graphic features which are considered to be desirable for optimal growth of rubber. 
, 546 
On the other hand, the poorer soils have properties which limit the optimal growth and 
performance of rubber and these limitations arc graded as minor, serious or very serious 
in their degree of limiting crop performance. 
The criteria considered in the classification system arc as follows : 
Desirable physical properties : 
Soil depth up to 100 cm, free of pan/rock outcrop hindrance 
Well drained 
Good soil aeration 
Good soil structure (strong, moderately strong and moderate medium and fine 
subangular blocky structure) 
Friable to firm consistency 
Good water-holding capacity 
No peat or acid peal deeper than 20 cm 
Soil texture with sufficient clay (preferably a minimum amount of 35% clay 
to retain sufficient moisture and nutrients and about 30% sand to allow for 
expression of good physical soil properties like aeration and drainage). 
Desirable environmental and physiographic features : 
Gently sloping or rolling terrain between 2 - 9 ° slopes (4 - 20 %), with minimal 
soil erosion and surface run-off 
Water-table should be deeper than 100 cm. 
Desirable chemical properties : 
At least medium levels of total nutrient contents of nitrogen (N), phosphorus (P) 
potassium (K), and magnesium (Mg) with no deficiency of trace elements 
A pH of around 4 5 
Absence of saline/acid sulphate conditions. 
Limitations are properties which limit good crop performance and can be graded 
as minor, serious and very serious in accordance to its degree of severity on affecting 
crop performance. 
547 
Minor limitations.* 
Weak soil structure within 90 cm 
Moderate drainage conditions 
Massive thick hard-pan below 50 cm from the surface or loosely packed gravels 
within 50 cm from the surface 
Less than 50% rock outcrop in a unit area 
Susceptibility of soil erosion 
Sub-optimal soil nutrient status, reflected by low contents of nitrogen, phosphorus, 
potassium and magnesium. 
Serious limitations: 
Slopes steeper than 9° (20%) but less than 16° (36%) 
Massive thick hard-pan between 20 and 50 cm of the surface 
Between 50 % and 75 % of rock outcrop in a unit area 
Permanent water-table between 20 and 50 cm of the surface 
Strong compaction influencing permeability and infiltration 
Poor structure (too sandy resulting in no structure or massive due to hcav; 
clay in waterlogged conditions) 
Susceptible to moisture stress as reflected by tcxtural and structural qualities 
Poor nutrient status. 
Very serious limitations: 
Slopes steeper than 16° (36 %) 
Massive thick hard-pan al or very close (within 20 cm) to the surface 
More than 75% of rock outcrop in a unit area 
Permanent water-table at or near (within 20 cm) the surface 
Acid peat layer thicker than 20 cm at or near the surface 
Land disturbed by mining activities (tin tailings) and very sandy deposits, wi 
more than 90 % sand 
Very poor nutrient status. 
548 
The consideration of these criteria allows the classification of the soils into the follow­
ing soil suitability classes. 
Class I—The soils in this class have no limitations to rubber cultivation 
Class / /—The soils in this class have one or more minor limitations to rubber cultivation 
Class III—The soils in this class have at least one serious limitation to rubber cultivation 
Class IV—The soils in this class have more than one serious limitation to rubber cultivation 
Class V—The soils in this class have at least one very serious limitation to rubber cultivation 
Later attempts based on parametric approaches 
The first indication of a parametric approach towards the evaluation of Peninsular 
Malaysian soils for rubber cultivation was attempted by Chan et al in 1975. This 
system consists of assessment based on the evaluation of sixteen criteria, namely rock 
outcrop, effective soil depth, soil texture, wet consistency of the soil, moist consistency 
of the soil, soil structure, internal soil drainage, presence of peat, acid sulphate status, 
available water, permeability, erodibility, soil pH, terrain (slope %), susceptibility to 
flooding and stagnation of water at the surface (Table 1). 
Class limits are given for each criterion for the desirable range : minor, serious and 
very serious limitations. A score of 5 points is given to the desirable range ; 4 points to 
the range of minor limitation ; 2 points to the range of serious limitation ; and 1 point 
to the range of very serious limitation. 
The total scores are obtained by summation and the maximum score is 80 points 
(100%). In this Soil Suitability Technical Grouping System for Rubber, the following 
technical groups are proposed : 
la 9 6 — 100%, lb 9 1 — 9 5 % 
IIa 86 — 90% l ib 81 — 85% 
11176 — 80% 
IVa 71 — 75 % IVb 66 — 70 % 
Va 61 — 65 %, Vb 56 — 60 %, Vc 51 — 55 %, and Vd less than 50 %. 
The technical groups I to V, are defined in the same manner as the soil suitability 
classes of Chan and Pushparajah (1972). 
Parametric approaches for international application 
While earlier systems of soil suitability assessments were confined to evaluation within 
the country, more recent approaches were aimed for international use. As such, more 
recent approaches used systems that enabled international correlation. For this, the 
guidelines of the FAO Framework for Land-evaluation (FAO, 1976) were U6ed. 
549 
Degree in se verity 
_ of limitation 
Soil properties 
1. Physical properties rock 
outcrop (%) 
Table 1. Principal criteria in grading of soil suitability classes for rubber (after Chan et al, 1975) 
Desirable range'* Minor limitation4* Serious limitation" Very serious limitation 1* 
Effective depth (a) 
Texture 
Consistency dry 
Consistency moist 
Consistency wet 
(a) Stickiness 
(b) Plasticity 
Absence 
> 100 cm 
< 50 
> 60 — 100 cm 
> 50 — 75 
> 25 — 60 
> 7 5 
> 2 5 cm 
Almost proportionate amounts (i) Sandy loam (> 50—70% sand) (i) Very sandy 
of sand and silt + clay 
Soft 
Friable — Very Friable 
Slightly sticky 
Slightly plastic 
(ii) Clayey (> 50—70% clay) 
(iii Sifty clayey 
(> 50—70% silt + clay) 
Slightly hard 
Firm 
(i) Extremely sandy (> 90 % sand) 
( > 7 0 — 90 % sand) (ii) Extremely clayey (> 90 % cley) 
(ii) Very clayey (> 70—90% clay) (iii) Extremely silty clay 
(iii) Very silty clayey (> 90% silt + clay) 
(>70 — 9 0 % silt + clay) 
Very hard 
Sticky 
Non plastic 
Non-sticky 
Plastic 
Structure Prismatic Columner Weak moderate fine very fine ( l ) Moderate strong medium 
(2) Weak coarse 
Angular blocky 
Internal drainage (b) 
Peaty characteristic 
Acid sulphate characteristic 
Moisture retention (based on 
available water in cm/m) 
Penneability(b) 
Erodibility0>) 
II. Chemical properties pH 
III. Physiographic features 
Terrain ( c) 
Susceptibility to flooding 
(1) Moderate strong fine very 
fine 
(2) Moderate strong medium 
Class D well drainage 
Absence of peat 
Absence of acid 
> 150 cm 
Moderate 
Class l (slightly eroded) 
Mean 4, 5 range 4-3—4-6 
0 — 8° 
No flooding 
Weak coarse (for texture 
coarser than clay loam) 
Class C Moderate well drained 
Hard Loose 
Very firm, Loose 
Very sticky Very plastic 
Strong Coarse 
Strong Coarse 
Extremely firm 
Extremely sticky extremely plastic 
Strong very coarse 
Strong very coarse 
Weak coarse (finer than clay loam) 
Class B Imperfectly drained Class A Poorly drained 
Class E Somewhat excessively Class F Excessively drained 
drained 
Acid peat > 50 cm from surface: Acid peat 25 cm to 50 cm from Acid peat layer <25 cm from 
> 25 cm thick peat 
Acid S O 4 layer > 50 cm from 
surface : > 25 cm thick 
> 100— 150 cm 
Moderately slow or moderately 
rapid 
Class 2 (Moderately eroded) 
> 4 - 6 — 5 0 
> 8 — 15° 
surface : 25 > 50 cm thick 
peat 
surface : < 50—100 cm thick 
peat 
Acid S O 4 layer 25 cm - 50 cm Acid S O 4 layer > 25 cm from 
from surface ; > 25—50 cm surface; >25—100 cm thick 
thick 
> 50—100 cm 
Slow or rapid 
Class 3 (Severely eroded) 
> 5 0 — 6 0 
> 15—33° 
> 5 0 cm 
Very slow or very rapid 
Class 4 (Very severely eroded) 
> 6 0 
>33° 
Floods after very heavy down­
pours 
Stagnation of water at surface No stagnation 
Flodded after heavy downpours Floods after light downpours 
Water stagnates for a few hours Water stagnates for 3 days Water stagnates for > 3 days 
* denotes point scores given 
(a) considers depth to hard pan (limestone, laterite layers, quartz vein, compact parent material, etc.) 
or to permanent water table 
(b) based on the definitions of U.S.D.A. 1960 
(e) refers to more commonly occurring slopes only. 
5 5 1 
Recognising that climate forms the first major part of any land evaluation exercise, 
a method to assess the suitability of climate has been proposed (Yew and Sys, 1984) and 
shown in Table 2. The evaluation of climate should be used in combination with the 
evaluation of the soils and landscape features (Table 3). 
Table 2. Climatic requirements for Hevea cultivation* 
Degree of limitation 
Climatic characteristics 0 1 2 3 4 
Mean temperature (°C) 28 — 25 24 
29 
— 22 
— 30 
21 
31 — 
20 
32 
19 — 18 
33 — 34 
< 18 
> 3 4 
Mean daily maximum temp. 34 — 29 28 — 27 
> 3 4 
26 — 24 23 — 24 < 2 2 
Mean daily minimum 
temp. (°C) > 2 0 20 — 19 18 — 17 16 — 15 < 15 
Mean annual rainfall (mm) >2000 2000 — 1750 1749 — 1500 1499 — 1250 < 1250 
Expected total rain 
interference (days/yr) 0 — 30 31 — 60 61 — 90 > 9 0 
Sunshine (hours/yr) 2100 2100 — 1800 1799 — 1400 1399 — 1000 > 1000 
Mean annual R.H.(%) < 8 0 80 — 100 
Length of dry season 
(months/yr) 0 - 1 2 — 3 4 5 — 6 6 > 
•After Yew and Sys (1984) 
The evaluation of both climate as well as soils and landscape follows a parametric-
limitation approach whereby scores, ranging from 0—100, are accorded to the diagnostic 
criteria ; and land indices are obtained by a multiplication process, according to the 
method of Sys (1978). 
Land suitability is categorised into orders, classes, subclasses and units. Two orders 
arc recognised, being suitable (S) for rubber cultivation and not suitable (N). The sub­
divisions are as follows : 
Order S .'suitable. Land units with no, slight or moderate limitations and no more 
than two severe limitations which, do not exclude the use of the land. The land index 
is more than 25. 
Class SI : highly suitable. Land units with no or only slight limitations which, in 
combination, give land index values ranging from 75—100. 
Class S2: moderately suitable. Land units with slight or moderate limitations which, 
in combination, give land index values ranging from 50 to 74. 
Class S3: marginally suitable. Land units with moderate limitations or normally 
not more than two severe limitations which, in combination, give land index values rang­
ing from 25 to 49. 
Order N: not suitable. Land units with more than two severe limitations or with at 
least one very severe limitation which exclude the use of the land. The land index is 
normally 24 or less. 
553 
Tabic 3. Landscape and soil requirements for rubber * 
Land 
characteristics 
TOPOGRAPHY (t) 
Degree of limitations 
Slope % 
WETNESS (w) 
Drainage 
Fiat to 
undulating 
0—10 
Good 
Rolling 
11—20 
Hilly 
20—30 
Steeply 
dissected 
31—45 
Mountainous 
> 4 5 
Moderate Imperfect, Imperfect Poor, very 
temporarily with almost poor 
high water permanent 
table 
Flooding FO 
high water 
table 
Fl F2 + 
PHYSICAL SOIL CHARACTERISTICS (s) 
Fine earth 
texture/ 
structure 
CL, Co, SiC, SCL, SL, LSf 
SC, Cs L 
LSm, LSc, Sin, Sc 
Cm, SiCm, 
Sf 
Surface and Downgrading over six sections of 25 cm using the 
subsurface stoniness weighting rates : 2—1.50—1.0—0.75—0.50—0.25 
Soil depth (cm) 150 149—100 99—50 49—25 
FERTILITY NOT READILY TO BE CORRECTED 
Oxisols ( + ) 
< 2 5 
Weathering stage 
(CEC) 
Base saturation 
Top soil (%) 
Organic carbon 
( % , C , 0 - 1 5 c m ) 
* After Sys, 1982 
Entisols 
lnceptisols 
Ultisols 
Oxisols 
( - ) 
any (—) 
20—35 
> l - 5 
- 1 6 ( + ) 
36—50 
< 2 0 
1-5—0 6 
51—80 
< 0 - 6 
> 8 0 
For texture/structure, the suffixes : 0 = weak structure and consistence of the oxic 
horizon, s = angular or subangular blocky structure, m = massive, f = fine, m = medium, 
c — coarse. 
554 
Class N : Currently not suitable. Land units with severe or very severe limitations 
which may be overcome in time but which cannot be corrected with existing knowledge 
at currently acceptable cost. 
Class N2 : permanently not suitable. Land units having limitations which appear 
so severe as to preclude any possible use of the land in the given manner. 
The kinds of limitations are reflected in the land suitability subclasses. 
Allowances for the consideration of current land suitability classification (without 
major land improvements) or the potential land suitability classification (after major 
land improvements have been made) are in-built within the framework. 
Test applications 
Soil suitability classification system for rubber 
The classification of the suitability of the soils for rubber cultivation by this system 
has been found to be in good agreement with field experience. In the study by Chan 
and Pushparajah (1972) the mean yield for any of the given clones is always higher 
on the shale derived Munchong series soil (Haplorthoxes) and lower on the marine 
alluvial soil of the Selangor series (Sulfaquepts). Munchong series soil has been classified 
as a Class I soil and Selangor series soil as a Class V soil for rubber cultivation in that 
study. 
Soil Suitability Technical Grouping System for Rubber 
Based on a study of this Technical Grouping System by Chan et al (1975), it has 
been shown that soils with balanced composition of particle size distribution, friable 
consistencies, strong and moderately strong medium and fine subangular blocky structures 
and good drainage features, such as Munchong, Segamat and Kuantan series (all of these 
soils being Oxisols) scored the highest (above 72 points). On the other hand, soil units 
with a shallow effective depth, unbalanced composition of particle size distribution, very 
firm or very loose consistencies, high compaction, poor structure and poor drainage scored 
very low (42 — 52 points), for example, Briah series, Selangor series and Linau series 
(Inceptisols/Entisols), while peat (Histosols) scored the lowest (36 points). Higher techni­
cal group ratings were generally related to higher yield patterns in the field. Tables 4 and 5 
summarise the system. 
Land evaluation; system for rubber 
Land indices, obtained in the system proposed in 1982, have been shown to have 
yield predictive values. Higher land indices are related to the yield of rubber (Fig. 1). 
As such, the expected yield of rubber for various land units can be computed ; those for 
four land suitability classes are shown in Fig. 2. 
555 
556 
557 
Table 4. Scores of common soils under Hevea in peninsular Malaysia* 
Total scores from sum of 
Classification 16 differentiating criteria Technical 
Soil series after soil taxonomy (maximum, total — 80 points) group rating 2 
Scores Relative % 
Munchong Haplorthox 78 98 la 
Jeram Tropudult 
Bungor Paleudult 76 95 
Jerangau Haplorthox 74 93 
Rengam Paleudult 73 91 lb 
Segamat Haplorthox 73 91 
Kuantan Haplorthox 73 91 
Harimau Paleudult 72 90 
Senai Haplorthox 70 88 11a 
Batang Merbau Tropudult 68 85 
Subang Tropaquept 68 85 l i b 
Kulai Tropudult 66 83 
Serdang Paleudult 64 80 
Ulu Tiram Dystropept 64 80 
Pohoi Dystropept, Tropudult 63 79 
Holyrood Quartzipsamment, 
Haplorthox 63 79 111 
Tampoi Haplorthox 63 79 
Lunas Dystropept 62 78 
Kuala Brang Dystropept 61 76 
Durian Tropudult 59 74 
Batu Anam Tropudult 59 74 
Malacca Haplorthox 57 71 IVa 
Pasir Puteh Tropaquept 55 69 
Kampong 
Chempaka Dystropept 55 69 
Tok Yong Dystropept 55 69 
Sogomana Tropaquept 55 69 IVb 
Sitiawan Tropaquept 55 69 
Briah Fluvaquept 52 65 
Selangor Sulfaquept 50 63 Va 
Sg. Buloh Quartzipsamment 48 60 Vb 
Linau Sulfaquept 42 53 Vc 
Peat Fibrists, Hemists 36 45 Vd 
* After Chan et al (1975) 
1 Maximum = 80 points 
2 Technical group rating (by relative %) 
96 — 100 1a 81 — 85 l i b 
91 — 951b 76 — 80 111 
86 — 90 Ha 
558 
Table 5. Proposed soil suitability technical grouping system for soils under rubber* 
Soil 
suitability 
class 
Yield 
categories1 
kg/ha 
High yield 
(> 1350) 
Series '• 
(a) Munchong 
Jeram 
Prang 
(b) Segamat 
Kuantan 
Rengam 
Jerangau 
Yong Peng 
Bungor 
Classification 
Great Group Level 
Soil taxonomy After Thorp and Smith, 1949 
I I 
I I I 
IV 
Above average (a) Senai 
yielding Hariniau 
(> 1200—1350) Merbau 
Batang Subang 
Kulai 
Average 
yielding 
(> 1050—1200) 
Serdang 
Holyrood 
Typic Paleudults 
Typic Paleudults 
Tropeptic Haplorthoxes 
Tropeptic Haplorthoxes 
Tropeptic Haplorthoxes 
Typic Paleudults 
Typic Paleudults 
Typic Paleudults 
Typic Paleudults 
(Orthoxic)6 Palehumulls 
Typic Paleudults 
(Orthoxic) Tropudults 
(Oxic) Humitropepts 
(Oxic) Plinthodults 
Typic Paleudults 
Oxic Dystropept 
Latosol 
Latosol 
Latosol 
Latosol 
Latosol 
Red and yellow podzolic 
Red and yellow podzolic 
Latosol 
Red and yellow podzolic 
Red and yellow podzolic 
Red and yellow podzolic 
Red and yellow podzolic 
Hall-bog 
Yellow podzolic 
Red and yellow podzolic 
Pale yellow podzolic 
Below 
average 
yielding 
(> 900—1050) 
Low yielding 
(>900) 
(a) Balu Anam (Aquoxic)5 Dyslropcpts 
Durian 
(b) Malacca 
Gajah 
Mati 3 
Marang 
Kedah 
Seremban3 
(a) Briah 
Selangor 
(b) Sungei 
Buloh 
(c) Linau 
(d) Peat 
(Oxic)5 Piinthudults 
Plinthic Haplorthoxes 
Petroplinthic 
Haplorthoxes 
Petroplinthic 
Paleudults 
Lithic Oxic Dyslropcpts 
Petroplinthic Paleudults 
Tropic Fluvaquents 
Typic Sulfaquepts 
Orthoxic Quartzipsammcnls 
Typic Sulfaquepts 
Tropofibrist 
Grey podzolic 
Yellow podzolic 
Latosol 
Latosol 
Grey podzolic 
Reddish brown podzolic 
Reddish brown podzolic 
Low humic glei 
Low humic glei 
Alluvial 
Low humic glei-bog 
Bog 
1. Considering only average management standards : with higher levels of management standards and 
modern clones, productivity is significantly higher (Chan and Pushparajah, 1972) 
2. After Chan, H. Y. 1975. 
3. Included on the basis of limited data and their close likeness of physical and chemical properties to 
the ones studied. 
4. It is qualified that there are soil variation like soil texture existing within a soil series. The physio­
graphy of the soil series can also vary, e.g. slope and soil depth. These variations influence yield 
and growth (Chan et. at., 1972 & 1974). As such, the soil scries mentioned in this table are the model 
or standard soil series only. Further investigations of such variations on Hevea performance will 
provide the basis for updating and revision of these recommendations to incorporate any necessary 
refinements. For current practice and purpose, it is sufficient if a plantation has its property soil-
mapped at the series level. This is an essential pre-requisite for any initial proper management-
planning. 
Parenthesis under ' classification' column denotes suggested new sub-groupings. 
No soil series name has been given to " Peat" for purposes of this paper, the general grouping of 
" Peat" is used. 
5. 
6. 
• After Chan et. al.; 1975. 
S U M M A R Y A N D DISCUSSIONS 
The need to develop methods that can be used to diagnose the suitability of land for 
agricultural purposes, including rubber, is now assuming greater proportions of promi­
nence. This situation has arisen in Peninsular Malaysia, whereby, with the tempo of 
5 5 9 
agricultural activity increasing rapidly in the past three decades, scarcity of good land and 
conflicting interests arising from competition for land by various users have led to the 
need to evaluate land according to its intrinsic qualities and merits. 
In Peninsular Malaysia, the first schematic soil classification system was suggested 
by Owen in 1951. Only in 1966, was the first official documentation of soil series present 
in the country made available. This knowledge, coupled with the information accreted 
over a period of time regarding the performance of rubber on various soil situations and 
their responses to fertilisers provided the avenue to understand better the relationships 
of rubber tree crop/soil adaptability and suitability. 
The first Soil Suitability Classification System for rubber has been proposed of 
Chan and Pushparajah in 1972. The classification is based on the number and types by 
soil limitations and is non-parametric in approach. 
Three years later, the Soil Suitability Technical Grouping System for Hevea (Chan 
et al., 1975) was proposed. The approach was based on an addition of scores accorded 
to the evaluation of the sixteen diagnostic criteria. 
Since both the systems were developed for local use, the effect of climate has been 
considered to be small and not significant when compared to the contribution of the soil 
effect on rubber performance, as shown by Chan and Pushparajah (1972). 
More updated information (Yew, 1982), although cursory in nature, indicated that 
yield differences attributable to climate could be significant and has to be considered in 
land evaluation. 
With an aim towards global use, an evaluation system of climate for rubber has been 
proposed (Yew and Sys, 1984), following the FAO (1976) guidelines for land evaluation. 
A system for evaluating soils and landscape (Sys, 1982). following the same FAO 
guidelines, has also been developed. 
A consideration of both the climate as well as soils and landscape leads to an effec­
tive land evaluation as shown by the significant relationship between land indices and 
yield of rubber (Yew, 1982). 
An added advantage of the Land Evaluation System is that land suitability can be 
considered in its present state, without any land improvements (actual land suitability), 
or in its future state, after land improvements have been made (potential land suitability). 
It is, thus, possible to upgrade a land unit to a higher potential land suitability class 
by appropriate agro-management manipulations and land improvements. However, the 
final arbiter is the economics of the operation itself. 
Finally, all three systems described, can be used to group land units into suitability 
classes, from which expected yields of rubber production can be predicted. All three 
systems have been ' test-applied' and found to have agronomic logic. 
560 
Further considerations 
While the foregoing has attempted to establish the logistics and soundness of the 
basic structure of the three crystallised systems on land evaluation for Hevea brasiliensis 
cultivation, it is likely that further precision can be derived from finer quantitative defini­
tions in the degree of range of the considered criteria selected for evaluation. Furthermore, 
other criteria, based on the local experience of features common to a particular environ­
ment, may add more strength to the overall completeness of the systems. Thus, for exam­
ple, the experience gained on the influence of strong winds on rubber trees on Hainan 
Island as reported by Huang and Zheng (1983) can form the foundation towards intro­
ducing an additional criterion into the proposed systems (Table 6) following the principles 
of Table 1. 
Table 6. Consideration of wind scale as an additional criterion for land evaluation for 
Hevea brasiliensis cultivation* 
Degree in severity 
of limitation 
Desirable 
Minor 
Moderate 
Serious 
Very serious 
Wind scale 
(Beaufort force value) 
8 and below ( < 20 -7 m/s) 
9 (20 8 — 24 4 m/s) 
10 (24-5 — 28 4 m/s) 
11 — 12 (28 5 — 36 9 m/s) 
13 and above ( > 37 0 m/s) 
Wind damage % (as 
experienced in Hainan 
Island) 
2 — 5 
6 — 1 0 
11 — 16 
17 — 33 
> 33 
* Based on Huang and Zheng (1983) 
Point scores can be awarded to the criterion in accordance with the same principles 
as established in the systems discussed in the foregoing. 
In the same vein, other criteria, environmental or otherwise, can be included for 
assessment criterion in question. However, such additions should be further confirmed 
by formal tests and trials. 
Following the same line of reasoning, a cross-check of experience gained on a specific 
criterion but operating in different environments will attest further to the quantitative 
validity of the chosen criterion. Thus, for example, it is interesting to note that Huang 
and Zheng's experience that, in China, the temperature ranges of 27 — 30°C (optimum 
range for photosynthesis) and 22 — 28°C (favourable for latex flow) agrees very much 
with the desirable range for temperature as established by Yew and Sys (1984) based on 
experience in hot Tropical Malaysia. Perhaps, such follow-up work should form the 
basis of future work in land evaluation studies for rubber. 
A C K N O W L E D G E M E N T S 
The authors wish to thank the Director and Deputy Director (Research) of the Insti­
tute for their encouragement in these studies and their kind permission to publish this 
paper. Further, the co-operation received from all organisations who supplied the ground 
yield data and the assistance of the supporting staff within the Division are deeply appre­
ciated. 
561 
R E F E R E N C E S 
C H A N , H. Y . A N D P U S H P A R A J A H , E. (1972). Productivity potentials of Hevea on West 
Malaysian soils. Proc. Rubb. Res. Inst. Malaysia Plrs. Conf., Kuala Lumpur, 97. 
C H A N , H . Y . , P U S H P A R A J A H , E., Y E W , F. K . A N D Z A I N O L E U S O F (1975). A soil suitability 
technical grouping system for Hevea. 3rd ASEANSoil Conf, Kuala Lumpur, 277—287. 
F A O (1976). A framework for land evaluation. Soils Bull. No. 32, FAO, 72 pp. 
H U A N G , Z . D. A N D Z H E N G , X. Q. (1983). Rubber cultivation in China. Proc. Rubb. 
Res. Inst. Malaysia Plrs. Conf, Kuala Lumpur, 31. 
O W * E N , G. (1951). A provisional classification of Malayan soils. Jour. Soil Sc., 2, 
20 — 42. 
P U S H P A R A J A H , E. (1977). Nutritional status and fertiliser requirements of Malaysian 
soils for Hevea brasiliensis. D.Sc. Thesis Submitted to State Univ. Ghent, Belgium. 
S Y S , C. (1978). Evaluation of land limitations in the humid Tropics. Pedologie XXVIII, 
3, 307 — 335. 
S Y S , C. (1982). In : Y E W , F. K . (1982). Contribution towards the development of a 
land evaluation system for Hevea brasiliensis Muell. Arg. cultivation in Peninsular 
Malaysia. D.Sc. Thesis Submitted to State Univ. Ghent, Belgium, pp. 327. 
W Y C H E R L E Y , P . R. (1963). Variation in the performance of Hevea in Malaya. / . Trop. 
Geog.,17, 143—171 . 
Y E W , F. K . (1982). Contribution towards the development of a land evaluation 
system for Hevea brasiliensis Muell. Arg. cultivation in Peninsular Malaysia. 
D.Sc. Thesis Submitted to State Univ. Ghent, Belgium, pp. 327. 
Y E W , F. K . A N D S Y S , C. (1984). Land evaluation for Hevea brasiliensis cultivation using 
land characteristics : 1. The evaluation of climate. Draft to be submitted to 
RRIM Journal, pp. 16. 
$62 
NODULATION OF PUERARIA PHASEOLOIDES BY 
INTRODUCED RHIZOBIA IN COMPETITION WITH 
NATURALIZED STRAINS IN THREE DIFFERENT SOILS 
C . K . J A Y A S I N G H E , * C . A . P A R K E R , * * D . K . K I D B Y * * A N D S . A . K U L A S O O R I Y A * * * 
(Rubber Reserch Institute, Sri Lanka * University of Western Australia, Australia** 
Univesity of Peradeniya, Sri Lanka.***) 
INTRODUCTION 
Inoculation of legume seeds with selected strains of rhizobia before sowing is a general 
practice in many parts of the world. For such inoculations to be successful the introduced 
strain must not only be effective but also should be able to compete with indigenous rhizobia 
and other soil microflora. 
In most of the temperate countries native rhizobial populations are very low in num­
bers and inoculation with effective rhizobia is an obligatory procedure in legume pro­
ductivity. But this is not the case in tropical countries, specially with the Rhieobium 
cowpea complex. There is a considerable number of native flora which can compete 
with introduced inocula. However published information on success of strains applied 
as inocula in soils containing native flora appear to be lacking for tropical soils. 
The investigations reported in this paper were undertaken to find competition bet­
ween introduced strains and naturally occuring populations of Rhizobium cowpea complex 
under temperate and tropical conditions. 
The competitiveness is generally studied by examining strains, recovered from no­
dules on inoculated plants. Therefore, the most important factor in this study is the identi­
fication of introduced strains in the presence of other strains. 
Different methods have been used by various workers over the last 40 years for this 
purpose. Serology was the most commonly used method and other techniques include 
morphology in culture or in nature, symbiotic properties, biochemical properties, sensi­
tivity to bacteriocins, antibiotics and bacteriophages, and use of genetic markers such as 
antibiotic resistance. 
The usefulness of any of these techniques depends on the specificity and stability 
of the marker with time. 
Considering the nature of my study I have decided to use genetic markers arid sero­
logical markers for the identification of introduced strains. 
563 
MATERIALS A N D M E T H O D S 
Soils 
Studies were carried out in three different soils and details of these soils are given in 
Table 1. 
Table 1. Details of the soils 
Soil type Description 
Lancelin brown sand (Australia) pH 5 5 ; Nitrogen content—very low 
Agalawatta soil series (Sri Lanka) pH 5 1 ; Nitrogen content — 0 118% 
Homagama soil series (Sri Lanka) pH 4 8 ; Nitrogen content — 0 • 107 % 
Rhizobium strains 
Details of the rhizobia used in this study are given in Table 2. 
Table 2. Rhizobia! strains used in this study 
Strain Description 
RRISL 09 Isolated from Pueraria phaseoloides. Found to be effec­
tive on P. phaseoloides. 
RRISL 09S a o o C L 1 0 0 Streptomycin (200 ppm) and Chloramphenicol (100 ppm) 
resistant mutant of RRISL 09. Effective on P. phaseo­
loides as its parent 
CB 756 Received from culture collection in Queensland. Effective 
on P. phaseoloides. 
Production of mutant strain 
Mutants were obtained as described by Schwinghamer and Dudman (1973) except 
that multiple steps were used in my study. When mutants had acquired the expected 
level of streptomycin resistance, they were subjected to chloramphenicol and their resis­
tance developed upto 100 ppm. 
Production of serological markers 
CB 756 and RRISL 09 were introduced into rabbits to obtain antisera against them. 
Young growths of bacteria from surface of yeast manitol agar were harvested in sterile 
physiological saline and were injected into 6 months old rabbits at 14 days intervals. 
Rabbits were bled after four injections from the marginal vein of the ear. 
564 
Host plant 
Pueraria phaseoloides plants were grown in plastic pots containing unsterile soils. 
Basal nutrients without nitrogen were added and pots were maintained at about 20% 
water holding capacity. 
Estimation of natural rhizobial populations 
The method used for counting rhizobia was the same as the five fold serial dilution 
plant infection test of Brockwell (1963). 
Sowing and inoculation 
Acid treated seeds were introduced at about 2 cm depth. Seedlings received one ml 
of appropriate bacterial suspension containing approximately 1 x 10 8 cells per ml. 
Sampling 
Nodule occupancy by the introduced strain after 1 1/2 months was the criterion for 
the competitiveness. 
Rhizobial strains which were introduced to different types of soils and methods of 
identification after recovery are given in the Table 3. 
Table 3. Rhizobial strains which were introduced to different types of soils and methods 
of identification after recovery 
Type of soil Strain Method of identification 
Lancelin sand RRISL 09 Using the genetically marked mutant of RRISL 09. 
Agalawatta series RRISL 09 Gel immune diffusion 
CB 756 Gel immune diffusion 
Homagama series RRISL 09 Gel immune diffusion 
CB 756 Gel immune diffusion 
All nodules from selected plants were pooled and 20 nodules were selected randomly 
from each treatment to determine the percentage of nodule occupancy by the introduced 
strains. 
Identification 
(a) Using genetic markers 
Nodules were surface sterilized and streaked on yeast manitol agar and antibiotics 
incooperated yeast manitol agar. 
565 
Using serological markers 
Growths from the nodules were used for immunodiffusion reactions. Immunodiffu­
sions were performed in 4 mm thick gels as described by Dudman in 1964. The undiluted 
antisera were introduced into the central wells and suspensions of the appropriate antigens 
were put into two of the outer wells, diametrically opposite each other. The strains to 
be identified were placed in the four remaining wells. 
R E S U L T S 
Population estimations 
Naturalized rhizobial populations in experimental soils are given in Table 4. 
Table 4. Estimation of naturalized populations 
Estimated number Confidence limits 
Type of soil (Per gram of soil) (95%) 
Lancelin brown sand 4 0 1 2 — 12 8 
Agalawatta soil series 14-1 X 10 1 5 3 — 3 8 1 x 10 1 
Homagama soil series 10 8 x 10 1 4 - - 28 7 x 10 1 
Percentage of nodule occupancy by the introduced strains in different soils is given 
in the Table 5. 
Table 5. Percentage of nodule occupancy by the introduced strains in different soils 
Soil type Strain % of occupancy by introduced strain 
Lancelin sand RRISL 09 90 
Agalawatta soil RRISL 09 3 5 
CB 756 25 
Homagama soil RRISL 09 30 
CB 756 20 
Competitive success of introduced strains in Sri Lankan and Australian soils is 
given in Table 6. 
566 
Table 6. Competitive success of introduced strains in Sri Lankan and Australian soils 
Sample 1 Lancelin brown Sample 2 Agalawatta and 
sand (Australia) Homagama soils 
(Sri Lanka) 
No . of nodules formed by 
introduced strains 18 
N o . of nodules formed by 
indigenous strains 2 
Competitive success of introduced 
Rhizobia 0 90 
(This difference is highly significant, using normal approximation to binomial distribution) 
DISCUSSION 
These studies indicate that introduced rhizobia establish differently in different soil 
conditions. The introduced strains produced a majority of nodules in Lancelin brown 
sand in Australia and fewer number of nodules were produced by the introduced strains 
in Sri Lanka. 
This may be due to the wide spread of cowpea type of' Rhizobium and their high com­
petitive ability in Sri Lankan soils. Plant infeclion counts revealed that Agalawatta and 
Homagama soils in Sri Lanka have thirty times higher bacterial populations than 
Australian soils. 
These data clearly show why promiscuous host plants in tropical soils rarely respond 
to inoculation. On the other hand introduced rhizobia produced a major proportion 
of nodules in Lancelin brown sand with P. phaseoloides even though this host is a promis­
cuous type. 
This disparity may be due to the presence of very low numbers of indigenous rhizobia 
in tested soils. In 1980 Ikram and Broughton also noticed the sametrend with Psopocarpus 
tetragonolobus when inoculated with Rhizobium cowpea complex strains in soils with 
low numbers of naturalized flora. 
These results clearly show us why introduced rhizobia under Sri Lankan conditions 
do not produce higher yields as in Australia or in other temperate countries. 
Therefore the poor response towards the inoculation in tropical soils is mainly due 
to the poor competitive ability of introduced strains and not due to their ineffcctivity. 
567 
A C K N O W L E D G E M E N T S 
We thank Dr O. S. Peries, Director, RRISL and Dr A. de S. Liyanage, Deputy Director, 
RRISL for their encouragement and guidance. We also thank Mr D . S. Wettasinghe 
and Mrs P. C. Wettasinghe for their technical assistance. 
R E F E R E N C E S 
Brockwell, J . , ( 1 9 6 3 ) . Accuracy of a plant infection technique for counting populations 
of Rhizobium trifolii. Appl. Microbiol., 1 1 , 3 7 7 — 3 8 3 . 
D U D M A N , W. F., ( 1 9 6 4 ) . Immune Diffusion Analysis of the Extracellular Soluble Antigens 
of Two Strains of Rhizobium meliloti J. Bad., 88, 7 8 2 — 7 9 4 . 
I K R A M , A. A N D B R O U G H T O N , W. L. J. ( 1 9 8 0 ) . C. Rhizobia in Tropical legumes — I X 
Pot and Field Trials with Inoculants for Psophocarpus tetragonolobus (L) Dc. Soil 
Biol. Biochem., 1 2 , 2 0 3 — 2 0 9 . 
S C H W I N G H A M E R , E. A. A N D D U D M A N , W. F. ( 1 9 7 3 ) . Evaluation of Spectinomycin Re­
sistance as a Marker for Ecological Studies with Rhizobium. J. Appl. Bact., 36, 2 6 3 . 
5 6 8 
SESSION 10. M A R K E T I N G 
INVESTMENT OPPORTUNITIES IN THE RUBBER 
INDUSTRIES IN SRI LANKA 
By 
M . N A D A R A J A H * A N D G . T . U N A M B O O W E * * 
(Mackies Ltd., Sri Lanka* and Industrial Development Board, Sri Lanka**) 
A B S T R A C T 
Sri Lanka's connection with rubber started in 1876 when Sir Henry Wickham brought 
70,000 rubber seeds from Brazil to Kew Gardens, England, germinated them and brought 
2,800 plants to Sri Lanka. They were planted in the Botanical Gardens at Peradeniya and 
at Heneratgoda where some are still alive today. These trees provided the seeds for the 
rubber plantations in Sri Lanka. 
Natural rubber (NR) is of vital importance to the economy of Sri Lanka. Next to 
tea, it is the most important commodity being the second major contributor to agricultural 
export earnings and employment generation. The natural rubber sector accounts for about 
10% to Sri Lanka's total gross export earnings and provides employment for about 500,000 
persons. It also contributes a base for the production of industrial rubber products, thus 
providing additional employment to several thousand persons. 
Investment opportunities 
Table 1 shows the extent of rubber cultivated in Sri Lanka from 1904 onwards. The 
rapid expansion of rubber cultivation at that time necessitated the setting up of the Rubber 
Research Institute in 1909, and by 1930, an acreage of 235,000 hectares were planted 
with rubber. Please note that the acreage in 1982, 52 years later was only 206,000 
hectares, well below the 1930 figure. This decrease would mean that there is now 
no investment opportunity for new rubber planting in Si i Lanka and only replanting. 
Hence future investment in the rubber industry in Sri Lanka would be, 
(1) To improve our raw rubber processing 
(2) To manufacture specialised grades of natural rubber and 
(3) To manufacture rubber products. 
Table 1. History of cultivation of rubber in Sri Lanka 
Year Acres Hectares 
1904 
1910 
1920 
1930 
1945 
1960 
1980 
1981 
1982 
25,000 10,125 
118,000 47,790 
460,000 186,300 
580,260 235,005 
659,803 267,220 
668,948 270,923 
561,757 227,512 
508,685 206,037 
508,620 205,991 
569 
Table 2 shows the consumption of N R & SR up to 1983. You would note that in 
thel930's and 1940's when Sri Lanka had its maximum acreage under rubber, there was 
no competition from SR. Today SR is consumed at a level of 8 million tonnes and N R 
at a level of 4 million tonnes and we are now in no position to set standards for our natural 
rubber. We will have to follow the standards set by synthetic rubber producers. 
D r B . C. Sekhar in his address to you said that N R deserves a price of 2 U S ? a kg but 
is only getting one US $ a kg. Why is this ? He also said about the 3P's namely purity, 
processing and performance and in formulating an investment programme for the rubber 
plantation industry in Sri Lanka these three P's must be kept in mind. 
Table 2. World NR and SR consumption (000 tonnes) 
Year N R SR Total rubber 
1930 722 . 722 
1940 1127 43 1,170 
1950 1750 589 2,339 
1960 2135 1850 3,985 
1970 2990 5635 8,625 
1980 3760 8685 12,445 
1981 3900 8465 12,365 
1982 3655 8005 11,660 
1983 3940 8185 12,125 
Table 3 shows the raw rubber processing facilities available in Sri Lanka. You 
would note that it is highly accented to latex crepe manufacture. It is predicted that 
because of replanting, the rubber production in Sri Lanka would reach 200,000 tonnes per 
year by the end of this century. It is recommended that without the addition of new 
latex crepe factories, the present crepe factories should be rehabilitated to run efficiently. 
If this recommendation is accepted, then the Rubber Controller should stop without 
delay the factory subsidy for new latex crepe factories. 
Table 3. Facilities available for processing NR in Sri Lanka 
Factories 
Organisation Latex RSS TSR Scrap Centrifuged 
crepe crepe latex 
State Plantations Corporation 
(SPC) 70 61 
— 
5 
— 
Janatha Estates Development 
Board (JEDB) 53 30 
— 
6 1 
State Rubber Manufacturing 
Company 6 
— 
2 — 1 
Co-operatives 
— 
100 
— — 
1 
Private sector 20 71,292 6 27 3 
roller units and 
132, 242 smoking 
facilities 
570 
It would be also seen that the SPC or JEDB have no alternative forms of4 manufacture 
for latex crepe. It is possible that at times latex crepe prices will be depressed and pro­
vision should be therefore made for alternative forms of raw rubber manufacture. The 
SPC or JEDB should build new factories based on centrifuged latex and TSR manufacture. 
Dr Tillekeratne in his paper today emphasised the need for more TSR manufacture in 
Sri Lanka. At present no subsidy is given for centrifuged latex manufacture. The 
Government should decide to give approval for centrifuged latex to be included in the 
Factory Development Scheme to be entitled for subsidy. The necessity for subsidy for 
factory development is because of all the N R producing countries, Sri Lanka has the 
highest export duty and if Sri Lankan rubber is to get good prices, the Sri Lankan Govern­
ment should divert part of this export earnings on investment in raw rubber processing 
to get a better return for the rubber exported. 
Table 4 shows the export of natural rubber by types. It will be seen that though 
the small holders produce most of the raw rubber in Sri Lanka, latex crepe production 
was 62,000 tonnes of the 124,000 tonnes of raw rubber exported. This means that a fair 
amount of small holders latex is being converted to latex crepe. Dr B. C. Sekhar's first 
P was purity. We would normally think, that latex crepe, satisfies this requirement easily, 
at all stages of production till it reaches the shippers stores. We do not think so. Purity 
is very important. The RRI officers should visit crepe factories and recommend to the 
management the steps to be taken in eaclt crepe factory to ensure purity. 
Table 4. Exports of natural rubber by types 
Rubber Crepe ubber Technically Latex 
Year smoked Sole Scrap Latex specified & master- Total 
sheet rubber batch 
(TSR) 
1977 77,051 4,218 16,557 34,878 1,825 — 134,529 
1978 77,434 4,569 14,111 36,776 5,156 
— 
138,046 
1979 62,611 4,851 13,740 33,881 13,105 
— 
128,188 
1980 60,597 4,472 8,117 38,253 8,451 52 120,942 
1981 70,898 4,107 8,354 34,307 14,644 213 132,523 
1982 72,040 4,145 7,342 35,850 10,788 1,156 131,302 
1983 51,770 3,021 2,508 62,013 5,427 492 124,441 
All flooring should be tiled, all crepe factory roofs should have a ceiling so that dirt 
does not fall from the roof on to the rubber, all crepe sent to brokers stores and to the 
auctions should be in perforated polythene bags to minimise dirt contamination in transit. 
As regards RSS produced by small holders it is possible to upgrade RSS in TSR factories 
and to bring their dirt content below 0 05%. N o rubber made from latex should be 
exported from Sri Lanka which has a dirt content greater than 0-05%. No scrap rubber 
having dirt content over 0 5 % should be exported. These can be implemented if we all 
agree that P — purity of NR is an important property to get better prices for our natural 
rubber. 
571 
When considering the next P i.e. processing, it is very necessary to manufacture Sri 
Lankan rubbers of different Mooney viscosities, as required by the consumer and this 
could vaiy from 20 to 70. This is possible by the new simple inexpensive process deve­
loped by the Ceymac Rubber Co., Ltd. 
In this context it is necessary to say a few words on rubber processing. A very serious 
defect of NR is that it is mostly supplied at rather high viscosities whilst SR does not have 
this viscosity defect. Because of this high viscosity, N R has to be premasticated before 
rubber chemicals can be mixed into it. Premastication is both time and energy consuming. 
At present only a very small amount of NR is supplied as viscosity stabilised grades and 
even these rubbers have a Mooney viscosity of 60 and 68 which in my opinion are on 
the high side. If the consumers were asked to specify the viscosity, they may request a 
Mooney viscosity of 45 and N R producers must give the viscosity which the consumer 
wishes to have. 
A very simple inexpensive method to obtain any required Mooney viscosity from 
any grade of dry rubber is urgently required and such a process has been developed by 
Ceymac Rubber Co., Ltd. Consideration of this must be given when future investment 
in raw rubber processing in Sri Lanka is considered. I would say that soon, at least 
50% of Sri Lankan rubber will be exported in Mooney viscosities requested for by the 
consumer. 
Coming now to the third P which is performance, I would state that the Rubber 
Chemists and Rubber Technologists have enough of academic experience and knowledge 
and in future their energies should be directed to visit consumers of Sri Lankan rubber, 
know their problems and tailor make our natural rubber so that in performance it would 
satisfy the consumers needs. This is done by SR producers and N R producers will have 
to follow suit if they wish to get good prices for their NR. 
Exports of N R from Sri Lanka 
More than 85 % of the natural rubber produced in Sri Lanka is exported and only 
less than 15% of the natural rubber produced is used to manufacture rubber products. 
The exports of natural rubber by types during the period 1977 — 1982 have been given 
in Table 4. 
Latex crepe 
Sri Lanka is the world's largest producer of latex crepe and markets it at present on 
visual specifications. It is recommended that special types of latex crepe based on con­
sumer requirements be manufactured and marketed on technical specifications. Invest­
ment opportunities exist for the manufacture of such specialised grades such as CV crepe, 
de-proteinised crepe etc. Yellow fraction crepe has different properties from normal 
latex crepe and should be shipped as a special grade preferably as a TSR grade in 33 1 /3 kg 
bales and if necessary as a CV grade. 
572 
Ribbed smoked sheet (RSS) 
RSS is mainly produced by smallholders in Sri Lanka and with the increased replant­
ing being done by smallholders the tonnage of RSS produced will increase. Most of 
the RSS produced in Sri Lanka is exported in 111 • 1 kg bare-back bales and is not techni­
cally specified. A serious drawback of Sri Lankan RSS is its high Mooney viscosity. 
The value of the RSS could be further considerably improved if it can be produced as a 
CV grade. Thus whilst smallholders RSS currently produced is of satisfactory quality, 
its value can be considerably enhanced for export and its marketability improved by 
converting it in TSR factories as a constant viscosity RSS and exporting it as a SLR 5 
(RSS) CV grade. There is considerable investment potential for this type of processing, 
using RSS produced by smallholders and carrying out the operation in TSR factories. 
Technically specified rubber (TSR) 
Technically specified rubber is exported in small bales (33 1/3 kg) in one tonne pallet 
crates. This type of export is suited to container shipment which is economical in freight 
rates. There exists investment potential for exporting converting and exporting RSS 
and latex crepe as TSR (CV) grades in small bales. 
Sole crepe 
Malaysia was the largest plantation sole crepe producer with about 20,000 tonnes 
in 1974 and with 8,000 tonnes in 1982. The sole crepe production in Malaysia is declining 
and sole crepe factories are closing down there due to the high cost and non availability 
of suitable labour to hand laminate the sole crepe and is expected to be below 1,000 tonne 
production this year. Sri Lanka can step in and fill this shortfall, as sole crepe manu­
facture is still profitable at Sri Lankan labour costs. 
To manufacture one tonne of finished sole crepe/day would need an additional work­
force of 50 persons per day over pale crepe. Since the manufacture of sole crepe gives 
employment to people in rural areas and earns more foreign exchange, investment oppor­
tunities for its manufacture should be aclively encouraged by the Sri Lankan Government. 
Very soon Sri Lanka may be the only N R producing country producing plantation sole 
crepe. Since producing countries have not been able to meet the world demand for 
sole crepe, the consuming countries have been compelled to produce industrial sole crepe 
which is inferior in quality to plantation sole crepe. This industrial sole crepe is produced 
at about 20,000 tonnes annually. The annual exports of sole crepe from Sri Lanka has 
never exceeded 5000 tonnes and was as low as 3021 tonnes in 1983. 
With proper marketing policies the production of Sri Lankan sole crepe can be easily 
stepped up. A reasonable target for increased sole crepe production in Sri Lanka is 
about 1,500 tonnes each year for about 3 years to give an increased output of 4,500 
tonnes per year at the end of 3 years. Sole crepe could be considered as a rubber 
product and to achieve the above target it is very necessary for the Sri Lankan Govern­
ment Authorities to decide to give sole crepe the incentives which rubber products enjoy 
at present such as duty relief, tax holidays and investment relief. 
573 
Supply of field latex, and of centrifuged latex and manufacture for export of latex based 
products 
During the past decade, the rubber product manufacturing industries in some of the 
N R producing countries have experienced very rapid growth. However the only NR 
exporting country which exports more rubber products than importing rubber products 
is Malaysia. For the rubber product industry to expand there'must be locally produced 
ancillary raw materials and also centrifuged latex available in quantity and at reasonable 
prices. 
Sri Lanka does not produce ancillary raw materials. Because of this lack of an­
cillary raw materials, the rubber based products most suitable to Sri Lanka are latex 
based products. The latex products identified as being most suitable are dipped pro­
ducts, latex thread and rubberised fibre. The world needs annually 105,000 tonnes of 
rubber as dipped products and 35,000 tonnes as latex thread. In 1982 Malaysia exported 
9,000 tonnes of rubber as dipped products and 3,500 tonnes of rubber as latex thread and 
earned around a billion rupees from the export of these latex products. 
Sri Lanka has been very backward in this respect as regards the manufacture of 
latex based products. A major problem is the supply of centrifuged latex at fair prices 
to users. Only the JEDB and to some extent the State Rubber Manufacturing Com­
pany sell centrifuged latex, but their production is not at all enough to meet the demands 
of the latex based industry. Hence rubber latex product manufacturing industrialists 
have to set up their own centrifuges to produce their own centrifuged latex. The supply 
of field latex to operate these centrifuges can be got from big estates or from smallholders. 
Nearly all the big estates in Government hands are managed by the JEDB and SPC. 
The two organisations namely the SPC and JEDB are agreeable to supply field latex at 
IX latex crepe prices. This is too high and private centrifuged latex manufacturers are 
reluctant to buy at this high price as a reasonable price would have to be worked out. 
It may be mentioned that in 1954 when the rubber estates in Sri Lanka were in pri­
vate hands, around 5,000 tonnes of rubber were produced as centrifuged latex and ex­
ported whilst now 30 years later even half that amount of centrifuged latex cannot be 
produced to feed latex based industries though the resulting products would earn on 
exporting a much higher foreign exchange than if the rubber was exported in raw rubber 
form. It is expected that a considerable amount of latex thread and dipped products 
manufacture would in the next decade shift to Malaysia, Thailand and Sri Lanka. The 
amount involved is upto 140,000 tonnes and Sri Lanka should endeavour to take its 
share. 
The Ministry of Plan Implementation has recommended in Feb. 1981 on pi 3 of 
their report on "The natural rubber latex based industry" in section 3 5 as follows. 
" Since JEDB and the SPC have the monopoly of the supply of high quality field latex, 
an appropriate government policy making body could direct these organisations to make 
available estate field latex to centrifuged latex manufacturing organisations at RSS minus 
C O P or average latex crepe minus C O P prices, whichever is higher. Such prices could 
be determined on average monthly figures and announced as fixed prices for each month." 
574 
It is suggested that the above recommendation be given careful consideration and 
implemented as there is still considerable investment opportunities for the manufacture 
in Sri Lanka for latex products for export if the raw material namely centrifuged latex 
or field latex for centrifuging is readily available at fair prices. At present Sri Lanka 
does not export even 1,000 tonnes of rubber as latex based products. The main products 
are gloves from Dipped Products Ltd and latex thread from Multistretch, The export 
of rubberised fibre is also being initiated in a big way. 
High export duty in Sri Lanka and its use in raw rubber processing 
Of the N R producing countries, the highest duty and cesses are in Sri Lanka followed 
by Thailand and Malaysia with there being little or no duty and cesses on Indonesian 
rubber. This is one of the factors which has limited investment in the raw rubber industry 
in Sri Lanka. It has been recognised that part of the duty should be ploughed back 
into the rubber industry with a view to developing it. This is being done by way of re­
planting subsidies and factory development subsidies. The Sri Lankan Government 
subsidies for factory development is paid by the Rubber Controller for factories which 
manufacture a premium grade rubber after 1974 and takes the form of l/3rd grant'on 
approved equipment. This scheme is intended to serve as an incentive to rubber pro­
ducers to increase the manufacturing capacity of rubber factories producing premium 
grades of rubber. The premium grades of rubber identified are latex crepe, sole crepe 
and TSR. This scheme was drawn up 10 years ago and needs revision to include con­
centrated latex such as centrifuged latex and the manufacture of speciality rubbers such 
as PA 80 and Hevea plus MG rubber and speciality latices. 
Investment opportunities to manufacture rubber based products 
The rubber products manufacturing industry in Sri Lanka has its beginnings in the 
1930's and 1940's with the tyre retreading industry and in 1947 the annual consumption 
of N R was 72 tonnes. In the 1960's the manufacture of latex based products, bicycle 
tyres and tubes and ebonite battery boxes commenced and annual consumption rose to 
807 tonnes in 1960 and 3,720 tonnes in 1970. The establishment of the Sri Lanka Tyre 
Corporation in 1968 which manufactures most of Sri Lanka tyre requirements boosted 
the consumption of N R considerably and by 1972, Sri Lanka's N R consumption was 
6,109 tonnes. After 1972 efforts were made by industrialists to export rubber products 
but no significant achievement was possible till 1977. In 1976, 7,269 tonnes of N R was 
used and Rs. 4-36 million of rubber products were exported. The value of the rubber 
products was Rs. 225 million and provided employment for about 15,000 persons. 
After 1977 because of the drastic changes effected in the economic policies of Sri Lanka, 
rubber product exports were developed. Initially liberalisation of imports enabled 
industrialists to procure vital raw materials without difficulty and to import new machinery 
and equipment. These factors enabled a higher quality finished products suitable for 
export to be produced. The establishment of the Greater Colombo Economic Commission 
(GCEC) in 1977 encouraged foreign investment in the rubber industry to manufacture 
rubber products for export. Some of the investment is from foreign collaborations who 
work on joint ventures projects with local industrialists and entrepreneurs. The opening 
of the Industrial Processing Zones (IPZ) in Katunayake and in Biyagama accelerated this 
investment. There are now about 180 rubber products manufacturing units in Sri Lanka 
of this 28 are large scale factories with six large scale factories being in the IPZ and 22 
outside. 
575 
The IPZ factories cater to the export market and they enjoy certain benefits such as 
duty exemption on imports but are also subjected to certain restrictions, the main restric­
tion being that they are not normally permitted to sell their products to the local market. 
In the past 1977 period, the output of the rubber product industry has increased from 6,700 
tonnes in 1977 to 16,300 tonnes in 1981. The large export oriented factories have contri­
buted most to this favourable change. These firms are making products such as latex 
thread, rubber gloves, rubber bands, canvas footwear, rubberised fibre and solid tyres. 
Promotion and development of the rubber industry in Sri Lanka 
It is very necessary to have organisations which understands problems and anxieties 
of those who have invested in the rubber industries in Sri Lanka. The GCEC is an orga­
nisation involved in the administrative setting up of rubber industries in the Free Trade 
Zone (FTZ). The Government organisations which can assist with technical problems 
encountered by the rubber products industrialists are the Rubber Research Institute of 
Sri Lanka (RRISL) and the Ceylon Institute of Scientific and Industrial Research (CISIR). 
In 1977 the Ministry of Industries and Scientific Affairs set up a Rubber Technical Service 
Centre under the Industrial Development Board (1DB) with the main target of uplifting 
the small and medium scale sectors. The necessary financial assistance for this project 
was provided by the World Bank and the International Development Agency. Since the 
accent now is on the production of rubber products for export it is now very necessary 
for the IDB, Technical Centre to also actively provide services for the big scale industries 
which are producing for export. The IDB should also assist in enabling small and medium 
scale industries to subcontract in the manufacture of certain items for big scale industries 
who produce goods for export. Further there should always be very active liaison bet­
ween the RRISL, IDB (TSC) and CISIR in solving problems faced by big rubber indus­
trialists. 
Government action to develop the rubber industry y 
The Ministry of Trade and Shipping on the recommendation of the Export Develop­
ment Board has appointed an Advisory Committee to monitor the progress and to advice 
on measures to be taken for the future advancement of the rubber and rubber products 
industry in Sri Lanka. This Committee has noted with concern the slow export growth 
of the rubber products industry in Sri Lanka. This Committee mentions that the absence 
of adequate export infrastructure, has impeded the export performance of the rubber 
products. Export infrastructure comprises those services which are basic to the efficient 
performance of export business and which individual enterprises cannot provide for them­
selves on an economic basis. Examples include shipping and freight handling, export 
credit, foreign market research, expoit promotion etc. Adequate provision of most of 
these services at reasonable rates is the proper responsibility of Government whose ini­
tiative and financial support are vital. The committee was of the view that the export 
of rubber in the form of finished products will give about 3 times the foreign exchange 
earnings of the same quantity of rubber. Hence the committee has set a target of 40,000 
MT tonnes of N R for industrial consumption in Sri Lanka by 1990. This should enable 
20,000 tonnes of rubber to be exported in the form of finished products from Sri Lanka. 
576 
This committee has also recommended monetary benefits, namely ; 
1. Interest free loans up to 50% of the capital cost of building and machinery 
2. Five year tax holiday on profits 
3. Investment relief of a minimum of 50% 
4. Duty free import of machinery 
5. Removal of BTT on imported machinery and chemicals 
6. Provision of promotional freight rates 
7. Working capital at interest rates not exceeding 12% 
The committee has also recommended that a Quality Control Unit be set up with 
UNIDO assistance to ensure adequate quality of products for export. 
REFERENCES 
Commonwealth Development Corporation (1979). Report on the rubber industry 
master plan study, 1 — summary report. 
NADARAJAH , M., D E SILVA , K. P. N. , D E SILVA , G. A. AND DHARMASENA , R. P. (198-*). 
Viscosity control of natural rubber in the solid state. Sri Lanka Patent Appl. No . 9421. 
577 
THE SPECIAL PROBLEMS OF MARKETING RUBBER 
By 
G U Y SIRIMANNE 
(Ceylon Trading Company, Colombo, Sri Lanka) 
It was not intended that this paper bears the title it does, but reflecting on a suitable 
subject to speak about to a diverse group it transpired that structural difficulties, in mar­
keting rubber emerged as if of their own accord. Such spontaneity was not to be denied, 
specially, if it did not fall on entirely deaf ears. It then occurred to me that I have been 
given a rare opportunity, to kill several birds — buyers, administrative officials, brokers 
and planters — with one stone. In this country there is, I was going to say, unconscious 
suspicion rather a very definite and conscious mistrust of the middle man who is supposed 
to be raking in the shekels. Licencing, customs, exchange control and every other con­
ceivable authority looks askance at the activities of the rubber exporter. The initiated, 
I would have expected to look beyond cliches. Unfortunately, in Sri Lanka where the 
export of rubber is concerned this phenomenon is very much entrenched and is reflected 
in the rules and regulations the rubber exporter is subject to. Pure trade factors get 
more involved as does the economic background against which trade takes place. 
In our day to day operations there is much criticism of the Colombo market by 
dealers and buyers in the major marketing centres, mostly concerning price. Justifiably 
so, since the same and many substitutable grades are most of the time, available at lower 
prices from other origins. 
Without seeming to be complaining about conditions as they exist, there are however 
some factors which stand out that preclude us from being quite as flexible as we should 
be as an exporter country. Some are so fundamental that they cannot be changed in 
the short term, others (more) should be subject to intensive scrutiny by policy makers. 
The most fundamental of these is that production in Sri Lanka amounts to a mere 
140,000 tonnes a year making it one of the smaller origins operating in the international 
market place. The quantity available for export operations is extremely limited in the 
context of the structural arrangements in the international field. The exporter in Sri 
Lanka has necessarily to be supply dominated which in itself influences individual deci­
sions re price by way of an inbuilt insurance. At any given time an accurate assessment 
of availability is crucial. While some buyers are perhaps more tolerant than others, 
contractual terms as they are, securing extensions on delivery periods is certainly not 
automatic. Permission has to be sought down the line and is not always forthcoming. 
Besides it is likely that all exporters have been similarly affected and a replacement im­
possible- This alone tends to be a restraining factor either by way, of the limiting of 
the quantity or increasing the price as a hedge against an unexpected change in the supply 
situation. 
In our daily communication with marketing centres or in Market Reports it is almost 
embarrassing to mention flood or drought with monotonous regularity, but this happens 
to be so. Prudent stockpiling early this year against the historical lean period turned 
579 
out the proverbial albatross and a windfall for buyers, as exporters endeavoured to divest 
themselves of languishing stocks, improve their liquidity position and continue operations. 
A more positive approach to marketing later in the year in anticipation of a weaker mon­
soon turned out that the rubber producing areas suffered the worst flood since 1957. 
Exporters spent an agonising 3 weeks thereafter endeavouring to secure quantities that 
were not available at high prices in turn making overseas sales at remunerative prices 
impossible. 
Rubber plantations are almost without exception situated within the South-west 
quarter of the country and are by and large subject to the same weather conditions. Be 
it flood or drought, an alternative source of supply is just not available. Stupid as it 
may seem, exhaustive efforts to secure quantities for immediate commitments only result 
in exorbiant prices with no real alleviation of the original circumstances. Other producing 
countries consist of considerable land mass, with widespread acreage and as such weather 
conditions do not have quite the same ramifications as they do here. While weather 
conditions are beyond our control an otherwise assiduously attended trading position 
does not necessarily buffer you from other shocks. For instance, an incorrect trading 
decision by a larger exporter to sell short could throw the market out of line with other 
origins, circumscribing turnover. Stocks of the commodity previously purchased at 
a lower price cannot be traded profitably as they would be subject to the licensing scheme 
which is based on current levels. 
Since Colombo is a small market, price fluctuations are wide and movements fast 
causing considerable distress. It is often said that wide and frequent fluctuations offer 
better opportunities for profit, perhaps true, but with definite limitations in Colombo in 
the absence of the usual tools available to traders at other origins. 
The contribution the export duty makes to the misalignment of Colombo prices is 
considerable. At present it comprises 25 % of the most expensive type of natural rubber 
exported from the country, sole crepe and 45 % of RSS5. So much so that the lower 
grades of crepe rubber and flat bark crepes, are rarely marketable except at exceptional 
times. Export markets for these grades have been lost to Africa. Fortunately, for the 
producer the gap has been filled by manufacturers of retreaded tyres for local consump­
tion. The export duty in this country is based on the Singapore price for physical RSS1 
for the current month and with the usual run-up in prices prior to the position being 
closed out, registers an increased export duty which is reflected in the price the following 
week, irrespective of whether international markets for Colombo grades are firmer or 
easier. 
Since a terminal market does not exist in Colombo and exchange control regulations 
prevent arbitrage, no hedging facilities are available. The idea has been mooted and 
perfunctorily examined but no information is forthcoming whether an intensive study 
has been carried out. The Colombo market is now expanding to include manufacture 
of rubber based goods and the full spectrum of requisite participants is available for the 
efficacious functioning of such a market. The presence of such a facility would consider­
ably extend the operation of marketing. Producers in a better position of a more pre­
dictable return are likely to develop a more benign attitude to forward business and to 
be less intransigent. At present forward operations particularly in the better grades of 
580 
iatex crepe are wholly inequitable with an open-ended price mechanism, it can be rea­
sonably expected that the setting up of such a facility would considerably mellow the 
extremes the Colombo market is so frequently subject to. If their buying programmes 
at present are anything to go by, the securing of raw material by industrialists on the 
physical market alone seems a nightmare. The duel operation on the contrary would 
make for a more constant cost factor of their raw material. 
Many of you are aware there is a licensing scheme in operation in Colombo and is 
operated by the Department of Commodity Purchase, for and on, behalf of the Depart­
ment of Exchange Control. For all the crepe grades and sole crepe the scheme is based 
on the price paid at the bi-weekly auctions. The scheme for ribbed smoked sheet is 
based on the prices paid on the London physical market. The scheme itself is probably 
necessary as malpractices are not unknown and is a safety net and an inducement to as­
piring rubber exporters, which is in line with government policy. From the purely com­
mercial point of view the disadvantages are obvious particularly the scheme for crepe 
rubber which presupposes a back-to-back operation. This could not be further from 
reality. The kind of organisation required to trade in rubber costs a great deal of money 
as does the actual operation including finance, store facilities, communication and staff. 
By the time operating costs and overheads are added to the price of the rubber, overseas 
offers can be expensive. 
There is no secret about the fact that exporters live on taking views, more correct 
ones than incorrect. All other factors being equal, exporters endeavour to buy rubber 
when they consider it cheap and sell when they conclude indications or bids are thought 
of as high. Between the two operations much can happen and an exporter can end up, 
selling at lower levels than cost and is eventually a salvage operation, when every month 
the rubber lying in a warehouse between purchase and shipment costs 20 cents per kilo 
per month. Neither is the commodity merely subject to supply and demand. The 
comment is increasingly heard that rubber exporters are now trading more in currencies 
with which they are not entirely conversant. 
Freight allocation in and out of Colombo is controlled by the Central Freight Bureau 
and whereas the overall management has resulted in a high degree of rationalisation in the 
port, the bias is in favour of shipping lines. They have aggressively monitored freight 
rates protecting exporters from arbitrary increases and with an effort at remunerative 
operations both for the lines and the users. The exporters' more parochial interests are 
not however served by way of competitive freight rates. The present system in its very 
essence does not permit the existence of non-conference vessels with the competition 
thereby implied. This facility, from all reports, exists at other origins enabling them to 
ship their produce at lower rates. The freight component in export prices from Malaysia 
& Indonesia, are 8 5 % and 8 % respectively compared to Sri Lanka's 9 5%. 
Financing in Sri Lanka comes very expensive and it could hardly be believed that in 
a country so dependant on its exports, as Sri Lanka is, rubber exporters are given so little 
encouragement in moving out of the country one of its major cash crops. Packing 
credits have been available for sometime and give some relief though they are restrictive 
in operation. A comparatively new re-financing scheme for export is in operation 
though this too is limited. The instercst rate payable by rubber exporters in Sri Lanka 
581 
is an average 20% compared to 13%, 6% and 17% in Malaysia, Indonesia and Thailand 
respectively. The best evidence of the inadequacy of financial support from the banking 
system is seen during the major and minor cropping seasons i.e. usually December/January ^ 
and July/August, when after periods of substantial rainfall there is a marked increase in 
the arrival of rubber in Colombo and are perennial periods of lower prices irrespective 
of international markets. The only way exporters can tide over these periods is to offer 
lower prices overseas in an effort to move th« commodity which operation the more astute * 
buyer is well aware of. If for any reason the shipment schedule is dislocated, it is common 
to find big exporters sitting out the auctions. This occurrence continues year-in year-out 
and means a lower price for the producer and does not necessarily mean a fruitful time 
for local exporters but cheap procurements for overseas markets. 
The emerging importance of local industry as a market factor cannot be underesti­
mated. Upto 1977 rubber based industry in the country accounted for a negligible amount 
of the total rubber produced and was a mere 6,800 tonnes. The most upto date figure 
that of 1983 indicates that 16,500 tonnes equals 12% of our production is now consumed 
locally. This has had a considerable impact on the price structure. The absence of a « 
terminal market necessitates that the industrial consumer ensures a continuous availability 
of the raw material by stockpiling their requirements and this they do with the finesse 
of a steam roller. The international rubber market is far more sensitive to costs and 
with the previously mentioned conditions also taken into consideration, certainly leaves •* 
the exporter at a disadvantage except for when shipment oriented hysteria prevails. Many 
of the crepe rubber grades are now preserves of local manufacture. At the beginning 
local industry produced green parrots, balloons, erasers and rubber bands all of indifferent 
quality. Now it runs to the manufacture of tyres, retreading included, masterbatch, 
rubber thread and dipped goods to mention just a few. The manufacturers of technically 
specified rubber invade the auctions for fodder and conventional rubber is in no position 
to compete with the incentives given to the marketing of these comparatively new grades. 
When the phenomenon commenced local manufacturers were limited to the lower grades 
of scrap crepe rubber which were in any case labouring under the high level of export 
duty payable. Their operations have intensified and widened towards the more sophis­
ticated and their requirements now run to the full gamut of scrap crepe grades and well 
into the lower grades of latex crepes. We have on occasion seen a situation where the 
lower grades required by local manufacturers are at a higher price then the next best 
grade required by the exporter. 
It must be made very clear at this point that value added operations with their socio­
economic ramifications are well understood as desirable and not a complaint but meant 
to illustrate the impact these operations have on overseas prices. 
582 
BROKERS ROLE IN THE MARKETING OF 
SRI LANKA RUBBER 
By 
RANJITH PERERA 
( Forbes and Walker Ltd., Colombo, Sri Lanka ) 
With the rapid expansion of the rubber industry from the early 1950's and the growth 
of the activities of the exporters, broking activity in the country has expanded. Colombo 
is the centre for rubber broking activity and all Companys, partnerships and individual 
organisations dealing with such activity operate from this centre. Structurally brokers 
in Sri Lanka could be classified into two groups : 
The more organised system of broking Companys handling state owned produce 
and those from private proprietory plantations. Majority of these firms arc mem­
bers of the Colombo Brokers Association whilst a few operate outside and handle 
a fair share of business. 
The other group of brokers who carry on this function mainly in an area popularly 
known to the rubber trade as " GRANDPASS " are mostly partnerships and indivi­
duals. A broad distinction is that the larger and more organised form of broking 
firms whose expertise date back to the early 20th century handle rubber produced 
by the two large state organisations, JEDB & SLSPC and the produce of private 
proprietory owners where often the extent of the land is limited to 50 acres and less. 
These broking firms between them handle nearly 40% of the rubber produced in 
Sri Lanka. The main broking firms sell their rubber through the auction system 
which is the pivot on which the rubber market revolvos. The other category of 
brokers activities are confined to the rubber which trickles into the city from the 
outstation dealers who in turn purchase from smallholders in rubber growing areas. 
They perform a very useful function and help the dealers to dispose of their rubber. 
Millers who manufacture low quality scrap crepes also seek the assistance of these 
brokers to sell their rubber. They are also engaged in negotiating forward sales 
on FOB contracts. 
A limiting factor to the organised system of broking is that a large amount of rubber 
bypasses the auctions. The 1983 statistics reveal that out of a total production of 
approximately 133,000 m tons a quantity of around 54,000 m tons Was sold through the 
auctions. Rubber which bypasses the auction was disposed in the following manner. 
(a) Purchased by the Commissioner of Commodity Purchase 
(b) Despatched by outstation traders and dealers direct to the shippers 
(c) Shipped directly by the producers 
(d) Sold forward by the main brokers in Colombo 
or (e) Sold by manufacturers direct to local consumers as liquid latex. 
583 
The methods by which the rubber brokers perform their rubber marketing functions 
(a) The Auction System 
(b) Forward Sales 
(c) By Private Treaty 
a 
It is through the auction system that the brokers perform their most important mar­
keting function. The Colombo rubber market had its origin at the outset of the 20th 
century with the first rubber auction taking place on 4 November 1910. Since 
then from a mere 22 tons, currently on the average almost 800 to 1,000 tons comes under 
the hammer weekly. The auction system is the most organised method of marketing 
rubber in Sri Lanka. The conditions of sale are laid down by a set of rules formulated 
by the Ceylon Chamber of Commerce. Even where the sale of rubber is effected outside 
the auctions (FOB, Forward Delivery & Private Treaty) the conditions or sale and forms 
of contract tend to remain the same. The brokers conduct these sales having obtained 
an Auctioneer's license. The types of rubber auctioned are mainly crepe, scrap crepe, 
sheet rubber and sole crepe. * 
Prior to auctioning the rubber, a lot of preparations are made by the brokers. The 
work may appear to be routine but a lot of expertise is involved in the initial preparations 
with well trained staff at hand to assist in the work. Samples have to be drawn from daily 
arrivals of consignments, invoiced and despatched from rubber plantations. These 
numerous samples of different types of rubber which illustrate the quality of the bulk 
and are finally examined by the buyers prior to purchase at the auctions comes under 
the careful scrutiny of the brokers who in keeping with the standards laid down by the 
" Green book " grade and value the rubber. The catalogues depicting the " Garden 
mark " invoice No . and grade despatched by the estate are prepared by brokers and made 
available to the entire trade. Buyers make use of these catalogues to purchase rubber 
offered by the different brokers. 
Although market trends, the general demand and the degree of competition affects 
prime levels at the auctions, brokers try hard to elicit the best possible bids. When the 
market is bouyant activities at the auctions could be most vociferous with the brokers 
having a nightmare of a task picking out the bids. Whilst on a day when (he market 
sags, he has the unenviable task of extracting prices from equally determined buyers who 
would like to purchase goods at the cheapest possible level. 
Price levels obtained at the auctions after weighted averages have been worked out 
are the official quotations and are published in the news papers. These price levels directly • 
influence all sales of rubber outside the auction system. Auctions do have their dull periods 
but more often active conditions are experienced and the conditions generated have re­
sulted in producers benefiting by way of enhanced prices. The presence of new compe­
titors who give the traditional buyers a good run for their money and the recent increase <# 
in demand by the local consumers have been heartening to the producers. With the 
liberalization of the sheet rubber trade arrangements were made for this commodity 
to be auctioned after a long lapse. During the inaugural year, the total sold was only 
7,339 m tons whilst in 1983 the quantity increased to 21,515 m tons. Bulk of the large 
584 
buantity of sheet which bypasses the auction is brought directly to shippers godowns 
by outstation dealers whilst a small quantity is purchased by the Commissioner of Com­
modity Purchase through his depots located in rubber growing areas. The auction prices 
have been at higher levels and the smallholders should be encouraged to bring their rubber 
to the auctions in order that they may benefit by competitive prices. It is encouraging 
to note that the Rubber Research Institute (RRI) together with the Export Development 
Board have launched a scheme whereby smallholders are currently being assisted in 
manufacturing better quality grades. Brokers too have pledged their support to assist 
them in grading and sorting sheet rubber. 
In grading and evaluating the different types of rubber despatched for sale by public 
auction and forwarding character reports to the producers, the brokers are able to high­
light any shortcomings in the quality of the produce. A thorough scrutinity by the brokers 
helps to identify dullness in colour or the presence of blemishes and the extent to which 
defects show up. Sri Lanka has the purest form of latex crepe and if adverse comments 
have to be made on manufacture, it is with the intention of improving quality. It is to 
the credit of the guardians of our better managed estates battling against natural elements 
such as unpredictable weather patterns, difficulties in manufacture caused by bulking 
different varieties of latex and machinery breakdowns. Sri Lanka has yet been able to 
maintain a remarkable consistency and supremacy in the white crepes. 
Rubber could be sold forward in Sri Lanka for a period upto 6 months for Export 
Duty registration purposes. The quantity handled by the Colombo Brokers may vary 
between 8,000 to 9,000 m tons annually with the types sold being mainly crepe, scrap 
crepe, thick and/or thin crepe, sole crepe and technically specified rubber. Currently 
forward contracts are negotiated in a particular grade at a fixed price or a premium over 
prices obtained at the auctions for that particular grade. By selling forward taking into 
account various market factors such as production, weather and other influences, producers 
arc able to program their tasks more efficiently by ensuring the sale of quantities of rubber 
at attractive price levels for forward months which ensures both a profit as well as the 
smooth programming of the various necessities of the estates concerned. Brokers in keeping 
with market trends make daily efforts to elicit interest at advantageous levels and keep 
estates covered with forward contracts. An after sales service too is rendered both to the 
buyers and sellers when attempts are made to overcome problems connected with delayed 
deliveries and manufacture of quality rubber. By syphoning rubber against forward 
contracts arrivals at the public auctions could be curtailed thereby ensuring stable price 
levels. 
Currently a commodity which commands a premium over all other grades namely 
sole crepe is manufactured essentially against forward contracts. However production 
has remained stagnant and restricted to approximately 4,000 m tons annually. The belief 
that sole crepe has a good future was strengthened by the news that Malaysia, perhaps 
the largest producer of plantations sole crepe has curtailed their production. The feeling 
prevails in Sri Lanka that we could capture a share of the market. However market 
information should be obtained whether a genuine demand still prevails for Plantation's 
sole crepe. Furthermore the bulk of the sole crepe produced in Sri Lanka is the narrow 
gauge type measuring 13"X36" whilst only a handful of estates manufacture the broader 
585 
variety. For several estates that used to manufacture sole crepe upkeeping outdated 
machinery has been a real problem whilst recent high price for blanket crepe which 
narrowed the price differential between the two grades lead to abandoning this pursuit. 
Selling forward upto a month or two has been a real constraint. Producers have 
currently realised the value of having long term forward contracts, the lack of which 
poses problems to the manufacturer's who are compelled to find alternative employment 
for a large workforce engaged in the production of types such as sole crepe. The shippers, 
too will welcome this idea of obtaining forward contracts covering a 5 to 6 month period, 
since they can assure their clients overseas of a continued supply. Buyers overseas have 
a habit of restricting their purchase to certain sole crepe estates which have acquired a 
reputation for consistency in quality. However even lesser known sole crepe estates 
too could produce extremely good rubber and with sellers now engaged in rehabilitating 
and improving machinery and assisting the factories which abandoned this lucrative pur­
suit, manufacture is bound to increase whilst one can hope for consistency in the stan­
dards of manufacture. 
The inability of the seller to honour contracted obligations results in shipping commit­
ments of buyers getting disrupted. This leads to heavy penalty claims from overseas 
clients. Delay in deliveries is more often due to the location of rubber growing areas 
in the South West Zone of Sri Lanka which comes under the influence of unpredictable 
weather patterns. Buyers cannot be expected to carry buffer stocks to counter this 
problem due to high financing costs. 
Sales by private treaty are rare and confined mostly to withdrawn lots at the Public 
Auction and unmanufactured scrap rubber. 
Regular weekly visits are undertaken by the Brokers to both buyers and sellers offices 
where market information is exchanged. More often broker's advice has been invaluable 
to the estates when visits are undertaken in order to improve standards of manufacture 
in relation to trade requirements. When disputes arise between the buyers and sellers 
in matters connected with defaults and claims the Broker is called upon to arbitrate and 
often provides a speedy solution. 
When rubber is sold by Public Auction the Brokers ensure that the seller is paid the 
full value of the Invoice by the " PROMPT DAY " which means 7 working days 
after sale. Buyers too are expected to pay within this period. The rubber Broker is 
often the " Financer " advancing payments when produce is delivered to his godowns. 
Brokers participate in numerous meetings held by organisations such as the CRTA 
to discuss matters affecting the trade. Highlighting and drawing official attention to 
any shortcomings that prove detrimental to the trade. 
Market information is provided by the Broker to the entire rubber trade by means of 
weekly, monthly, quarterly and annual publications. The reports call for intensive 
and laborious collection of data. Heavy subscriptions are made to various publications 
required to keep the seller will advised of market conditions. 
586 
A little known fact of a broker's service is the information he is called upon to 
provide various individuals and organisations outside and within the country. These 
sources such as the World Bank in the USA look to the Brokers for impartial views 
and statistics. 
Brokers also perform many functions ancillary to the broking business such as the 
provisions of storage facilities for a consideration. Many broking firms have provided 
warehousing facilities to the trade when there was a dire need for good storage in the city 
thereby rendering an invaluable service. 
587 
SESSION 11. E C O N O M I C S O F R U B B E R 
NATURAL RUBBER: PAST PROBLEMS, FUTURE PROSPECTS 
By 
P . W . ALLEN 
(International Rubber Research and Development Board, U.K.) 
For the year in which natural rubber (NR) research started in Sri Lanka, world N R 
consumption was 90,000 tonnes. Since that time it has grown to just under 4 million 
tonnes, the average annual growth rate over these 75 years being 5 per cent. This is super­
ficially a record of steady if unspectacular growth, over a long period which has seen 
two world wars and a number of sharp reversals in the world economy. 
This picture of steady growth ignores, of course, the impact of the synthetic rubbers 
(SR). Unknown in 1909, save as chemical curiosities, world production and consump­
tion of SR did not start on a serious scale until the early 1940s, and SR consumption 
overtook that of N R in 1960. Today, N R constitutes only some 32 percent of the " elas­
tomer " (NR + SR) market. Evidently N R has lost ground in the sense that its pro­
duction has not been able to match the growing demand for elastomers. The question 
we have to ask is where, roughly, will N R be by the end of the century ? Will its market 
share be about where it is now ? . . . or will it have fallen even lower ? . . . or is there a 
possibility that it could increase ? 
The 1970s 
A spectator of the NR/SR scene during the late 1960s might have concluded that 
NR's days were numbered — and indeed this view was held by not a few. By 1970, 
NR's share of the world elastomer market had fallen from 100 percent pre-1939 to under 
30 percent, in virtually a straight line. Given this, given various technical improvements 
with SR, and given also the great economic attraction of low-cost SR from low-cost oil, 
it would not have been difficult to conclude that N R was doomed to steady extinction 
— or, to be less pessimistic, that its share of the elastomer market would fall to some 
very low residual value representing the minimal usage of N R in a few applicational 
areas for which SR was technically unsuited. 
The " o i l shock" of 1973 — and repeat performances since then — changed the 
picture. The view taken by many commentators at that time was that escalation in the 
price of oil was a " bonus " for NR, in that it would greatly improve the competitiveness 
of NR. Was this a correct view ? In one sense, yes : the cost-dependency of N R on 
energy is only some 10 percent of that of SR, and the astronomical increases in energy 
costs since 1973 have of course considerably increased the production costs of SR vis a 
vis NR. 
This, though, is only half the picture. The 1973 and 1979 oil price increases — and 
the subsequent series of shock waves through the world economy — had a major effect 
on the demand side, an effect which can be very simply illustrated. Over the decade 
589 
of the 1960s, world elastomer consumption grew at almost 7 percent/year, and had in 
fact done so since shortly after the end of the 1939 — 45 war. Since 1974 to the present 
time, the annual growth rate has been little more than 1 percent: external events during 
the 1970s virtually destroyed growth in the world rubber industry. 
These events have harmed N R far more by their adverse effect on consumption than 
they have helped by improving NR's production cost competitiveness. NR, like SR, 
has had to struggle to exist in a world of declining markets. The very sluggish growth 
in the world elastomer market to which reference has just been made had the effect on 
N R that consumption actually fell for 3 years in succession over 1979 — 1982, and 
there was a price drop of 1000 Ml/tonne over 1980 —1982. 
Potential for the future 
In fact, as far as market share is concerned, there has been some evidence for an 
improvement in health in recent years : NR's market share has moved up from a low of 
29 8 percent in 1978 to almost 33 percent today. This, though, is largely a response 
to the very low N R price over much of this period, and this short-term trend could reverse 
if the N R price were to rise substantially over that of general-purpose SR. Wc need to 
look for more substantive signs of permanent health. 
Such a sign can be gleaned from a study carried out in the early 1970s by the Malay­
sian Rubber Research and Development Board on behalf of the ANRPC, to assess the 
potential market for NR. The purpose of this study was to try to assess what level of 
share would accrue to N R if could it compete on level economic terms with SR (and 
provided, of course, that N R production could be expanded). The outcome of this 
study was the " techno-economic norm " : a sensible estimate of the market share that 
N R might aim to achieve. This norm was at that time calculated to be 43 percent for 
the world (excluding the Communist-bloc countries), the upper limit being 50 percent. 
The minimum N R market share was estimated to be around 30 percent. 
The conclusions of this earlier study are reinforced by what now appears to be the 
pragmatic view of the rubber industry, that the elastomer market divides up on technical 
grounds into 40 percent for NR, 40 percent for general-purpose SR, 10 percent for spe­
cial-purpose SR, the remaining 10 percent being open to competition on the basis of 
price. Disregarding the precise percentages, the point to be made is that the decline in 
NR's share of the market in recent decades is reversable ; there is a lot of life left in NR, 
and its full potential has yet to be realised. 
Several contemporary features of the world elastomer scene, in fact, reinforce in a 
very tangible way the supposition that NR, far from being a commodity of diminishing 
value, continues to be of vital importance. Particularly significant is growing concern 
among major consuming regions over the security of the future supply of NR. This is 
evidonced by (1) moves in the USA to increase the US strategic stockpile of NR to 850,000 
tonnes, (2) continued interest in the potential of Guayule to provide another source of 
natural polyisoprene, and (3) the long-term intention of the USSR to develop a very large 
productive capacity of synthetic polyisoprene. 
590 
If N R is technically a " has been " then these situations would not exist. The USA, 
for example, with its prodigious technical and financial skills, would surely not be at all 
interested in maintaining large domestic stocks of N R if it were a simple matter to use 
some other elastomer. Current interest in Guayule (mis-placed though some of us in 
the Hevea business think this to be) would not exist unless there is the belief that poly-
isoprene rubber is a much needed commodity. The USSR approach further underlines 
the correctness of this view, even though their choice (for reasons that need not concern 
us here) lies with the synthetic version of NR, a choice which is certainly ruled out in the 
West by economic factors. The world, then, shows every sign that it will continue to 
want polyisoprene rubber. Since Hevea is without doubt the best all-round producer 
of polyisoprene (disregarding special situations) the prognosis for our industry can only 
be regarded as good. 
Possibilities for 2000 A D 
In order to arrive at some idea as to the likely N R share of the world elastomer (NR + 
SR) market at some future date it is necessary first to assess future elastomer demand. 
This is no easy matter, and it has been made much more difficult by the prolonged slump 
in demand since 1973 to which reference has been made. For this reason, forecasts made 
by various expert bodies over recent years have given progressively lower estimates for 
1990 and 2000 as each year passes. 
Forecasts produced since 1979 by the IRSG, the IISRP and others (including a com­
prehensive econometric study by the Dutch Economist, H. P. Smit), have given estimates 
for the year 2000 ranging from 14-3 to 21 1 million tonnes. Very recently, the " Task 
Force " appointed in 1983 by the Minister for Primary Industries in Malaysia has come 
up with the figure of 18 million tonnes, with a probable error of + / - 3 million tonnes. 
This can be regarded as a sensible, middle-of-the-road, benchmark. 
If the N R producers were to aim to increase the N R market share from its present 
level (32 percent) to say 40 percent — and, as indicated, this is feasible on techno-economic 
grounds, then N R production would need to be over 7 million tonnes/year by 2000. The 
present (1984) production rate is about 4 million tonnes/year, and it is inconceivable that 
production could be increased by 3 million tonnes/year in the intervening 16 years. 
The Malaysian Task Force suggests that the demand for N R in 2000 will be 6 million 
tonnes/year. This figure is reinforced by the econometric study just referred to, which 
includes analysis of N R production, using a method whereby allowance is made for the 
age and type of planted trees and hence of their future production potential ; this gives 
6 15 million tonnes/year for N R production in 2000. 
Accepting all the hazards of forecasting, all that can be said is that the consensus 
view of experts at this time is that, for the year 2000, world elastomer consumption will 
be about 18 million tonnes and N R production about 6 million tonnes. On this basis, 
therefore, NR's share of the market is expected to be 33 percent, distinctly higher than 
the ' l o w ' of 1978 and somewhat higher than the present level. 
591 
The outcome is quietly encouraging. Production constraints make it impossible 
that N R should rapidly regain a very much larger market share ; on the other hand, the 
various forecasts do all indicate that the share will certainly not diminish but will stabilize, 
moving up just a little. This reinforces the view put forward earlier, that N R remains 
an important and valuable commodity, and there is no case for pessimism as regards 
market share. 
Requirements for growth : the production scene 
To achieve the production level just indicated — 6 million tonnes/year by 2000 — 
will require careful and concerted effort on the part of the producers. Also, it must be 
said — very firmly — that producers are not so much interested in tonnages and market 
shares as in value (price times volume). For this reason, attention needs to be devoted 
to the nature and composition of N R consumption which — it will be shown — has an 
important bearing on price. 
On the production front, the " name of the game " will be production efficiency. 
In most producing areas, opportunities for turning over new land to Hevea are becoming 
increasingly limited ; there are other uses for land (other crops, industrial uses, housing) 
and the days when production of N R could be boosted by planting trees on newly-cleared 
jungle land at low levels of productivity are gone for ever. There is also a growing shortage 
of labour. The need therefore is to extract the maximum quantity of N R from existing 
areas (or even diminished areas) with no more (and if possible, less) manpower than that 
presently used. 
Two observations are pertinent. The first is that, despite the many years' work on 
tree breeding since the inception of the Institute in Sri Lanka, average yields around the 
world — as low as 200 — 300 kg/ha/year in some areas — are far below the best achieved 
in some areas (over 1000 kg/ha/year) and even these are less than the potential. The 
second is, to quote from the Malaysian Task Force Report, that " . . .the tapping and 
collection practices commonly practised today with the exception of the use of yield sti­
mulants are more or less those of 50 years ago ". 
As far as trees already in the ground are concerned, the need is to optimize produc­
tivity by making use of well-known and proven methods. There is no doubt that if the 
full potential of these trees can be realised, future production could meet future demands 
at the level postulated. For example, reduction in the frequency of tapping (from d/2 
to d/3) plus use of yield stimulation can, according to recent Malaysian experience, reduce 
the labour requirement by some 40 percent while maintaining or increasing the yield. 
Also, further development of mechanized tapping and collection methods is called for. 
As far as future of replanting is concerned, an imaginative approach is called for, 
coupled with full use of existing know-how. There is need to make much wider use of 
techniques such as those which reduce the period of immaturity, precocious yielding 
clones, three-part trees, optimal planting densities, and so on. All of these are well-
known to the industry. The problem seems to be that knowledge of all such methods 
is not matched by its application. 
592 
For the longer term, there is of course the need to extend the potential of Hevea 
by breeding research (with, for example, the 1981 material collected by the IRRDB in 
Brazil and now being distributed) and allied activities (e.g. in vitro cultivation techniques). 
For breeding, the target should be, not merely productivity, but adaptability to special 
conditions, and resistance to diseases (especially SALB). 
Overall, the plantation industry needs to take a wider view of its purpose. Instead 
of regarding itself as a producer of NR, it should plant Hevea with the intention of using 
the (reduced) immaturity period for the growing of cash crops, it should collect and use 
seeds during the 20-year life cycle of a tree for cattle feed, and it should use the tree at 
the end of its life as timber. Such an integrated approach will not only enable N R 
production to expand as envisaged, it will also greatly improve the socio-economic con­
tribution of the industry to its country. The technologies to do all of this exist, the need 
is to employ them. 
In general, the position appears to be that, despite the existence of a wide range of 
methods for improving production efficiency, these methods are not being implemented 
as fully or as widely a9 is desirable. Is there a "communications gap"? If so, the 
IRRDB is willing and able to help fill it. 
Requirements for growth : the consumption scene 
In this section, attention is focussed on patterns of N R consumption, and on the 
influence that these patterns have on the economic status of NR. 
As far as the N R producers are concerned, the influence of consumption on their 
well being is more diffuse and less obvious than is the influence of production changes. 
In respect of production, the producers are obviously their own masters, but they cannot 
dictate in what ways their N R should be used. All that they can do is to seek to persuade 
and encourage, by providing appropriate information to consumers. 
Tyres are the dominant sector of N R consumption. Over the past 20 years, the 
tyre industry's share of N R consumption has risen from under 60 percent to 73 percent 
(for the major industrialized countries taken together) ; this is a result of (i) conversion 
from cross-ply to radial-ply construction, and (ii) the increasing production of truck 
tyres vis a vis car tyres. There is no doubt that this trend will continue. The Malaysian 
Task Force suggests that, by 2000, tyres will require an additional 2 million tonnes/year 
of N R compared with 1982 ; this is about 80 percent of the total postulated increase in 
N R consumption over this period. 
It may well seem that this situation — whereby one consuming sector virtually " gua­
rantees " taking most of the projected production increase — is very satisfactory. To 
an extent this is so, but there is another feature. 
As between the tyre and " non-tyre " sectors of the NR market there is an important 
difference. NR/SR compositional requirements for tyres are not very responsive to price 
changes. This is not true in the non-tyre sector, where there are many " technically 
undemanding " applications for which either N R or SR can be used, according to price. 
(" Technical" applies to processing requirements as well as to service performance). 
593 
The dominance of the tyre sector ensures that its consumption largely determines 
the N R price, other things being equal. When there is a surge in demand for tyres, the 
tyre manufacturers have to find the extra N R either from world stocks or by (as it were) 
" stealing " some N R from the non-tyre sector ; this cannot come from extra N R pro­
duction because this is unable to be increased in the short term. The NR price will rise, 
and this will encourage consumers in the non-tyre sector to switch from NR to SR. Un­
fortunately — and this is- the important point — when tyre demand slackens, there is 
evidence that the reverse process (switching back from SR to NR.) in the non-tyre sector 
is not so strong. Thus, the existence of " technically undemanding " applications in 
the non-tyre sector has the effect of restraining the price rises which would otherwise 
occur as a result of booms in the tyre industry. 
As regards N R consumption, then, the requirements for healthy, growth are : 
(1) to continue to foster usage of NR in tyres by means of appropriate R & D and 
promotional activities, on the basis that it would be absurd to neglect or ignore this domi­
nant sector. It is especially important to be aware of possibilities for major substitution 
away from N R in tyres, and to carry out appropriate work to guard against this. 
(2) to develop a portfolio of non-tyre applications which are technically demanding, 
so as to strengthen this sector of the market. 
Concerning (2), the need is not only to establish new applications for N R which unique­
ly require it, but also to encourage use of added-value forms of NR.The former includes, 
for example, high-durability engineering components ; the use of NR in these is already 
respectable (some 100,000 tonnes/year) and needs to be increased. The latter includes 
materials such as thermoplastic, powdered and liquid forms of NR. Production and 
usage of these materials will help to strengthen the NR market, and it is especially desirable 
that their production should be concentrated in the NR producing countries. 
On this last point, it is important to appreciate the scale and significance of the changes 
taking place in the geography of NR consumption. Over the past 20 years, 50 percent 
of the increase in NR consumption has taken place in the developing countries, including 
those that produce NR. Of particular significance is the fact that consumption within 
the producing countries (about 8 percent today) is rising fast. This trend, in addition 
to serving the purposes of the countries themselves, is an important factor which serves 
to strengthen the market for NR. 
Final points C 
The long-term outlook for NR is quietly promising ; there is certainly no cause for 
pessimism, despite the problems of recent years. NR producers can, if all the expert 
prognostications are correct, expect to see annual consumption rise from the present 
level of 4 million tonnes/year to about 6 million tonnes/year by A D 2000, and this will 
maintain NR's share of the world elastomer market at slightly above the present level. 
594 
To achieve this production increase in a manner which will satisfy the national and 
individual needs of the producers will require strenuous efforts to improve production 
efficiency. It will also require the fostering of new outlets for N R in non-tyre applications 
so as to strengthen the market. Both aspects will require considerable inputs of R & D . 
Of course, no one can guarantee that there will not be problems ; for example, another 
round of world depression. Short term problems (let us hope that if they occur, they 
will be short-lived) affecting the N R price can of course be coped with via the INRO 
price stabilization operation. This latter, plus the longer term measures advocated in 
this paper, should together enable the N R producers to face the future with confidence. 
REFERENCES 
Opinions expressed in this paper are those of the author, and do not necessarily reflect 
the views of Member Institutes of the IRRDB. 
In preparing the paper, use has been made of the following sources : 
ALLEN, P . W . , THOMAS, P . O. AND SEKHAR, B. C . A study on competition between natural 
and synthetic rubber. MRRDB, 1974. 
ALLEN, P . W . The changing pattern of natural rubber consumption and its impact 
on the market. International Rubber Marketing Conference, Kuala Lumpur, October 
1983. 
Malaysian Natural Rubber Indutrsy 1983 — 2000. Report of the Task Force of Experts. 
MRRDB, 1983. 
SEKHAR, B. C. The natural rubber industry in the year 2000. Third International 
Rubber and Plastics Conference for Asia, Singapore, March 1984. 
SMIT, H. P. The world rubber economy to the year 2000, Amsterdam, Kanters B V 1982. 
YOUNG, C. How are natural and synthetic rubber prices related ? 95th Group Meeting 
of the IRSG, London, June 1983. 
595 
NATURAL RUBBER SUPPLY IN SRI LANKA: ANALYSIS 
AND PROJECTIONS USING VINTAGE APPROACH 
By 
H. P. SMIT 
(Free University, Netherlands) 
General information 
Sri Lanka produced about 100,000 tonnes of natural rubber (NR) until the middle 
of the 1960's (cf. Fig. 1.1). Sri Lanka was the fourth largest natural rubber producing 
country with 4 to 5 percent of world N R production. However, it lost its fourth place 
to India in 1980 while its N R production declined from 152-7 thousand tonnes in 1979 
to 133*2 thousand tonnes in 1980 (see Table 1.1) or 3*5 percent of world N R production, 
with even lower levels in 1981 and 1982. What are the reasons behind these developments 
and what will be the future ? That is the subject of this paper. 
Table 1.1. Production of NR in Sri Lanka (1,000 tonnes) 
Year Production 
1955 95-3 
1956 96-9 
1957 99-7 
1958 101.8 
1959 93-2 
1960 98-8 
1961 97-6 
1962 104-1 
1963 104-8 
1964 111-6 
1965 118-3 
1966 131-0 
1967 143-2 
1968 148-7 
1969 150-8 
1970 159-2 
1971 141-4 
1972 140-4 
1973 154-7 
1974 132-0 
1975 148-8 
1976 152-1 
1977 146-2 
1978 1 5 5 7 
1979 152-7 
1980 133-2 
1981 123-9 
1982 125-2 
1983 139-9 
597 

Area under rubber has been rather stable, as can be concluded from data from the 
office of the Rubber Controller of Sri Lanka. In 1945 officially reported area was about 
265,000 ha increasing to 270,000 ha in 1960. However, the agricultural census of 1962 
revealed that about 42,000 ha of reported rubber land was no longer under rubber. On 
the one hand rubber land has been neglected and has virtually turned into jungle. On 
the other hand, there is constant pressure on rubber land from urbanization and infra­
structure such as roads. In 1980 there was about 227,000 ha of rubber land of which 
about three quarters is high-yielding. However, the Department of Census and Statistics 
reports an area of 194,000 ha in 1980. In accordance with an often practised approach, 
we define a smallholding with 10 acres or less. The others will be called estates. The 
estate sector then comprises slightly less than 70 percent of total rubber area. 
Replanting and new planting have been marginal in the 1970's, partly due to uncer­
tainties regarding announced land reform programs in 1972 and 1975. In 1953 a govern­
ment subsidized replanting scheme was started. It was very successful in the beginning, 
but response dropped dramatically in later stages. In the 1970's the average was slightly 
over 3,000 ha per year, that is about 1 - 5 percent of total area and therefore far below 
3 percent required to maintain a properly composed rubber area with a maximum age 
of 33 years which is close to optimal. Replanting reached 5,400 ha in 1980 and is planned 
to increase substantially in 1985, to some 5 percent of total area, thus compensating for 
the lack of replanting in the past. After 1986 replanting is aimed at 8,000 ha annually. 
New planting is very limited e.g. replacing uneconomic coconut plantings. The level of 
new planting in the 1970's was in the order of 200 ha on average per year. Efforts to 
increase new planting in the 1980's have had some success already with 581 and 977 ha in 
1979 and 1980 respectively increasing to 1650 ha in 1982. Plans call for a level of around 
900 ha per year in the first half of the 1980's and 200 ha annually in the second half. 
Production has increased from around 100,000 tonnes in the fifties and early sixties 
to around 150,000 tonnes in the seventies. However, production only reached 133,000 
tonnes in 1980 and dropped further to around 125,000 tonnes in 1981 and 1982. Next 
to the low price of N R and unfavourable weather conditions, a major reason behind this 
development is the decrease in tappable area from about 190,000 ha in 1979 to around 
150,000 ha in 1985 if the replanting plans are realized. This means that a level of pro­
duction of about 150,000 tonnes may continue up to 1987. Afterwards there will be 
an increase. With some 45 percent of rubber area (to be) replanted in the 1970's and 
1980's and the other 55 percent to show a further reduction in yield, one may expect a 
substantial increase in average yield by the year 2000. How much N R output may be 
oxpected in the years ahead and what will bo the implications of certain planting policies 
on N R supply ? These questions will be answered in the following sections. 
A vintage approach for production analysis 
The need for a vintage approach 
Rubber production on a commercial basis, started in Sri Lanka in the beginning of 
this century. The area increased rapidly until the 1940's and decreased slightly after­
wards. In analyzing the relationship between production capacity of NR and area, it 
is essential to include a few aspects. 
599 
(a) Trees have an immaturity period of 5 — 8 years, reaching top levels of produc­
tions after a few years of tapping and becoming less productive afterwards. 
(b) Improvement in the quality of the trees leads to higher productivity for trees 
of vintages planted later. A tree planted in 1910 provided less rubber than 
a tree of the 1930-vintage. 
(c) A steep increase in productivity was achieved by the introduction of high-yield­
ing trees from the late 1930's onwards. Total rubber area in the past decades 
therefore consisted of high-yielding trees and ordinary trees strongly influencing 
rubber output. This is even more true since the ordinary trees, because of 
slaughter tapping during World War II, have seen their potential yields reduce 
drastically. 
(d) A further reason for including the year of planting in the analysis of the rela­
tionship between production and area, is that replanting has been lagging 
behind and that replanting schemes have not been effective before 1954 and 
in most of the late sixties and seventies. The strong increase in production 
in the 1960's was caused by the drastic replanting in 1950's. In the 1970's 
production stagnated and declined in 1980 amongst others because of successful 
replanting efforts. 
(e) Because of differences in productivity and planting, estates and smallholdings 
need to be distinguished. 
Planting data for estates and smallholdings 
Area data on natural rubber are shown in Table 2.1, columns (1), (2) and (3). Data 
do not always represent actual rubber area, as was e.g. revealed by the 1962 census. Up­
rooted area may still be registered as rubber area. Between 1945 and 1962 data have 
been adjusted proportionately in the analysis so as to eliminate a sudden drop in 1962. 
This has also been done for the data between 1963 and 1983, since the Department of 
Census and Statistics gives a figure of 194,000 hectares for 1980. Wherever aggregate 
data or the split-up between estates and smallholdings were not available, linear interpola­
tion has been applied. According to available data (Table 2.1), the smallholder sector 
comprises 30-5 % of total area in 1975. The Department of Census and Statistics in 
its recent " Highland crop survey " reports a percentage of 43. For this reason we have 
gradually increased the share of smallholder area from 1976 onwards. The results are 
shown in Table 2.2 and Fig. 2.1. 
Replanting data (cf. Table 2.1) for 1976— 1983 have been disaggregated using the 
division as applicable for 1975. Disaggregated data on new planting are constructed by 
using the shares, estates and smallholdings had in replanting. The sum of replanting 
and new planting is the area with new trees. Results are shown in Table 2.3 and Fig. 2.2. 
600 
Table 2.1. Area and planting of natural rubber (ha) 
Area Replanting New 
Small­ Small­ planting 
Year Estates holdings total Estates holdings total total 
(2) (2) (3) (1) (2) (3) 
1901 800 — 800 
1911 59,900 15,000 74,900 
1921 126,400 31,600 158,000 
1931 182,700 45,800 228,500 
1941 258,600 
1945 267,220 38 — 38 1,104 
1946 205,300 61,600 266,900 666 — 666 613 
1947 266,880 858 — 858 183 
1948 266,631 592 — 592 119 
1949 265,319 890 — 890 233 
1950 265,366 1,566 — 1,566 272 
1951 265,300 1,388 — 1,388 570 
1952 266,036 1,701 — 1,701 553 
1953 195,200 71,300 266,500 2,161 188 2,349 349 
1954 266,980 5,886 1,589 7,475 619 
1955 267,699 7,006 2,271 9,277 502 
1956 266,995 7,215 2,742 9,957 174 
1957 267,594 7,055 2,758 9,813 1,131 
1958 269,259 5,707 2,652 8,359 936 
1959 193,200 77,400 270,600 4,354 3,151 7,505 814 
1960 270,924 4,334 2,912 7,246 767 
1961 271,686 4,760 2,808 7,568 572 
1962 229,765 4,440 2,831 7,271 277 
1963 230,033 3,955 2,490 6,445 268 
1964 230,204 3,502 1,986 5,488 171 
1965 230,465 3,107 1,957 5.064 261 
1966 230,622 3,134 1,961 5,095 157 
1967 230,495 2,748 1,337 4,085 56 
1968 160,948 69,581 230,529 3,424 1,734 5,158 237 
1969 163,144 67,511 230,655 3,229 1,665 4,894 126 
1970 230,259 2,716 1,431 4,147 112 
1971 230,038 2,226 1,208 3,434 238 
1972 229,636 2,589 952 3,541 180 
1973 228,581 2,016 931 2,947 186 
1974 158,491 69,688 228,007 1,879 987 2,866 35 
1975 158,851 69,625 227,638 2,072 1,160 3,232 155 
1976 226,981 2,551 56 
1977 226,568 2,617 34 
1978 226,328 3,226 380 
1979 226,604 4,168 581 
1980 227,240 5,436 977 
1981 205,609 6,441 710 
1982 205,690 6,782 800 
Source : Administrative Reports of the Rubber Controller (various issues) 
Association of Natural Rubber Producing Countries (1976) 
John Keels' Rubber Statistics (various issues) 
Rubber Trends, Economic Intelligence Unit (1979) 
601 
Table 2.2. Estimated data on natural rubber area (ha) 
Area of Area of Total Year Area of Area of Total 
Year estates smallholdings area estates smallholdings area 
1901 800 0 800 1945 203,793 60,547 264,340 
1902 6,710 0 6,710 1946 203,425 61,037 264,462 
1903 12,620 0 12,620 1947 200,167 61,846 262,013 
1904 18,530 0 ' 18,530 1948 196,754 62,571 259,325 
1905 24,440 0 24,440 1949 192,606 63,020 255,625 
1906 30,350 2,500 32,850 1950 189,484 63,763 253,247 
1907 36,260 5,000 41,260 1951 186,307 64,454 250,761 
1908 42,170 7,500 49,670 1952 183,710 65,316 249,026 
1909 48,080 10,000 58,080 1953 180,937 66,090 247,027 
1910 53,990 12,500 66,490 1954 179,394 65,639 245,033 
1911 59,900 15,000 74,900 1955 178,005 65,242 243,248 
1912 66,550 16,660 83,210 1956 175,672 64,497 240,169 
1913 73,200 18,320 91,520 1957 174,198 64,066 238,264 
1914 79,850 19,980 99,830 1958 173,403 63,884 237,287 
1915 86,500 21,640 108,140 1959 172,381 63,617 235,997 
1916 93,150 23,300 116,450 1960 170,038 63,767 233,805 
1917 99,800 24,960 124,760 1961 167,955 64,026 231,981 
1918 106,450 26,620 133,070 1962 165,578 64,187 229,765 
1919 113,100 28,280 141,380 1963 163,555 64,498 228,054 
1920 119,750 29,940 149,690 1964 161,449 64,793 226,242 
1921 126,400 31,600 158,000 1965 159,391 65,125 224,516 
1922 132,030 33,020 165,050 1966 157,245 65,439 222,684 
1923 137,660 34,440 172,100 1967 154,894 65,685 220,578 
1924 143,290 35,860 179,150 1968 152,639 65,989 218,627 
1925 148,920 37,280 186,200 1969 153,318 63,445 216,762 
1926 154,550 38,700 193,250 1970 151,112 63,196 214,409 
1927 160,180 40,120 200,300 1971 149,035 63,289 212,224 
1928 165,810 41,540 207,350 1972 146,853 63,042 209,877 
1929 171,440 42,960 214,400 1973 144,274 62,672 206,946 
1930 177,070 44,380 221,450 1974 142,019 62,445 204,464 
1931 182,700 45,800 228,500 1975 140,565 61,610 202,175 
1932 184,207 46,853 231,060 1976 137,750 61,888 199,638 
1933 185,713 47,907 233,620 1977 134,181 63,144 197,325 
1934 187,220 48,960 236,180 1978 130,763 64,406 195,169 
1935 188,727 50,013 238,740 1979 127,682 65,775 193,457 
1936 190,233 51,067 241,300 1980 126,100 67,900 194,000 
1937 191,740 52,120 243,860 1981 124,160 69,840 194,000 
1938 193,247 53,173 246,420 1982 122,220 71,780 194,000 
1939 194,753 54,227 248,980 1983 120,280 73,720 194,000 
1940 196,260 55,280 251,540 
1941 197,767 56,333 254,100 
1942 199,273 57,387 256,660 
1943 200,780 58,440 259,220 
1944 202,287 59,493 261,780 
60g 
Table 2.3. Planting data on natural rubber (ha) 
Replanting New planting New trees 
Year Estates Smallholdings Total Estates Smallholdings Total Estates Smallholdings 
1951 1248 140 
1952 1541 160 
1953 2161 188 
1954 5886 1589 
1955 7006 2271 
.1956 7215 2742 
1957 7055 2758 
1958 5707 2652 
1959 4354 3151 
1960 4334 2912 
1961 4760 2808 
1962 4440 2831 
1963 3955 2490 
1964 3502 1986 
1965 3107 1957 
1966 3134 1961 
1967 2748 1337 
1968 3424 1734 
1969 3229 1665 
1970 2716 1431 
1971 2226 1208 
1972 2589 952 
1973 2016 931 
1974 1379 987 
1975 2072 1160 
1976 1635 916 
1977 1678 939 
1978 2068 1158 
1979 2672 1496 
1980 3936 1500 
1981 4941 1500 
1982 5282 1500 
1983 5500 1500 
Discarding systems and vintages 
Reasonably adequate data on area distribution by vintages, i.e. the number of hectares 
planted in year t and still existing in year t is not available for a proper period of time 
which might act as the period of analysis e.g. t = 1951 . . . , 1983. Of course these data 
might be calculated if the distribution in the starting year 1951 was known and if dis­
carding of trees was given by vintage. Unfortunately, neither is available. 
The method adopted in this analysis, is to use aggregate data on new planting, re­
planting and total area, and to derive mathematically formulated discarding systems, 
which accurately relate discarding of area to new planting and replanting in the past, 
resulting in a distribution of the area by age. Underlying this analysis is the argument 
that if one would know in year t what the area per vintage t is, say at t' and if one would 
also know which percentage according to a theoretical discarding system would be dis­
carded for each vintage t depending on the age of the vintage, t - 1 , then one could calculate 
the area per vintage as well as the total discarded area. However, the parameters of the 
discarding system are not kown. Now turning around the argument : given data on a 
distribution by vintages in a year, total discarding of area in the following year may act 
as a criterion to select among theoretical mathematically formulated discarding systems. 
Afterwards, discards of trees can be allocated to vintages, using the selected discarding 
1388 513 57 570 1761 197 
1701 501 52 553 2042 212 
2349 321 28 349 2482 216 
7475 487 132 619 6373 1721 
9277 379 123 502 7385 2394 
9957 126 48 174 7341 2790 
9813 813 318 1131 7868 3076 
8359 639 297 936 6346 2949 
7505 472 342 814 4826 3493 
7246 459 308 767 4793 3220 
7568 360 212 572 5120 3020 
7271 169 108 277 4609 2939 
6445 164 104 268 4119 2594 
5488 109 62 171 3611 2043 
5064 160 101 261 3267 2058 
5095 97 60 157 3231 2021 
4085 38 18 56 2786 1355 
5158 157 80 237 3581 1814 
4894 83 43 126 3312 1708 
4147 73 39 112 2789 1470 
3434 154 84 238 2380 1292 
3541 132 48 180 2721 1000 
2947 127 59 186 2143 990 
2866 23 12 35 1902 999 
3232 99 56 155 2171 1216 
2551 36 20 56 1671 936 
2617 22 12 34 1700 951 
3226 244 136 380 2312 1294 
4168 372 209 581 3045 1704 
5436 707 270 977 4643 1770 
6441 817 248 1065 5758 1748 
6782 1285 365 1650 6567 1865 
7000 876 214 1000 6286 1714 
€03 


Table 2.4. Area distribution for estates (ha) 
Year of Year of tapping 
planting 
1955 1956 1957 1958 1959 1960 1961 1962 1963 1964 1965 1966 1967 1968 1969 1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982 1983 
1901 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 
1902 3 2 1 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 
1903 7 4 2 1 I 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 
1904 14 8 5 3 2 I 1 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 
1905 26 15 9 6 4 3 2 1 1 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 
1906 44 25 14 9 7 4 3 2 1 1 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 
1907 68 40 22 15 10 7 4 3 2 1 1 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 
1908 103 59 34 22 15 10 6 4 3 2 2 1 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 
1909 150 87 49 32 23 15 9 6 4 3 2 2 1 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 
1910 214 124 70 45 32 21 14 9 6 4 3 2 2 1 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 
1911 . 298 173 98 63 45 29 19 12 9 6 5 3 3 2 2 1 0 0 0 0 0 0 0 0 0 0 0 0 0 
1912 454 263 149 96 68 44 29 19 13 9 7 5 5 3 2 2 1 0 0 0 0 0 0 0 0 0 0 0 0 
1913 608 352 199 129 91 59 38 25 17 13 9 7 4 4 3 «- 2 1 0 0 0 0 0 0 0 0 0 0 0 
1914 831 481 272 176 125 81 52 34 24 17 13 10 7 5 4 3 3 2 2 0 0 0 0 0 0 0 0 0 0 
1915 1173 679 384 248 176 114 74 48 34 24 18 13 10 7 6 5 4 3 2 2 0 0 0 0 0 0 0 0 0 
1916 1570 1003 567 367 260 169 109 71 50 36 27 20 15 11 9 7 6 4 3 3 2 0 0 0 0 0 0 0 0 
1917 2001 1391 872 564 399 260 168 110 76 55 41 31 23 17 14 11 9 7 5 4 3 3 0 0 0 0 0 0 0 
1918 2453 1822 1249 870 616 400 259 169 118 85 63 47 36 26 22 17 13 10 8 6 5 4 3 0 0 0 0 0 0 
1919 2916 2283 1677 1248 935 608 393 257 179 129 96 71 54 39 33 25 20 15 12 9 8 6 5 4 0 0 0 0 0 
1920 3381 2762 2143 1680 1325 927 600 392 273 197 146 109 83 60 51 39 30 23 18 14 12 9 7 6 4 0 0 0 0 
1921 3843 3249 2635 2155 1770 1321 921 601 419 302 224 167 127 92 78 59 47 36 27 22 18 14 11 8 6 5 0 0 0 
1922 3780 3287 2763 2339 1986 1561 1162 815 568 409 304 - 227 172 125 105 80 63 48 37 30 24 19 15 11 8 6 4 0 0 
1923 4257 3789 3279 2854 2491 2042 1600 1196 884 637 473 353 268 195 164 125 98 75 58 46 38 30 24 18 13 10 6 3 0 
1924 4741 4301 3813 3396 3032 2570 2100 1652 1287 977 726 541 411 299 251 192 151 115 89 71 58 46 36 27 20 15 9 5 2 
1925 5216 4808 4347 3946 3589 3128 2644 2168 1761 1402 1092 814 618 449 377 288 227 173 133 106 87 69 55 41 30 22 14 7 3 
1926 5732 5353 4921 4538 4191 3737 3248 2753 2315 1915 1555 1214 921 670 563 430 338 259 199 158 130 104 81 62 45 33 21 1 ] 5 
1927 6261 5912 5509 5147 4816 4376 3891 3389 2931 2501 2103 1712 1356 986 828 632 497 381 293 233 191 152 120 91 66 48 30 15 7 
1928 6830 6509 6135 5795 5480 5057 4582 4082 3614 3162 2735 2303 1897 1451 1219 931 732 560 431 343 282 224 176 133 97 71 44 23 10 
1929 7419 7125 6779 6462 6166 5765 5305 4815 4345 3883 3437 2976 2531 2027 1748 1335 1050 804 618 491 404 322 253 191 139 102 64 33 14 
1930 7943 7677 7363 7073 6795 6426 5992 5522 5062 4604 4153 3680 3212 2670 2360 1881 1480 1132 871 692 570 454 356 269 196 143 90 46 20 
1931 8480 8241 7957 7692 7441 7096 6589 6244 5800 5352 4905 4428 3950 3384 3053 2530 2057 1581 1216 967 796 633 498 376 274 200 125 64 28 
1932 3309 3231 3136 3048 2965 2849 2710 2556 2401 2242 2082 1910 1734 1523 1397 1196 1011 807 621 493 406 323 254 192 140 102 64 33 14 
1933 3616 3544 3457 3375 3296 3188 3055 2908 2757 2600 2442 2269 2091 1874 1744 1531 1332 1105 886 704 580 461 363 274 200 146 91 47 20 
1934 3948 3881 3801 3725 3652 3550 3424 3284 3137 2984 2827 2656 2478 2259 2127 1908 1698 1454 1211 998 821 654 514 388 283 207 129 66 29 
1935 4301 4240 4166 4096 4028 3933 3814 3681 3539 3390 3237 3058 2592 2675 2542 2322 2105 1851 1590 1354 1148 914 719 543 396 289 181 93 40 
1936 4674 4618 4550 4485 4423 4335 4223 4096 3959 3815 3657 3502 3330 3116 2985 2767 2550 2290 2017 1764 !53S 1267 996 752 548 401 251 129 55 
1937 5060 5009 4947 4887 4830 4749 4655 4535 4403 4264 4120 3960 3792 3584 3457 3244 3030 2768 2490 2227 1986 1688 1370 1035 754 551 345 177 76 
1938 5450 5403 5347 5292 5240 5166 5080 4980 4854 4720 4582 4427 4265 4064 3943 3737 3529 3272 2995 2728 2479 2162 1807 1420 1034 756 473 243 104 
1939 5834 5792 5740 5691 5644 5577 5498 5407 5315 5187 5054 4905 4750 4557 4442 4247 4048 3799 3527 3262 3011 2683 2296 1869 1425 1041 652 334 144 
1940 6201 6163 6117 6073 6030 5969 5898 5816 5733 5644 5517 5375 5226 5043 4936 4753 4565 4328 4067 3809 3561 3230 2820 2363 1877 1429 895 459 197 
1941 6541 6508 6467 6427 6389 6335 6271 6198 6123 6043 5956 5821 5681 5508 5410 5241 5065 4844 4597 4351 4113 3784 3360 2883 2367 1872 1259 645 277 
1942 6849 6819 6783 6748 6714 6666 6610 6544 6478 6406 6329 6239 6107 5945 5856 5701 5540 5336 5107 4876 5651 4331 3899 3412 2879 2348 1682 968 416 
1943 7117 7090 7059 7028 6998 6956 6906 6849 6790 6726 6658 6579 6490 6340 6259 6120 5974 5789 5579 5366 5157 4852 4418 3930 3391 2837 2137 1356 682 
1944 7343 7320 7293 7266 7240 7203 7160 7109 7058 7002 6942 6872 6794 6691 6519 6495 6354 6197 6008 5815 5623 5336 4905 4423 3888 3321 2606 1784 1022 
1945 1508 1504 1499 1495 1490 1484 1476 1467 1458 1449 1438 1426 1412 1394 1382 1360 1337 1307 1273 1239 1204 1151 1067 973 868 755 612 445 281 
1946 1190 1187 1184 1181 1178 1173 1168 1162 1156 1150 1143 1135 1126 1113 1105 1090 1075 1054 1032 1008 984 946 883 813 735 648 540 412 280 
1947 959 957 955 953 951 948 944 940 936 932 927 921 915 907 901 891 880 866 850 834 817 790 741 687 628 560 477 379 274 
1948 610 608 607 606 605 603 601 599 597 595 592 589 586 581 578 572 567 559 550 541 532 517 487 455 419 378 328 268 204 
1949 990 987 985 983 982 980 977 974 971 968 964 960 956 949 945 937 929 919 907 895 882 859 813 763 709 644 568 477 376 
1950 1687 1683 1679 1676 1673 1670 1667 1663 1658 1653 1648 1642 1636 1627 1620 1609 1598 1583 1565 1547 1529 1493 1417 1337 1250 1144 1021 876 712 
.195! 1751 1747 1744 1740 1738 1735 1731 1728 1724 1720 1715 1710 1704 1696 1691 1681 1671 1657 1642 1626 1610 1576 1500 1421 1335 1229 1109 968 808 
1952 2033 2029 2024 2020 2017 2014 2010 2006 2003 1999 1994 1989 1983 1976 1970 1961 1951 1937 1922 1906 1890 1869 1784 1695 1600 1480 1347 1193 1017 
(953 2474 2468 2463 2458 2455 2450 2446 2441 2438 2433 2429 2424 2418 2410 2404 2394 2384 2370 2355 2338 2322 2300 2275 2168 2054 1908 1749 1569 1361 
1954 6362 6348 6334 6323 6314 6302 6291 6280 6270 6261 6251 6239 6227 6209 6197 6176 6153 6124 6089 6054 6017 5969 5916 5843 5553 5176 4774 4325 3807 
1955 7385 7370 7353 7340 7329 7316 7303 7290 7279 7268 7259 7248 7235 7218 7206 7185 7163 6134 7100 7065 7029 6982 6930 6857 6762 6322 5861 5355 4772 
1955 0 7341 7325 7312 7301 7288 7275 7262 7251 7241 7231 7222 7211 7197 7187 7169 7151 7126 7098 7068 703S 6998 6954 6893 6812 6723 6261 5761 5185 
1957 0 0 7868 7854 7843 7829 7815 7801 7789 7778 7768 7758 7748 7735 7726 7710 7693 7671 7645 7618 7591 7555 7514 7458 7385 7303 7162 6629 6018 
1955 0 0 0 6346 6337 6326 6314 6303 6294 6285 6277 6269 6261 6252 6246 6235 6223 6208 6191 6172 6153 6129 6101 6063 6012 5956 5859 5705 5216 
1959 0 0 0 0 4826 4818 4809 4801 4794 4787 4781 4775 4769 4762 4758 4751 4744 4734 4723 4711 4699 4683 4665 4640 4607 4571 4508 4408 4272 
i960 0 0 0 0 0 4793 4784 4776 4769 4762 4756 4750 4744 4738 4734 4728 4721 4713 4704 4694 4684 4670 4655 4634 4607 4576 4523 4439 4323 
1961 0 0 0 0 0 0 5120 5111 5103 5096 5090 5083 5077 5070 5066 5059 5054 5046 5038 5029 5020 5008 4994 4975 4950 4923 4874 4797 4692 
1962 0 0 0 0 0 0 0 4609 4602 4596 4590 4584 4579 4572 4569 4563 4558 4552 4546 4539 4532 1523 4512 4498 4479 4458 4421 4362 4281 
1963 0 0 0 0 0 0 0 0 4119 4114 4109 4103 4098 4093 4089 4084 4080 4075 4070 4065 4060 4053 4045 4034 4020 4004 3976 3931 3869 
1964 0 0 0 0 0 0 0 0 0 3611~ 3607 3602 3598 3593 3590 3586 3582 3577 3573 3569 3565 3560 3554 3546 3535 3524 3503 3470 3424 
1965 0 0 0 0 0 0 0 0 0 0 3267 3263 3259 3255 3252 3248 3245 3241 3237 3233 3230 3226 3222 3216 3208 3199 3183 3157 3122 
3966 0 0 0 0 0 0 0 0 0 0 0 3231 3227 3222 3220 3216 3212 3209 3205 3201 3198 3195 3191 3186 3179 3172 3158 3137 3108 
1967 0 0 0 0 0 0 0 0 0 0 0 0 2786 2782 2780 2776 2773 2770 2767 2764 2761 2758 2756 2752 2747 2742 2732 2716 2695 
1968 0 0 0 0 0 0 0 0 0 0 0 0 0 3581 3578 3574 3570 3566 3562 3558 3555 3551 3548 3544 3538 3532 3522 3505 3481 
3969 0 0 0 0 0 0 0 0 0 0 0 0 0 0 3908 3903 3899 3894 3890 3886 3882 3873 3875 3870 3865 3860 3850 3834 3813 
1970 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 2789 2786 2783 2780 2777 2774 2772 2769 2766 2762 2759 2753 2744 2731 
1971 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 2380 2377 2375 2372 2370 2368 2365 2363 2360 2357 2353 2346 2337 
1972 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 2721 2717 2715 2712 2710 2707 2704 2700 2697 2692 2686 2677 
1973 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 2143 2141 2139 2137 2135 2133 2130 2127 2124 2118 2112 
1974 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1902 1900 1898 1897 1894 1892 1890 1887 1882 1877 
1975 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 2171 2169 2167 2165 2162 2160 2156 2151 2145 
1976 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1671 1670 1668 1666 1664 1661 1657 1652 
1977 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1700 1698 1696 1694 1691 1687 1682 
1978 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 2312 2309 2306 2302 2297 2290 
1979 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 3045 3041 3036 3029 3020 
I98O 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 4643 4635 4625 4612 
I98I 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 5758 5745 5729 
1Q82 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 6567 6549 
J 983 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 628& 
607 
Year o f 
planting 
1901 
1902 
1903 
1904 
1905 
1906 
1907 
1908 
1909 
1910 
1911 
1912 
1913 
1914 
1915 
1916 
1917 
1918 
1919-
1920 
192! 
1922 
1923 
1924 
1925 
1926 
1927 
1928 
1929 
1930 
1931 
1932 
1933 
1934 
1935 
1936 
1937 
1938 
1939 
1940 
1941 
1942 
1943 
1944 
1945 
1946 
1947 
1948 
1949-
1950 
1951 
1952 
1953 
1954 
1955 
1956 
1957 
1958 
1959 
1960 
1961 
1962 
1963 
1964 
1965 
1966 
1967 
1968 
1969 
1970 
1971 
1 9 7 2 
1973 
1974 
1975 
1976 
1977 
1978 
1979 
1980 
1981 
1982 
Table 2.5. Area distribution smallholdings (ha) 
Year of tapping 
1955 1956 1957 1958 1959 1960 1961 
0 
1 
2 
4 
8 
284 
368 
461 
564 
694 
364 
758 
897 
1037 
1175 
1310 
1440 
1565 
1686 
1802 
1914 
1712 
1725 
1740 
1778 
1787 
1801 
1797 
1796 
1885 
1980 
1364 
1423 
1487 
1557 
1635 
1719 
1811 
1910 
2014 
2124 
2236 
2349 
2459 
1069 
77 
73 
96 
126 
140 
197 
212 
215 
1719 
2394 
0 
0 
0 
0 
0 
0 
0 
0 
0 
0 
0 
0 
0 
0 
0 
0 
0 
0 
0 
0 
0 
0 
0 
0 
0 
0 
0 
0 
o 
1 
1 
3 
6 
196 
254 
318 
389 
478 
397 
556 
691 
832 
975 
1118 
1259 
1396 
1528 
1657 
1782 
1611 
1637 
1664 
1711 
1729 
1751 
1753 
1759 
1851 
1949 
1346 
1407 
1472 
1544 
1623 
1708 
188! 
1901 
2006 
2116 
2230 
2343 
2454 
1067 
76 
72 
96 
125 
140 
197 
211 
215 
1716 
2390 
2790 
0 
0 
0 
0 
0 
0 
0 
0 
0 
t\ 
0 
0 
0 
0 
0 
0 
0 
0 
0 
0 
0 
0 
0 
0 
0 
0 
0 
o 
o 
1 
2 
4 
131 
169 
213 
260 
319 
399 
372 
494 
628 
770 
915 
1062 
1208 
1351 
1492 
1629 
1492 
1533 
1573 
1630 
1658 
1690 
1701 
1713 
1810 
1912 
1324 
1387 
1454 
1528 
1608 
1695 
1789 
1890 
1996 
2107 
2221 
2336 
2447 
1065 
76 
72 
95 
125 
140 
196 
211 
215 
1713 
2386 
2785 
3076 
0 
0 
0 
0 
0 
0 
0 
0 
n 
0 
0 
0 
0 
0 
0 
0 
0 
0 
0 
0 
0 
0 
0 
0 
0 
0 
0 
o 
1 
1 
3 
89 
115 
144 
176 
217 
271 
252 
336 
455 
587 
728 
875 
1025 
1175 
1324 
1472 
1368 
1423 
1476 
1544 
1583 
1623 
1643 
1663 
1764 
1871 
1299 
1364 
1434 
1510 
1592 
1680 
1776 
1877 
1985 
2097 
2212 
2327 
2440 
1062 
76 
72 
95 
125 
140 
196 
210 
214 
1710 
2382 
2781 
3071 
2949 
0 
0 
0 
0 
0 
0 
0 
0 
0 
0 
0 
0 
0 
0 
0 
0 
0 
0 
0 
0 
0 
0 
0 
0 
0 
0 
0 
0 
1 
2 
52 
68 
85 
104 
128 
159 
149 
198 
268 
380 
507 
646 
794 
948 
1104 
1261 
1199 
1272 
1341 
1422 
1476 
1529 
1561 
1592 
1699 
1811 
1264 
1332 
1405 
1483 
1568 
1658 
1756 
1859 
1969 
2082 
2199 
2315 
2429 
1058 
76 
72 
95 
125 
139 
196 
210 
214 
1707 
2377 
2775 
3065 
2943 
3493 
0 
0 
0 
0 
0 
0 
0 
0 
0 
0 
0 
0 
0 
0 
0 
0 
0 
0 
0 
0 
0 
0 
0 
0 
0 
0 
0 
1 
1 
33 
42 
53 
65 
80 
100 
93 
124 
168 
238 
344 
467 
605 
753 
909 
1069 
1042 
1129 
1211 
1304 
1370 
1434 
1478 
1519 
1632 
1749 
1227 
1299 
1375 
1456 
1543 
1636 
1735 
1840 
1951 
2066 
2184 
2302 
2417 
1053 
76 
72 
95 
125 
139 
195 
210 
214 
1703 
2373 
2770 
3059 
2937 
3486 
3220 
0 
0 
0 
0 
0 
0 
0 
0 
0 
0 
0 
0 
0 
0 
0 
0 
0 
0 
0 
0 
0 
0 
0 
0 
0 
0 
0 
20 
26 
33 
40 
49 
62 
58 
77 
104 
147 
213 
315 
435 
572 
721 
879 
884 
982 
1077 
1182 
1261 
1339 
1396 
1449 
1570 
1695 
1191 
1267 
1348 
1433 
1524 
1609 
1711 
1818 
1931 
2048 
2167 
2287 
2403 
1048 
75 
71 
95 
124 
139 
195 
209 
213 
1700 
2368 
2764 
3053 
2932 
3480 
3214 
3020 
0 
0 
0 
0 
0 
0 
0 
0 
0 
0 
0 
0 
0 
0 
0 
0 
0 
0 
0 
0 
0 
0 
1962 1963 1964 1965 1966 1967 1968 1969 1970 1971 1972 1973 1974 1975 1976 1977 
t 
1978 1979 1980 1981 1982 1983 
0 0 0 O 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 
0 0 0 O 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 O 
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 
12 7 5 3 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 
15 9 6 4 3 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 
19 12 8 5 4 3 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0^  
24 15 10 7 4 3 2 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 
29 18 12 8 5 4 3 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 
36 22 15 10 7 5 4 1 1 0 0 0 0 0 0 0 0 0 0 0 0 O 
34 21 14 10 6 5 4 1 1 1 0 0 0 0 0 0 0 0 0 0 0 O 
45 28 19 13 9 7 5 2 1 1 1 0 0 0 0 0 0 0 0 0 0 O 
61 38 25 17 12 9 6 2 2 1 1 1 0 0 0 0 0 0 0 0 0 0 
87 53 36 - 24 16 !3 9 3 2 2 1 1 1 0 0 0 0 0 0 0 0 0 
126 77 52 35 24 18 13 5 3 2 2 1 1 1 0 0 0 0 0 0 0 0 
186 114 77 52 35 27 20 7 5 3 3 2 2 1 1 0 0 0 0 0 0 0 
283 174 117 79 54 41 30 10 7 5 4 3 3 2 1 1 0 0 0 0 0 0 
399 267 180 122 83 64 46 16 11 8 6 5 4 3 2 2 2 0 0 0 0 0 
534 382 276 187 126 98 70 24 17 12 10 7 6 4 3 3 3 2 0 0 0 0 
684 517 393 285 192 149 107 37 26 19 15 11 9 6 5 5 4 3 3 0 0 0 
715 566 450 344 248 192 139 48 33 24 19 • 15 12 8 7 6 5 5 4 3 0 0 
822 676 559 446 340 274 198 68 47 35 21 21 17 11 9 9 8 7 6 5 4 0 
927 788 672 557 444 370 282 97 67 50 39 30 24 16 13 13 11 10 9 7 6 6 
1042 910 797 681 564 485 386 168 117 86 68 52 41 27 23 21 19 16 14 12 11 9 
1135 1015 910 799 682 602 497 256 188 139 109 84 66 44 37 34 30 26 22 19 17 14 
1226 1118 1023 919 807 727 619 361 279 216 170 130 103 68 58 53 47 39 34 30 25 22 
1297 1204 1121 1027 923 847 742 475 382 308 252 193 153 101 86 78 69 58 51 44 38 32 
1364 1286 1215 1132 1039 969 868 600 499 416 350 280 221 147 124 113 100 84 73 63 54 46 
1493 1425 1363 1288 1200 1135 1037 761 651 557 482 398 327 217 184 166 146 123 107 92 79 68 
1627 1568 1515 1449 1369 1309 1216 937 820 719 635 540 457 324 275 248 217 182 157 136 116 99 
1146 1105 1067 1021 969 928 868 697 622 556 500 436 378 284 245 214 183 151 127 107 91 77 
1228 1193 1160 1121 1074 1038 982 815 740 672 614 546 484 380 336 302 258 212 179 152 129 109 
1313 1284 1257 1224 1183 1151 1101 940 865 797 738 667 602 491 443 408 359 296 250 212 179 153 
1403 1380 1359 1331 1296 1269 1224 1070 997 929 870 798 731 615 565 532 481 410 347 294 249 212 
1498 1481 1465 1443 1414 1393 1353 1206 1135 1069 1010 939 870 751 702 674 624 549 479 406 345 293 
1588 1576 1567 1550 1528 1512 1478 1341 1273 1209 1151 1081 1014 894 846 829 785 709 637 556 472 402 
1681 1675 1671 1660 1645 1633 1606 1477 1413 1352 1297 1229 1162 1045 1000 997 963 891 822 737 643 547 
1791 1761 1763 1759 1749 1744 1723 1604 1544 i486 1434 1370 1306 1192 1153 1167 1148 1086 1027 943 842 735 
1906 1875 1852 1854 1851 1851 1836 1727 1672 1618 1569 1508 1448 1339 1307 1341 1339 1293 1251 1173 1070 956 
2025 2000 1976 1949 1952 1957 1948 1849 1797 1748 1702 1645 1589 1486 1461 1516 1534 1507 1488 1425 1325 1207 
2147 2124 2102 2077 2047 2057 2055 1964 1917 1872 1830 1777 1725 1629 1611 1687 1727 1721 1730 1688 1600 1484 
2268 2247 2227 2204 2177 2156 2159 2077 2034 1993 1954 1906 1858 1769 1758 1857 1918 1934 1973 1955 1886 1782 
2386 2367 2349 2328 2304 2285 2257 2183 2145 2107 2072 2028 1983 1902 1899 2018 2101 2140 2210 2220 2174 2089 
1042 1035 1028 1020 1011 1004 993 965 951 936 923 906 889 857 849 866 874 870 873 "863 840 809 
75 74 74 74 73 72 72 70 69 68 67 66 65 63 70 99 124 146 175 194 205 210 
71 7t 70 70 70 69 69 67 66 66 65 64 63 62 67 91 113 132 157 175 186 194 
94 94 93 93 92 92 91 90 89 88 87 86 85 83 86 102 115 127 144 156 163 168 
124 123 123 122 122 121 121 119 118 117 116 115 114 111 117 144 167 189 218 241 256 269 
138 138 138 137 137 136 135 134 133 132 131 130 129 126 137 185 228 267 322 366 398 426 
914 194 193 193 192 191 191 188 187 186 185 184 182 180 191 241 287 329 388 435 472 504 
209 208 208 207 207 206 205 203 202 201 200 199 198 195 194 254 309 361 434 493 540 584 
213 212 212 211 211 210 210 208 207 206 205 204 203 201 199 198 270 339 436 516 581 644 
1696 1693 1690 1687 1683 1680 1675 1663 1656 1650 1644 1636 1628 1613 1604 1595 1583 1760 2016 2226 2399 2570 
2363 2358 2354 2350 2346 2342 2336 2322 2314 2306 2299 2289 2280 2262 2251 2240 2225 2206 2520 2782 3002 3221 
2759 2753 2748 2744 2739 2735 2730 2715 2707 2699 2692 2682 2673 2654 2641 2632 2617 2598 2576 2861 3103 3348 
3046 3040 3035 3030 3025 3021 3016 3002 2995 2987 2980 2971 2962 2944 2934 2923 2909 2890 2870 2848 3134 3428 
2926 2920 2915 2910 2905 2902 2897 2886 2879 2873 2867 2860 2852 2838 2830 2820 2809 2793 2776 2758 2738 2997 
3472 3465 3460 3454 3448 3444 3439 3427 3421 3414 3408 3401 3393 3378 3370 3360 3348 3332 3315 3296 3276 3253 
3207 3201 3196 319! 3185 3181 3177 3166 3161 3156 3151 3145 3139 3127 3120 3113 3103 3090 3076 3061 3045 3027 
3014 3008 3003 2998 2993 2990 2985 2975 2970 2966 2962 2958 2952 2943 2937 2931 2924 2913 2902 2890 2877 2862 
2939 2933 2928 2923 2919 2915 2911 2901 2897 2893 2889 2885 2881 2873 2868 2863 2857 2848 2838 2828 2817 2805 
0 2594 2589 2585 2581 2578 2574 2565 2561 2558 2555 2552 2549 2543 2539 2535 2530 2523 2516 2508 2500 2491 
0 0 2048 2045 2041 2039 2036 2029 2026 2023 2021 2018 2016 2012 2010 2007 2004 1999 1994 1989 1983 1977 
0 0 0 2058 2054 2052 2049 2042 2039 2036 2034 2032 2029 2026 2024 2022 2019 2015 2010 2006 2001 1995 
0 0 0 0 2021 2019 2016 2010 2006 2004 2002 1999 1997 1993 1992 1990 1987 1984 1980 1976 1972 1967 
n 
V 
0 0 0 0 1355 1353 1349 1347 1345 1344 1342 1341 1338 1337 1336 1334 1333 1330 1328 1326 1323 
0 0 0 0 0 0 1814 1808 1805 1803 1801 1798 1797 1793 1792 1790 1789 1786 1784 1781 1779 1775 
0 0 0 0 0 0 0 1708 1705 1703 1701 1699 1697 1694 1693 1691 1690 1688 1686 1684 1682 1679 
0 0 0 0 0 0 0 0 1470 1468 1466 1464 1463 1460 1459 1458 1457 1455 1454 1452 1450 1448 
0 0 0 0 0 0 0 0 0 1292 1290 1289 1287 1285 1284 1283 1282 1281 1279 1278 1277 1275 
0 0 O 0 0 0 0 0 0 0 1000 999 998 996 996 995 994 993 992 991 990 989 
0 0 0 0 0 0 0 0 0 0 0 990 989 987 986 985 985 983 982 982 981 980 
0 0 0 0 0 0 0 0 0 0 0 0 999 997 997 996 995 994 993 992 991 990 
0 0 0 0 0 0 0 0 0 0 0 0 0 1216 1215 1214 1213 1211 1210 1209 1208 1207 
0 0 0 0 0 0 0 0 0 0 0 0 0 0 936 935 934 933 932 931 93! 930 
0 0 0 p 0 0 0 0 0 0 0 0 0 0 0 951 951 950 949 948 947 946 
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1294 1293 1292 1290 1289 1288 
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1704 1703 1701 1700 1699 
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1770 1768 1767 1765 
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1748 1747 1745 
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1865 1863 
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1714 
608 
1983 
system, and the distribution by vintages can be developed year by year. Thus, for each 
year, theoretical discarding by vintage is adjusted to take care of e.g. diseases or high 
levels of replanting : this is the basis for discarding for the following year. 
The above analysis requires a vintage distribution in the starting year, say 1951. 
Using data on planting in the years up to 1955 as well as a number of feasible mathemati­
cally formulated discarding systems, one may derive a number of feasible vintage distri­
butions in the starting year. That discarding system and the corresponding starting 
year vintage distribution may now be selected, which provides an optimal fit to data on 
discarding for the sample period, from the starting year onwards. 
A graphical presentation of a possible discarding system is shown in Fig. 2.3. 
An elaborate description of possible discarding systems and the selection of the optimal 
one can be found in Smit (1984). 
For Sri Lanka a good system appeared to be a linear decrease with 5 % for the first 
10 years and an exponential decrease afterwards with exponent 0 17% age of the trees 
for estates and 0* 15% age of the trees for smallholdings. The resulting area distributions 
for estates and smallholdings are given in Tables 2.4 and 2.5. 
A supply function for natural rubber 
Basic elements of a supply function 
The basis for a supply function is a production function. This production function 
relates production to : 
— area, disaggregated into vintages 
— yield profiles, so that each vintage can be multiplied with the normal yield of 
a vintage of planting year T in tapping year t 
— technical progress factors shifting yield profiles upwards for later planted, more 
productive vintages 
— flexible inputs such as manpower 
— exogenous variables like the weather condition 
— the status of the trees in view of tapping in the past : more or less than normal 
tapping will have adverse effects on later years ; if e.g. in year t slaughter tapping 
is applied, then this may slightly reduce the yield in year t + 1 and strongly in 
year t + 4 in view of bark consumption and renewal. Slaughter tapping during 
a number of consecutive years will dramatically reduce the yield of the tree for 
the rest of its life (cf. World War II). 
609 
0 20 40 60 0 20 40 GO 
age k ace k 
Fig. 2.3. Percentage of remaining area, discarded in year k (p ) and percentage of original area 
remaining after k years (r k) with a constant reduction in r k to 95% after 20 years. 
6 1 0 
The factor to be determined by the rubber producer is the level of flexible inputs. 
This will depend on the market situation. Classic approaches argue, that profit maxi­
mization is the key determining factor. This leads to the inclusion of price elements in 
the supply function to be derived from the production function on the basis of profit 
maximization, while flexible inputs, on which no data are available, can be eliminated. 
The production function 
Yield profiles: Vintages have been discussed in the previous section. In order to arrive 
at the yield for each vintage in the year of tapping the vintage needs to be multiplied by 
the average yield appropriate for the age of the vintage. The study on " The Agro-
Economic Norm for Natural Rubber " by the Association of Natural Rubber Producing 
Countries (1976) contains a chapter on Sri Lanka. From this chapter an ideal yield profile 
is derived as presented in Table 3.1 and Fig. 3.1. 
Table 3.1. Ideal yield profile (kgs per ha) 
Year of Year of 
tapping Yield tapping Yield 
1 420 29 900 
2 730 30 850 
3 920 31 800 
4 1,090 32 750 
5 1,180 33 700 
6 1,230 34 650 
7 1,290 35 600 
8 1,350 3 6 550 
9 1,400 3 7 500 
10 1 , 4 0 0 3 8 450 
1 1 1,400 3 9 400 
' 12 1,400 40 350 
13 1,400 41 300 
14 1,400 42 280 
15 1,350 43 260 
16 1,350 44 240 
17 1,350 45 220 
18 1,350 46 200 
19 1,350 47 180 
20 1,230 48 160 
21 1,230 49 140 
22 1,230 50 120 
23 1,230 51 100 
24 1,170 52 80 
25 1,100 53 60 
26 1,050 54 40 
27 1,000 55 20 
28 950 56 0 
611 
kgs.per ha. 
to 
Fig. 3 . 1 . Ideal yield profile. 
Partly, the yield profiles have been derived by straightforward extrapolation in order 
to complete the picture for very old trees. These yields are obtained under experimental 
conditions. In commercial practice, yields will be somewhat lower, although the shape 
of the yield profile may still be valid. The ideal yield profile therefore needs to be multi­
plied with a certain factor, in order to reduce the ideal yield profile to actual levels. This 
multiplication factor will be different for different countries, and, within countries, for 
estates and smallholdings. A reason may be the selection of clones. This selection of 
clones of which a certain vintage is composed may vary over time, implying that the multi­
plication factor may need to increase over time in view of technical progress. 
The assumption now is that each vintage will have an average yield profile, which is 
a constant fraction of the ideal yield profile : if a yield profile is estimated to be for example 
0 • 5 times the ideal yield profile, then the profile is suppressed to 50 percent of the original 
shape with a value of 365 kg per ha in the second year of tapping. Of course, later 
vintages may be composed of better clones, thus increasing average yield. In the example, 
the fraction of 0 - 5 may become 0-6 after a number of years. The fractions are estimated 
per vintage in the following sub-section, by relating area and ideal yield profile to pro­
duction and then deriving the fractions which create actual yield profiles that are consis­
tent with area composition and production. As a matter of clarification, it is derived 
that trees planted e.g. in 1950 on estates have a yield profile which is about 75 percent 
of the ideal yield profile. This vintage, of course, will only reach high production levels 
around 1962. 
Technical progress: There are a number of factors which have contributed to the increase 
in productivity of rubber land. By far the most important one is the improvement of 
the quality of trees. This partly needs to be combined with the application of fertilizers 
and possibly stimulants. All this may be put under the collective name "embodied 
technical progress ". There is another group of factors improving yield of rubber land 
which may be called " disembodied technical progress ". To this group belong such 
aspects as improved tapping methods. Embodied technical progress, embodied in the 
quality of the trees and thus in the vintages is by far the most important group. It has 
therefore been decided to emphasize the inclusion of embodied technical progress in the 
analysis, which then can be specified as a function of the vintage, the year of planting. 
The production function may therefore be specified as : 
S f C O y t - T a k (3.1) 
k = l 
production capacity of N R in Sri Lanka 
estates 
smallholdings 
year of planting 
year of tapping 
embodied technical progress function 
ideal yield profile, age = t - x 
area of vintage T still remaining in year f 
with q* _ 
t 
k = 1 
k = 2 = 
T = 
t = 
fk (T) 
Y t - T 
a^T = 
613 
The ideal yield profile is adjusted for the length of the immaturity period, which 
may vary around 6 years. There are many possible specifications for f k (x). In this 
study only linear functions are applied. 
In view of the lack of data on production capacity or " normal production " and 
on flexible input, it has not been attempted to estimate this production function. Below 
the supply function, including prices, will be estimated directly. Then factors like weather 
conditions and slaughter tapping will be investigated as well. 
The supply function 
Moving from the production function to the supply function, special attention needs 
to be paid to those factors determining the degree of capacity utilization u c t 
qt = u V q * t (3.2) 
with qt = production 
*qt = production capacity or " normal production " 
Above a number of factors determining ut has already been mentioned : 
f(P) = function of prices 
f (E c ) = the influence of the rubber-rice-barter with China until 1982 
f(R) = rainfall 
f((°,/°.*)t - 1 ) = " abnormal" tapping in the past 
The supply function is now broadly specified 
2 
qt = f(P) f(E°) f(R)• f (q/q*) t - x) 2 f k (x) Y t - xa k t T (3.3) 
k = 1 
Presentation of the numerous detailed specification and estimation results would 
require many uninteresting pages. Only the optimal results are given below. An itera­
tive procedure has been used to estimate the parameters. Note that in view of the number 
of parameters it has not been possible to obtain parameters for the factors influencing u c t 
separately for estates and smallholdings. Technical progress functions, however, have 
been estimated separately for estates and smallholdings. The resulting value for every 
tenth year is shown in Table 3.2. 
Table 3.2. Estimated values of embodied technical progress functions {equation (3.3)) 
Year Estates Smallholdings 
1910 0 1 6 5 0 149 
1920 0 168 0 152 
1930 0 1 7 1 0 155 
1940 0 646 0 592 
1950 0-673 0 620 
1960 0-701 0 647 
1970 0-728 0-675 
1980 1-414 1-317 
1990 1 469 1-372 
2000 1-525 1-427 
6 1 4 
these technical progress functions, together with the " ideal" yield profile and the 
area distribution determines " n o r m a l " production as given in equation (3.1). The 
parameters of the technical progress functions are estimated simultaneously with the 
parameters of the functions determining u° t . The year 1970, being rather exceptional, 
has been excluded. The results are : 
0 1 9 9 4 - 0 4525 0 1 4 1 3 0 5893 
qt = 1 0937(P t +) • ( P r ) • (E ct) • ( R ~ ) 
—0 4114 
q t _ . 0 - 4 ) 
• ( ) • q*t 
q * t _ 4 
R 2 = 0 984 
with P t + = Pt/Pt_! if Pt > P«_, 
= 1 otherwise 
P - = P t / P ^ i f P t < P . _ ! 
= 1 otherwise 
E ct = (1 + E t C h i n a /E t ) 
E t C h i D a = exports of rubber to China, effective until 1982 
Et = exports of rubber 
R~t = actual rainfall divided by normal rainfall, of less than normal 
= 1 otherwise. 
The theoretical background of the specification of the above equation is based on 
assumptions of consumer and producer theory, in particular that in a period of declining 
prices, producers may need to produce more, if they have no alternative sources of in­
come, in order to keep their consumption at an adequate level. The observed and the 
estimated values are shown in Table 3.3 and Fig. 3.2 
Table 3.3. Observed values and estimated values (equation (3.4) of rubber production 
(1000 tonnes) 
Year Observed production Estimated production 
1955 95-3 93 3 
1956 96-9 97-5 
1957 99-7 101-6 
1958 101-8 95-4 
1959 93-2 93 6 
615 
Table 3.3 (Continued) 
Year Observed production Estimated production 
1960 98 8 97 9 
1961 97 6 101-9 
1962 104 1 100 7 
1963 104 8 108-1 
1964 1 1 1 6 114 9 
1965 1 1 8 3 123-2 
1966 131 0 130-3 
1967 143-2 143-2 
1968 148 7 144-4 
1969 150 8 147 0 
1970 159-2 150 4 
1971 141-4 143-5 
1972 140-4 139 3 
1973 154-7 152-9 
1974 132 0 135 3 
1975 148-8 149 8 
1976 152-1 153-2 
1977 146-2 144 6 
1978 155-7 159-3 
1979 152-7 150 3 
1980 133-2 135-7 
1981 123 9 123-5 
1982 125-2 121-9 
1983 139 9 139-1 
It has not been possible to estimate standard errors for the parameters. An approxi­
mation can be found by using " normal capacity " q*t as estimated above and by trans­
forming (3.4) into natural logarithms. Estimation results with t - values in brackets are : 
In q t = 0 2521 + 0-9864 In q* t + 0 2030 !n P + t - 0 4339 In P ~ t + 
(27-402) (4-241) (4 651) 
+ 0 1329 In E c t + 0 5666 In R - — 0 4278 In q t - 4 (3.5) 
(1-984) (6-252) (3 853) q* t - 4 
R 2 = 0-982, D W = 2 38 
While the coefficient of In E t c is almost significant, all other coefficients are significant 
and all are very close to the results of equation (3.4). 
616 
o O b s e r v a t i o n 
A E s t i m a t e d v a l u e 
1 0 0 0 r . o n n e s 
1 6 0 "I 
Fig. 3.2. Observations and estimated values for rubber production. 
Projections of natural rubber production 
An optimistic planting scenario 
The analysis described in Chapter 3 may now be used as a tool to assess what the 
effects of planting policies on future natural rubber production may be. Projections 
will be made up to the year 2000. 
Such projections depend on a lot of factors which can be broadly divided into two 
groups : 
—- planting policies 
— other factors. 
To the first group, planting policies, belong such variables a s ; 
— total area under rubber 
— replanting with rubber 
— new planting 
— uprooting for alternative land use. 
Any consistent set of assumption for 1984 — 2000 may of course be fed into the model. 
Here, we have used the figures as shown in Table 4.1. Replanting and new planting levels 
may be considered rather optimistic. For the other variables the following assumptions 
have been used : 
— the prices of N R in Colombo will run parallel to the world market prices as 
projected in Smit (1984)-for the standard scenario ; 
— rainfall will be normal; 
Using the above assumptions, projections are derived. They are presented in Table 
4.2 and shown in Fig. 4.1. Production in 1000 tonnes will reach levels of around 150 
in the second half of the 1980's, 200 in the first half of the 1990's when the replanting efforts 
of the 1980's reach their maximum momentum, production braking the 200 barrier around 
1993. 
It is interesting to have a look at area productivity as shown in Table 4.3. Yield 
per ha shows a dramatic increase in the sixties because of the replanting efforts in the 
1950's. A similar development is shown in the 1980's and 1990's, showing the success 
of the current replanting schemes. Around the year 2000 average yield per ha will exceed 
1,200 kg. 
618 
8 
lfiOO t o n n e s 
220 -I 
200 4 
ON 
VO 
1S55 1960 1970 1980 1990 
Fig. 4.1. Observations and projections of natural rubber production. 
Table 4.1. Assumptions on planting policiesfoi-naturalrubberarea (ha) 
Estates Smallholdings * 
Total New New 
Year area Area Replanting planting Area Replanting planting 
1983 194,000 120,280 5,500 786 73,720 1,500 214 
1984 193,000 117,730 6,000 800 75,270 1,500 200 
1985 192,000 115,200 6,500 813 76,800 1,500 188 
1986 191,000 114,600 6,500 406 76,400 1,500 94 
1987 190,000 114,000 5,500 393 76,000 1,500 107 
1988 188,500 113,100 4,500 375 75,400 1,500 125 
1989 186,000 111,600 3,500 350 74,400 1,500 150 
1990 185,000 111,000 3,500 350 74,000 1,500 150 
1991 184,000 110,400 3,500 210 73,600 1,500 90 
mi 183,000 109,800 3,500 210 73,200 1,500 90 
1993 182,000 109,200 3,500 210 72,800 1,500 90 
1994 181,000 108,600 3,500 210 72,400 1,500 90 
1995 180,000 108,000 3,500 210 72,000 1,500 90 
1996 179,000 107,400 3,500 140 71,600 1,500 60 
1997 178,000 106,800 3,500 140 71,200 1,500 60 
1998 177,000 106,200 3,500 140 70,800 1,500 60 
1999 176,000 105,600 3,500 140 70,400 1,500 60 
2000 175,000 105,000 3,500 140 70,000 1,500 60 
Table 4.2. Projections of natural rubber production (1000 tonnes) 
Year Production 
1984 141-3 
1985 142-2 
1986 148-8 
1987 145 1 
1988 149-1 
1989 158 8 
1990 164-6 
1991 1 7 9 1 
1992 193-5 
1993 202-8 
1994 212-5 
1995 225-6 
620 
fable 4.2 (Continued1) 
Year Production 
1996 229 6 
1997 224 2 
1998 226-3 
1999 224 9 
2000 228 4 
Table 4.3. Productivity of rubber area 
Average production Average area Average yield per ha 
per year 1 per year 4 per year* 
1956 — 1960 98 1 237 414 
1961 — 1965 107 -2 228 470 
1966 — 1970 146 6 219 670 
1971 — 1975 143 5 207 692 
1976 — 1980 148 0 196 755 
1981 — 1985 1 3 4 5 193 697 
1986 — 1990 153-3 188 815 
1991 — 1995 202-7 182 1,114 
1996 — 2000 226-7 177 1,281 
(1) based on Tables 1.1 and 4.2, in thousand tonnes. 
(2) based on Tables 2.2 and 4.1, in thousand ha. 
(3) derived from the previous columns, kg, per ha. 
Effects of alternative planting scenarios 
In section on 'An optimistic planting scenario' planting policies were described 
which may be considered rather optimistic. The resulting projections of natural rubber 
production may be viewed accordingly. They are summarized again in Table 4.4. 
Table 4.5 shows results of a pessimistic scenario with replanting levels of around 1-6 % 
and Table 4.6 depicts a very pessimistic scenario with replanting levels of around 1 %. 
Of course, low levels of replanting leave more trees in the ground, resulting in higher 
production levels in the medium term. In 1990, the optimistic, pessimistic and very 
pessimistic scenarios give 164 6,173 • 8 and 177-2, respectively. However, in the first half 
of the 1990's the order reverses with the optimistic scenario giving the highest figure : 
225 • 6 as against 211-5 and 195-5 for the other scenarios. For the very pessimistic scenario, 
the highest value is reached in 1995 with a rapid and steady decline afterwards. The 
pessimistic scenario provides a slight decline in the second half of the 1990's, while the 
optimistic scenario yields rather stable levels of production. If planting is carried out 
at a reasonable level, production will reach levels of over 200,000 tonnes in the 1990's. It 
will depend on the level of replanting and new planting whether that level can be main • 
tained after the year 2000. 
621 
1 
Table 4.4. An optimistic outlook on planting (1000 ha) and production (1000 tonnes) 
Area Replanting Newplanting Production 
1983 194 0 7 0 1 0 139-9 
1984 1 9 3 0 7-5 1 0 141-3 
1985 192 0 8 0 1 0 142-2 
1986 191 0 8 0 0-5 148-8 
1987 190 0 7 0 0 5 145 1 
1988 188-5 6 0 0 5 149 1 
1989 186 0 5 0 0-5 158-8 
1990 1 8 5 0 5 0 0-5 164-6 
1991 184-0 5 0 0 3 179-1 
1992 1 8 3 0 5 0 0-3 193-5 
1993 182-0 5 0 0-3 202-8 
1994 1 8 1 0 5 0 0-3 212-5 
1995 180 0 5 0 0-3 225-6 
1996 179 0 5 0 0-2 229-6 
1997 178-0 5 0 0-2 224-2 
1998 177 0 5 0 0-2 226-3 
1999 1 7 6 0 5 0 0-2 224-9 
2000 1 7 5 0 5 0 0-2 228-4 
Table 4.5, A pessimistic outlook on planting (1000 ha) and production (1000 tonnes 
Area Replanting New planting Production 
1983 1 9 4 0 7 0 1 0 139-9 
1984 1 9 3 0 6 0 0-5 142-4 
1985 192 0 5 0 0-5 146 0 
1986 1 9 1 0 4 0 0-3 155-9 
1987 190-0 4 0 0-3 154-1 
1988 188-5 4 0 0-2 1 5 9 1 
1989 I 8 6 0 3 0 0-2 169 1 
1990 1 8 5 0 3 0 0-2 173-8 
1991 1 8 4 0 3 0 0 1 184-7 
1992 183 0 3 0 0 1 193-8 
1993 1 8 2 0 3 0 0 1 197-2 
1994 1 8 1 0 3 0 0 1 202-2 
1995 180-0 3 0 0 1 211-5 
1996 179 0 3 0 0 1 212-1 
1997 1 7 8 0 3 0 0 1 204-6 
1998 1 7 7 0 3 0 0 1 204-4 
1999 1 7 6 0 3 0 0 1 201-4 
2000 175 0 3 0 0 1 203-2 
622 
Table 4.6. A very pessimistic view on planting (1000 ha) and production (1000 tonnes) 
Area Replanting New planting Production 
1983 194 0 7 0 1-0 139-9 
1984 193 0 3 0 0 3 144 0 
1985 1 9 2 0 3 0 0-3 149 0 
1986 191 0 2 0 0 2 160-3 
1987 190 0 2 0 0 1 159-4 
1988 188 5 2 0 0 1 164-8 
1989 I 8 6 0 2 0 0 05 174-1 
1990 1 8 5 0 2 0 0 05 177-2 
1991 1 8 4 0 1 0 0 0 5 183-8 
1992 1 8 3 0 1 0 0 05 189 1 
1993 182 0 1 0 0 0 5 188-9 
1994 1 8 1 0 1 0 0 0 5 190-2 
1995 180 0 1 0 0 05 195-5 
1996 179 0 1 0 0 05 193-4 
1997 178 0 1 0 0 05 183-7 
1998 177 0 1 0 0 0 5 179-8 
1999 176 0 1 0 0 05 173-4 
2000 1 7 5 0 1 0 0 0 5 171-2 
REFERENCES 
Association of Natural Rubber Producing Countries (1976). The agro-economic norm 
for natural rubber production, Kuala Lumpur. 
Economic Intelligence Unit (1979). The Rubber Sector of Sri Lanka, in Rubber Trends, 
London, March, 1979. 
Economic Intelligence Unit (1982). Sri Lanka "Rubber Trends", London, June, 1982. 
KEEL, J. (various issues). John Keels Rubber Statistics, Colombo. 
Rubber Controller (various issues). Administration Report of the Rubber Controller, 
Colombo. 
SMIT, H. P. (1984). Forecasts for the world rubber economy to the year 2000, Macmillan. 
623 
THE INTERNATIONAL NATURAL RUBBER AGREEMENT : 
AN ASSESSMENT 
By 
C. SUAN.TAN* 
(The WorldBank, Washington\D.C.t USA) 
INTRODUCTION 
The International Natural Rubber Agreement (INRA) is the first commodity agree­
ment concluded under the U N C T A D Integrated Programme for Commodities. This 
5-year agreement came into force in October 1980 and was ratified and made operational 
at the first meeting of the International Natural Rubber Council (INRC) at Geneva in 
January 1981. At this meeting the principal Council officers were appointed and Kuala 
Lumpur named as the location for the INRC headquarters. Under this agreement price 
stabilisation was to be by the sole instrument of buffer stock (BS) operations of the Inter­
national Natural Rubber Organisation (INRO). The agreement also provides for a 
2 year extension upon its expiry in October 1985.». 
The BS operations began in November 1981 when natural rubber prices fell below 
the INRA-stipulated floor price of % 2100 per tonne in Malaysian and Singapore dollar 
currency. These BS purchases continued through to April 1983 when prices came within 
the stabilisation price band after purchasing some 270,000 tonnes of BS purchases. Since 
April 1983, the BS operations have been dormant as price remained within the stabilisa­
tion band. 
The eighth INRC meeting in July 1984 decided that instead of a 2 year extension 
from October 1985, the INRA will be renegotiated^). From the mandate granted by 
all INRA member governments to renegotiate a " successor " agreement, it would seem 
that despite some earlier mixed feelings about the efficacy of the first INRA, member 
countries now believe in striving to improve this efficacy. It is therefore useful at this 
juncture to analyse the strategies by which BS operations can reduce the natural rubber 
price volatility over the longer term, and to indicate the associated operational difficulties. 
Tins paper assesses the long term efficacy of the INRA by simulating the INRA 
rules in an econometric annual model of the world natural and synthetic rubber markets. 
In Section II the operational rules of the INRA are presented and translated into rules 
for simulations in the model (outlined in the Appendix). Aspects of INRA that need 
* The author is a World Bank staff member. The World Bank does not accept responsibility for the 
views expressed herein which are those of the author and should not be attributed to the World Bank 
or to its affiliated organizations. The findings, interpretations, and conclusions are the results of research 
supported by the Bank ; they do not necessarily represent official policy of the Bank. The designations 
employed and the presentation of material in this document are solely for the convenience of the reader 
and do not imply the expression of any opinion whatsoever on the part of the World Bank or its 
affiliates concerning the legal status of any country, territory, city, area, or of its authorities, or 
concerning the delimitation of its boundaries, or national affiliation. 
(1) A Preparatory Committee will meet during November 1 2 — 16,1984 in Kuala Lumpur to prepare the 
materials for, and organise the Negotiating Conference scheduled for 1985 at Geneva. 
625 
consideration are then discussed in Section III. Section IV describes the rubber market 
briefly to highlight the underlying market dynamics that must be considered when opera­
ting the BS. The results of ex ante (for 1970 — 1980) and ex post (1985 — 1995) simu­
lations of BS operations are presented in Section V. Finally, caveats for successful BS 
operations are given in Section VI. 
Overall the simulation results highlight the role of the lagged structure of the rubber 
market in constraining the speed with which BS operations can support price, and hence 
the short run effectiveness of the BS. Conversely, the effectiveness of the accumulated 
BS in defending the ceiling price in any economic recovery depends on its size relative 
to the speed of the recovery. 
Stabilisation aims and intervention price bands 
The declared aims of the INRA (UNCTAD, 1979) are (i) to stabilise natural rubber 
price and (ii) to obtain steady growth in the natural rubber export revenues of the pro­
ducing countries. The sole instrument for this market intervention is a buffer 6tock of 
550,000 tonnes total capacity. 
To guide the BS management on the timing of intervention, four sets of alternative 
price levels, and hence price bands, relating to optional and compulsory INRA interven­
tions are specified. As shown in Fig. 1, these are (a) a reference price which approxi­
mates to the average price, (b) the intervention price band defined by upper and lower 
intervention prices delineated by the reference price plus and minus 15% respectively, 
(c) the trigger price band defined by the upper and lower trigger prices delineated by the 
reference price plus and minus 20 % respectively, and (d) the indicative price band defined 
by upper and lower indicative prices delineated by the reference price plus and minus 
28 5%. Thus the indicative prices are the floor and ceiling prices. The reference price 
is the key to successful stabilisation since it provides the nexus between the BS and the 
market and is the price upon which the remaining prices are determined. 
In practice, BS operations will be conducted on the basis of the daily market indi­
cator price (DMIP) which is a composite weighted average of daily official current-month 
prices on the Kuala Lumpur, Singapore, London and New York markets. For each 
market, the DMIP is an equally-weighted average of the prices for the RSS1, RSS3 and 
TSR20 grades of the natural rubber quoted in Malaysian or Singapore currency. (All 
prices are quoted in Singapore dollars hereafter). To monitor price behaviour, the average 
market price for 5 or more consecutive days provides the yardstick for evaluating the 
DMIP in relation to the INRA-specified price levels. When the DMIP remains higher 
than the reference price for 5 or more consecutive days, the DMIP is deemed to exceed 
the reference price. 
To provide a measure of the extent of price support required at anytime, the total 
BS is differentiated into a normal BS of 400,000 tonnes and a contingency BS of 150,000 
tonnes. In this two-part designation, the normal BS is expected to be adequate for de­
fending the indicative price band under "normal circumstances". Should "normal 
circumstances " not obtain (which is tautologically defined to be when BS sales/purchses 
amounting to 400,000 tonnes fail to bring price within the trigger price band where inter­
vention is optional), the contingency BS will be mobilised when either (a) the DMIP 
626 
ON 
Upper Ind ica t ive Price (2 1/2 years) $2100 + 28.5% » $2700 
Pr ice _for_bri_nSing_ in _Contlngency_BS $J6_10 \ Compulsory BS 
Upper_Tr i^ep Price ' $2100 + 24.0% » $2520 :} 
Upp,e«r Intervent ion Pr ice 
Reference Price • (1 1/2 years) 
$2100 + 15.0% = $2415 
$2100 
 J Optional BS Sales 
No BS Operations Except 
Stock Rotat ion 
Lower Intervent ion Pr ice $2100 - 15.0% = $1790 
, ™ , r,
 4 o m n <•>/ n«/ EMAN 1 Opt ional BS Purchases Lower Tr igger Pr ice $2100 - 24.0/t = $1680 [ 
Pr ice for Br inging Contingency BS _ $1590 
Lover Ind ica t ive Pr ice (2 1/2 years) $21.00 - 28.5% = $1500 
Compulsory BS Purchases 
Fig. 1 . Price ranges specified for INRA BS operations. 
lies in the band where intervention is compulsory or (b) the DMIP reaches the trigger 
level. Unless the INRC votes to the contrary, the contingency BS will be mobilised 
when the DMIP reaches one of the averages of the trigger and indicative prices (i.e. 
$ 1,590 and $2,610). Thus when necessary, the entire 550,000 tonnes BS will be used to 
ensure stabilisation within the floor and ceiling prices. 
Basically, the reference price is to remain for 1£ years at $ 2,100, while the indicative 
(floor and ceiling) prices of $ 1,500 and $ 2,700 are valid for 2\ years. Provision for inter­
vention price revisions during the 5-year agreement period is given by clauses couched 
in terms of the speed of BS accumulation/releases. To check that the" predetermined 
reference price is appropriate, the net change in BS volume cannot exceed 100,000 tonnes 
within the \ \ year validity period of the reference price. To ensure effective utilization 
of the BS resources, the net total BS sales/purchases since the INRA enforcement or since 
any reference price revision may not exceed 300,000 tonnes. Thus the need for revising 
the reference and indicative prices is determined by movements in the DMIP and/or 
the net change in BS holdings. 
Should the 6 month average DMIP remain outside the trigger price band despite 
the BS intervention, the reference price will be revised up/down by 5 %. Furthermore, 
should the net change in BS exceed either 100,000 tonnes within \ \ years or 300,000 tonnes 
at any time, then the reference price will be revised accordingly by 3 %, unless the INRC 
decides otherwise. 
Following any reference.price revision, the indicative prices can be revised whenever 
the reference price is either revised by more than 5% within \ \ years or more than 3 % 
within 2\ years and the DMIP still falls outside the indicative price bands in the ensuing 
60 days. 
Some unexplored aspects of the INRA 
Assuming the desirability of price stabilisation, the first aspect of INRA to consider 
is that of feasibility. Although the INRA is intended to function for a 5-year term, the 
INRA price bands are specified up to terms of \ \ and 2\ years only after which they are 
subject to review and revision. While frequent price/band revisions should enhance the 
effectiveness of the BS operations, some more automatic rule for adjusting the reference 
price to take account of the price trend forecast may be called for to avoid time-consuming 
and costly negotiations. 
Another aspect of INRA to consider is that of prospective BS operations from a 
longer term perspective. Prospective BS operations refer to operating the BS where 
the INRA rules stipulate such operations to be optional. Given the market dynamics, 
prospective operations may be more efficient in reducing the price volatility than later 
intervention, after the market pressure on price has gathered momentum. As will be 
seen, stabilisation oyer the longer term tends to entail prospective buying or selling con­
siderably in advance of the anticipated adverse price movement. It is here that the in­
sights and foresight of the BS management about the speed of supply response and demand 
adjustments becomes critical. 
628 
The INRA is aimed at partial stabilisation to reduce, rather than to eliminate, price 
variability. For a given BS, the success of partial stabilisation depends on (a) the choice 
of the reference price level and (b) the effective interaction between the BS and market 
price. 
The advantage of partial stabilisation is that the long run trend price about which 
prices are to be stabilised need not be known to the degree of precision required for success­
ful complete stabilisation. In this respect partial stabilisation is easier to operate since 
a higher margin of error is tolerable for the reference price. However, the larger the 
difference between the teference and trend prices, the larger the BS required and the more 
onesided the intervention will be. This raises the question of whether the stabilised price 
bands have to be equidistant from the reference price. 
While INRA specifies equidistant price band about the $ 2,100 reference price, no 
explanation is given for the choice of the price level nor the equidistant principle. For 
a self-liquidating BS, equidistant price bands are valid only when prices are either 
(a) symmetrically distributed about a stationary trend or (b) distributed about a non-statio­
nary trend and skewed to the left (right) in the lower (upper) price range. If these strin­
gent conditions are not met, then a self-liquidating BS need not invoke stabilizing about 
equidistant price bands. Hence the choice of equidistant/non-equidistant price bands 
cannot be determined independently of the expected trend price behaviour. As the 
simulation results will later show, the rubber market dynamics are such that insistence on 
equidistant price band stabilisation would mean higher BS accumulation than necessary 
to defend the floor price. 
On the question of the BS capacity, it is not explained how the 400,000 tonnes " normal 
stocks " for market stabilisation under " normal circumstances " is determined, and the 
definition given in the INRA is tautological. Furthermore, effective interaction between 
the BS and market price involves assumptions about private stockholding behaviour in 
the presence of BS interventions. In general the effect of public storage supply on private 
stockholdings depends on (a) the existing levels of public and private stockholdings, (b) the 
level of financial resources for the supply of public storage and (c) the sizes of (a) and (b) 
relative to the market fluctuations arising from random disturbances (for example, see 
Siamwalla, 1981). Consequently, storage supply may shift between the two sources, 
with the scope for public displacement of private storage generally being proportional 
to the financial resources available, for public storage. 
These considerations of the correct identification of the long run trend price and 
the behaviour of private stockholdings highlight the reliance of BS operations on the 
judgement of the BS manager. In turn this leads to the problems of understanding 
the dynamics inherent in the system, the ability of the BS manager to distinguish reliable 
from false price signals, and the discretionary use of prospective BS operations. Thus 
the problems associated with the feasibility of BS operations and management to attain 
these ends cannot be overemphasised. 
Besides price stabilisation, the INRA is also aimed at the steady growth of export 
incomes of the producing countries. This may be attained (a) by obtaining a higher 
. average price under stabilisation and (b) by inducing more consumption (via input substi­
tution) and production of natural rubber through reduced price fluctuations and 
629 
uncertainty. Since export income growth may be viewed as an income transfer from 
consuming to producing countries, this raises the question of consumers' agreement to 
and participation in such price stabilisation arrangements, bearing in mind the known 
availability of synthetic rubbers. Even with consumers' participation, the compatibility 
of price stabilisation and income growth cannot be assumed. 
Model structure reflecting the world rubber market 
The econometric model of the world natural and synthetic rubber market by Tan 
(1984) is used to simulate the INRA rules. The model is essentially a supply-demand 
model, disaggregated by the main natural rubber producing, and natural and synthetic 
rubber consuming countries. With the synthetic rubber market modelled explicitly, the 
87-equation model can be viewed as consisting of two submodels, one for each group 
of rubber, and interacting through their relative price effects on consumption. The 
causality between natural and synthetic rubber prices in the market is reflected in the 
causal structure of the model which indicates co-determination of synthetic rubber pro­
duction and price. Synthetic rubber production and price then determine natural rubber 
price, production and consumption as well as some consumption of synthetic rubbers. 
The remaining synthetic rubber Consumption is then determined from which total synthetic 
rubber consumption and stocks are determined. In summary, natural rubber price for­
mation in the historical period has been causally dependent on synthetic rubber price, 
and co-determined with natural rubber production and natural as well as synthetic rubbers 
consumption. 
Several assumptions are implicit in using the model to simulate BS operations a la 
INRA. These are the continued validity of the existing market structure as represented 
by the model, and of the leading role played by the London market in natural rubber 
price formation. For in the historical period, natural rubber prices in the various primary 
and terminal markets have been determined ultimately by the London spot price which, 
to a large extent, was influenced by variations in the natural rubber stockholdings in the 
consuming regions. Thus the behaviour of private stockholdings in the consuming regions 
has a bearing on the potential impact of BS operations since the stabilisation programmes 
could affect private stockholding behaviour. 
The reasons for a possible reduction in the influence of the London spot price in 
future are that the oil crises and ensuing higher synthetic rubber prices have led to greater 
reliance on natural rubber supply through long term contracts. The potential shift in 
focus to other markets is also suggested by the recent establishment of a commodities 
exchange in Kuala Lumpur and by the fact that Singapore, an active centre for natural 
rubber trading, is not party to INRA. Consequently, the longer term reactions and spe­
culative activities of Malaysian and Singapore traders to BS operations remain to be seen 
and thus have not been incorporated in the simulations presented here. (In terms of 
model simulations, a different trading behaviour would necessitate a change in the values 
of the parameters used in the stockholding equations). Finally, it is assumed that the 
BS manager operates with sufficient financial resources. 
To understand the dynamics of the world rubber market, model simulations of the 
BS rules were conducted. The basis for simulating BS operations is the RSS1 - grade 
630 
price of natural rubber. The findings from ex post and ex ante simulations of BS operations 
with prospective buying and selling will now be discussed. 
Simulations of BS operations for price stabilisation 
This section discusses the results of two cases of ex-post stabilisation (from 1970 
to 1980) and two of ex-ante stabilisation (from 1985 — 1995). The 1970 — 80 decade 
was marked by, amongst other events, the sharp rise in oil price. This bears directly ori 
the natural rubber market through higher synthetic rubber prices and lower growth of the 
transport sector which is a major outlet for natural rubber. It is therefore instructive to 
appraise the schedule of BS operations that would have been required to stabilise price 
during this eventful period (Phase III) when, as can be seen from Fig. 2 , the long term 
natural rubber price trend reversed, and the amplitude of price fluctuations widened 
from those during the previous decade of low and stable oil price (for further details see 
Tan, 1984 : 6.6). 
Case 1 : BS operations during 1970— 1980 were simulated to stabilise price along the 
long term trend price with unchanged private stockholding behaviour. The actual and 
the stabilised price paths over time are shown in Fig. 3 with the levels of BS purchases 
(positive values) and sales (negative values) shown against their respective intervention 
years. Over the stabilisation period the BS holding will average 65,000 tonnes. The 
effects of this stabilisation programme, measured in terms of the average volumes and 
values of production and consumption and their instability indices are presented in Tables 
1 ( a ) - ( c ) . 2 
At the global level, Table 1(a) shows that stabilisation along the long term trend price 
will increase the average and total natural rubber output during the period, but wiil lower 
the average and total natural rubber consumption. The market intervention will also 
reduce both production and consumption instability. While stabilisation will reduce 
the instability indices for export earnings and consumption expenditures, the results also 
show that the higher production will yield higher export earnings but that the lower con­
sumption will cost more. The simultaneous increases in export earnings and consumption 
expenditures reflect the scope for income transfer from the consuming to the producing 
countries. This annual income transfer of US $ 92 million from price stabilisation is 
about 3 % of the average annual consumption outlay. Assuming no change in the private 
stockholding behaviour, Table 1(a) shows that while private stockholdings in the producing 
regions and afloat (on board ships) arc reduced under intervention, the average private 
stockholdings in the consuming countries will increase as will its level of instability. 
Table 1 (b) presents the stabilization effects on the major producing countries. Except 
for Indonesia and Thailand, the stabilisation programme is seen to benefit the producing 
countries by reducing the instability of production and export earnings. For Indonesia 
production instability is not improved by the stabilisation programme while price stabili­
sation will lead to lower output and export earnings for Thailand. 
(2) The instability index used here is the UNCTAD index, given by 
Z | Y - Y | / Y 
i • 
where Y and Y are actual and trend values, respectively 
i i 
631 

'3000-J 
;.6co J 
1 8 0 0 
R£SI-Price 
f.o.b. Singapore 
(SS/conne) 
1400 J 
1000 J 
Without Stabilisation 
1,'oper Intervention Fries 
With Stabilisation 
Beference Price 
Lower Intervention Price 
+180 +50 BS Operations (Th. Tonnes> 
BS operations with unmodified private stockholding behaviour. 
-80 *180 -60 -140 
1976 1970 1972 
Fig. 3 . 
Table 1(a), Summary of effects ofprice stabilization around the long term trend price, 1970 —1980 
Average level under 
Variable 
World NR production 
(thousand tonnes) 
World NR consumption 
(thousand tonnes) 
NR stocks in producing 
regions and afloat 
(thousand tonnes) 
NR stocks in consuming regions 
(thousand tonnes) 
World NR export earnings 
(US 8 million) 
World N R consumption 
expenditures 
(US S million) 
Free 
market 
3512 00(0 0055) 
3518-50(0-0046) 
788-65(0-0028) 
764-50(0-0018) 
2,541-30(0-2156) 
2,927-00(0-2356) 
World NR real export earnings 
(million 1983 US 8) 3,431-50(0-0420) 
World NR real consumption 
expenditures 
(million 1983 US $) 
Market 
intervention 
3543 -40(0 0028) 
3445-10(0-0039) 
743-69(0-0010) 
776-54(0-0107) 
2,633-30(0-1401) 
2,935-10(0-1662) 
2,643-40(0-0054) 
3,929-20(0-0465) 4,008-40(0-0141) 
Change in 
average 
level due to 
market 
intervention 
31-39 
—73-40 
—44-96 
12-04 
92 00 
8-10 
211-90 
Instability index under 
Free Market 
market intervention-
79-20 
0-6207 
0-5662 
0-4669 
0-3362 
3-695 
3-8949 
1-6928 
1-7857 
0-4522 
0-5192 
0-2573 
0-8923 
3-1929 
3-5142 
0-5540 
0-9199 
Note :— Figures within parentheses denote coefficients of variation of the distributions of values from which the cor­
responding averages are derived. 
Table 1(b). Summary of price stabilization effects on major producers, 1970 — 1980 
Country 
Africa 
Brazil 
Indonesia 
Malaysia 
Estates 
Smallholders 
Sri Lanka 
Thailand 
Average natural rubber output (a) 
Without stabilization 
Instability 
Volume 
207-14(0-0045) 
23-08(0-0145) 
866-50(0 0055) 
632-64(0-0005) 
804-13(0:0175) 
145-71(0-0032) 
412-23(0-0263) 
index 
0-5075 
0-9977 
0-5717 
0-4508 
1 0356 
0-4928 
1 • 3643 
With stabilization 
Instability 
Volume 
209-54(0-0036) 
24-11(0-0849) 
905-56(0-0051) 
638-18(0 0018) 
803-64(0 0091) 
145 00(0 0009) 
386-53(0-0191) 
index 
0-4999 
0-7404 
0-6177 
0-3293 
0-8300 
0-2272 
1-1741 
Average natural rubber export earnings (b) 
Without stabilization 
Instability 
Value 
168-98(0-1464) 
18 00(0-2923) 
682-06(0-2692) 
393-75(0-0809) 
481-60(0-1732) 
113-92(0-1879) 
337-37(0-3501) 
index 
2-7947 
4-3539 
4 0713 
2-3215 
3-4162 
3-5336 
4-8097 
With stabilization 
Instability-
Value 
174 03(0 1000) 
19-55(0-1922) 
726-53(0-1640) 
405 16(0 0510) 
509-12(0-1135) 
116-15(0-1135) 
322-67(0-2286) 
index 
2-5890 
3-8623 
3-5023 
1-8182 
2-8064 
2-8551 
4-1250 
Notes: (a) Average natural rubber output volume is given in thousand tonnes. 
(b) Average natural rubber export earnings value is given in million US dollars. 
Figures within parentheses denote coefficients of variation of the distribution of values from which the corresponding averages are derived. 
Table 1(c). Summary of price stabilization effects on major consumers, 1970 —1980 
Average natural rubber consumption (a) Average natural rubber consumption expenditure (b) 
Country Without stabilization 
Instability 
Volume index 
With stabilization 
Instability 
Volume index 
Without stabilization 
Instability 
Value index 
With stabilization 
Instability 
Value index 
France 165-85(0-0029) 0-4092 165 07(0 0023) 0-4104 140-37(0-2339) 3-7217 141-21(0-1569) 3-4374 
Fed. Rep. of 
Germany 190-87(0 0021) 0-3970 188-180(0 0006) 0-2121 157-45(0-1554) 3-0374 157-54(0-1406) 3-1236 
Italy 124 30(0 0037) 0-5066 122-49(0-0008) 0-2387 104-22(0-2057) 3-6010 104-15(0-1419) 3-2420 
Japan 333-32(0 0165) 1 0424 332-65(0-0099) 0-8634 267-92(0-3384) 4-6689 273-82(0-2067) 3-9378 
United Kingdom 162-23(0 0133) 0-9763 160-10(0-0087) 0-7901 128-65(0 1029) 2-4969 130-74(0-0775) 2-3877 
United States 695 08(0 0093) 0-8299 684-36(0-0073) 0-7508 610-33(0-1779) 3-4847 618-77(0-1411) 3-2205 
Notes : (a) Average natural rubber consumption volume is given in thousand tonnes. 
(b) Average natural rubber consumption expenditure is given in million US dollars. 
Figures within parentheses denote coefficients of variation of the distributions of values from which the corresponding averages are 
Table 1 (c) presents the stabilisation effects on the major consuming countries and 
show that countries benefit differentially. While stabilisation will lower global consump­
tion, consumption in Japan, U K and the US are seen to increase. For the majority o f 
countries, lower instability indices for consumption volumes and values are obtained. 
Some insights into the question of stabilisation about equidistant or non-equidistant 
price bands can also be gleaned from Fig. 3. The behaviour of price in the historical 
period is such that for the BS to successfully defend the ceiling price during 1979 — 1980, 
the BS purchases in 1971 and 1975 have to be higher than warranted by the floor support 
price. Thus the resulting price path could equally have been an outcome of stabilisation 
about non-equidistant price bands. 
Case II: Since it is known that private stockholding behaviour can change in the presence 
of a stabilisation programme, an assessment of the sensitivity of stabilisation effects to 
an assumed change in private stockholding behaviour is attempted. In line with the 
model structure of consuming region stocks playing a key role in the London spot price 
formation, the assumption in this simulation is that stockholdings in the consuming regions 
are shifted out partially. Using the accumulated BS of about 270,000 tonnes which 
approximates 30% of the stocks in the consuming regions in recent years as a guideline, 
the consuming region's stocks will be adjusted downwards by a factor of 0 3 to proxy 
the shift of stockholding burden to the INR.O. 
Fig. 4 illustrates the actual and stabilised price paths and the attendant schedule 
of BS operations. In comparison with the BS operations in Case I, the stabilised price 
path is smoother with the average BS holding of 80,000 tonnes, and the BS authority, 
will end the stabilisation period with a positive balance of 50,000 tonnes stocks in 1995. 
Furthermore, although the schedule of BS operations may differ by as much as 100,000 
tonnes from those in Case I for each corresponding year, the mid-period BS holdings 
in 1975 are identical at 200,000 tonnes. 
Tables 2 (a) — (c) quantifies the stabilisation effects. Table 2 (a) reveals that the global 
stabilisation effects are the same qualitatively as in Case I. Quantitatively, however, the 
Case II stabilisation yields slightly higher total output and total export earnings than for 
Case I, with slightly lower world consumption volume and at higher cost. As for Case I 
the price stabilisation programme effectively stabilises export earnings. 
The major producing countries benefit from stabilisation in terms of higher output 
and export earnings, both with reduced instability. However, intervention reinforces 
the Indonesian production instability and Thai export earnings instability. For the major 
consuming countries the stabilisation effects in Case II are similar to those for Case I, 
both qualitatively and quantitatively. 
These simulation results vindicate the hypothesis that the natural rubber market 
instability stems in part from the stockholding behaviour in the consuming regions. 
Case III: Given that the INRA is scheduled for renegotiation in 1985 from the likely 
position of having an initial BS of about 270,000 tonnes (assuming no BS releases bet­
ween August 1984 and October 1985), ex-ante simulations with the model can be used to 
explore alternative schedules of BS operations to stabilise an assumed scenario for the 
coming decade. 
637 
Variable 
Table 2 (a). Summary of effects of price stabilization around the long term trend price, 1979 —r-1980 
Average level under 
Free 
market 
World NR production 
(thousand tonnes) 
World NR consumption 
(thousand tonnes) 
NR stocks in producing 
regions and afloat 
(thousand tonnes) 
N R stocks in consuming regions 
(thousand tonnes) 764- 50(0 0018) 
World NR export earnings 
(US S million) 
3,512-00(0-0055) 
3,518-50(0-0046) 
788-65(0-0028) 
World NR consumption 
expenditures 
(US S million) 
2,541-30(0-2156) 
2,927-00(0-2356) 
World NR real exoort earnings 
(million 1983 US S) 3,431 • 50(0 • 0420) 
World NR real consumption 
expenditures 
(million 1983 US $) 
Market 
intervention 
3,553-70(0-0029) 
3,438-60(0-0038). 
746-27(0-0011) 
756-87(0-0078) 
2,681-20(0-1424) 
2,983 00(0-1663) 
3,706-00(0-0068) 
3,929-20(0 0465) 4,073-8O(0 0147) 
Change in Instability index under 
average level 
due to market Free Market 
intervention market intervention 
41-79 0-6207 0-4519 
—79-90 0-5662 0-5223 
—42-38 
—10-37 
139-90 
56-00 
274-50 
144-60 
0-4669 
0-3362 
3-6950 
3-8949 
1-6928 
1-7857 
0-2523 
0-7515 
3-2016 
3-5211 
0-7329 
10013 
Note : Figures within parentheses denote coefficients of variation of the distribution of values from which the corres­
ponding averages are derived. 
\ 
Table 2 (b). Summary of price stabilization effects on major producers, 1970 —1980 
v o 
Country 
Africa 
Brazil 
Indonesia 
Malaysia 
Estates 
Smallholders 
Sri Lanka 
Thailand 
Average natural rubber output (a) 
Without stabilization 
Volume 
207-14(0-0045) 
23-08(0-0145) 
866-50(0 0055) 
632-64(0-0005) 
804-13(0-0175) 
145-71(0-0032) 
412-23(0-0263) 
Instability 
index 
0-5075 
0-9977 
0-5717 
0 4508 
10356 
0-4928 
1-3643 
With stabilization 
Volume 
209-83(0-0035) 
24-29(0-0081) 
910-66(0-0053) 
640-01(0-0017) 
803-67(0-0092) 
144-87(0-0008) 
386-77(0-0196) 
Instability 
index 
0-5008 
0-7054 
0-6102 
0-3235 
0-8240 
0-2286 
11789 
Average natural rubber export earnings (b) 
Without stabilization 
Value 
168-98(0-1464) 
18-00(0-2923) 
682 06(0-2692) 
393-75(0-0809) 
481-60(0-1732) 
113-92(0-1879) 
337-37(0-3501) 
Instability 
index 
2-7947 
4-3539 
4 0713 
2-8215 
3-4162 
3-5336 
4-8097 
With stabilization 
Value 
177-41(0-1011) 
20-10(0-1963) 
744-80(0-1675) 
410 03(0 0501) 
512-26(0-1161) 
118-09(0 1135) 
329-14(0-2335) 
Instability 
index 
2-5827 
3-8788 
3-5105 
1-7276 
2-8040 
2-8287 
4-1710-
Notes : (a) Average natural rubber output volume is given in thousand tonnes. 
(b) Average natural rubber export earnings value is given in million US dollars. 
Figures within parentheses denote coefficients of variation of the distribution of values from which the corresponding averaees 
are derived. & 
Table 2 (c). Summary of price stabilization effects on major consumers, 1970 —1980 
Country 
France 
Fed. Rep. of 
Germany 
Italy 
Japan 
United Kingdom 
United States 
Average natural rubber consumption (a) 
Without stabilization 
Volume 
165-85(0 0029) 
190-87(0 0021) 
124-30(0-0037) 
333- 32(0 0165) 
162-23(0-0133) 
695-08(0-0091) 
Instability 
Volume index 
0-4092 164-56(0-0023) 
0-3970 187-51(0-0006) 
0-5066 121-89(0-0007) 
1-0424 332-05(0-0100) 
0-9763 159-03(0-0089) 
0-8299 683-03(0-0074) 
With stabilization 
Instability 
index 
0-4069 
0-2099 
0-2322 
0-8696 
0-7769 
0-7618 
Average natural rubber consumption expenditures (b) 
Without stabilization 
Instability 
Value 
140-37(0-2339) 
157-45(0-1554) 
104-22(0-2057) 
267-92(0-3384) 
128-65(0-1029) 
610-33(0-1779) 
index 
3-7217 
3-0374 
3-6010 
4-6689 
2-4969 
3-4847 
With stabilization 
Instability 
Value 
143-25(0-1595) 
161-25(0-1187) 
105-47(0-1441) 
278-45(0-2107) 
132-10(0-0784) 
627-82(0-1397) 
index 
3-4608 
2-9223 
3-2517 
3-9620 
2-3423 
3-230Q. 
Notes: (a) Average natural rubber output volume is given in thousand tonnes. 
(b>
 ^^SSSSS^S^ S S S i n of values from which the corresponding averages are derived. 
RSSI-Prlce 
f . o . b . Singapore 
(SS / tonne) 
RSSI-?rice — 
f-o.b. Singapore 
Fig. 5. BS operations for price stabilisation, 1985 -1995. 
The scenario employed here assumes GNP growth in the industrial countries averaging 
2 • 8 % during the 1984 — 86 period and 3 • 2 % during the 1987 — 1995 period. As Fig. 5 
for this scenario shows, the natural rubber price will fall from 1986 through to 1989 and 
rise from 1990 to 1995. This projected price behaviour can be explained largely in terms 
of the dynamics of supply response in rubber production. Apart from having a gestation 
period of some 5 to 8 years, the yield of rubber trees tends to peak about 5 to 6 years 
after first production so that the productivity of a tree becomes maximal about 10 — 14 
years after planting. The low prices projected for the second half of the 1980s reflect 
the lagged supply response to the high rubber prices in the early 1970s and the expectation 
soon after the oil crisis that synthetic rubbers had lost their price advantage. Since natural 
rubber competes with synthetic rubbers, the low prices in the 1980s will induce material 
input substitution favouring natural rubber as its price fall in relation to those of synthetic 
rubbers. As natural rubber consumption increases while its supply is inelastic in the 
short to medium term, the tightening of the natural rubber market causes price to rise 
again. Assuming the industrial countries continue to grow, the continued increased 
demand for natural rubber will reinforce the tightening market and increase price further. 
Fig. 5 shows two alternative schedules of non-stochastic ( 3 ) simulations of BS 
operations to stabilise price with the initial BS of 1985 being liquidated by 1995. To 
support the floor price the BS management will need to add onto the existing stockpile 
by buying about 80,000 tonnes annually in 1987 and 1988. By 1992 the BS releases will 
be required to defend the ceiling prices through to 1995. Two stabilised price paths are 
shown for the 1992 — 1995 period to illustrate the impact of different rates of BS releases. 
Tables 3 (a) — (c) summarises the effects of a stabilisation programme ending on 
the lower price path in Fig. 5. On this path the average BS holding is 280,000 tonnes. 
Qualitatively and quantitatively, the results indicate that the effects are similar 
to those for Cases I and II. More importantly, they show that the non-release of the BS 
during late 1983 and early 1984 (when the market price exceeded the upper intervention 
price) was appropriate to the stabilisation strategy about the long term trend price of 
this scenario. However, it should be mentioned that BS operations also incur operational 
(carrying and rotation) and administrative costs. These costs must be considered in an 
overall evaluation of the net result of BS price stabilisation. Where the average BS 
holding during the stabilisation is high, as is the case here, it may be envisaged that gains 
from stabilisation may be offset by the BS operation outlays. 
Caveats for successful stabilisation 
The three main results of ex post and ex ante simulations of the BS operations may 
now be summarised. 
(3) The simulations reported here are non-stochastic in nature. Although stochastic simulations are 
preferable for realism, the use of non-stochastic simulations does not detract from the findings because 
the model used here is annual in nature. Since an annual model is essentially a long term model, the 
non-stochastic simulations permits one to concentrate the analysis on the dynamic effects of market 
interventions arising from the long lagged structure of the model. However, the INRA intervention is 
based on daily market prices and all the constraint clauses are tied to these prices. Since these daily 
price variations are assumed away in non-stochastic simulations, it is obvious that the problems of 
price stabilisation are understated here. In this sense, the non-stochastic approach may be seen to 
provide the most favourable test of the INRA proposals. 
643 
Table 3 (a). Summary of effects of price stabilization around the long term trend price, 1983 — 1995 
Average level under 
Variable 
World NR production 
(thousand tonnes) 
World NR consumption 
(thousand tonnes) 
NR stocks in producing 
regions and afloat 
(thousand tonnes) 
Free 
market 
5,281 -20(0 0244) 
5,206-60(0-0228) 
770-24(0-0007) 
NR stocks in consuming regions 
(thousand tonnes) 1,187-30(0-0572) 
World NR export earnings 
(US $ million) 
World NR consumption 
expenditures 
(US S million) 
6,135-20(0-1121) 
6,691-30(0-1232) 
World NR real export earnings 
(million 1983 US $) 3,889-40(0 0168) 
World NR real consumption 
expenditures 
(million 1983 US $) 
Market 
intervention 
5,297-80(0 0226) 
5,172-90(0-0221) 
769-39(0-0014) 
1,171 -60(0 0674) 
6,144 00(0 0630) 
6,631-80(0-0697) 
3,944-50(0 0045) 
4,237-20(0-0220) 4,254-80(0-0067) 
Change in Instability index under 
average level 
due to market Free Market 
intervention market intervention 
16-60 1-7410 1-7112 
—33-70 1-7392 1-6886 
— 0-85 
—15-70 
8-80 
—59-50 
55-10 
17-60 
0-2744 
2-8089 
3-7541 
3-9803 
1-5140 
1-7625 
0-4279 
2-8938 
2-8456 
3 075 
0-7275 
0-8890 
Figures within parentheses denote coefficients of variation of the distribution of values from which the corresponding 
averages are derived. 
Table 3 (b). Summary of price stabilization effects on major producers, 1983—1995 
as 
Country 
Africa 
Brazil 
Indonesia 
Malaysia 
Estates 
Smallholders 
Sri Lanka 
Thailand 
Average natural rubber output (a) 
Without stabilization 
Instability 
Volume 
372-66(0-1578) 
33-43(0-0109) 
1,301-20(0 0225) 
711-98(0-0182) 
1,256-90(0 0197) 
140-43(0-0020) 
826-38(0-0335) 
index 
4-7026 
1-2023 
1-5371 
1-4203 
1•6282 
0-4177 
2-1201 
With stabilization 
Instability 
Volume index 
4-6922 
1-5785 
1-5758 
374-42(0-1561) 
34-56(0-0179) 
1,309 00(0-0205) 
716-82(0 0180) 
1,250-20(0 0170) 
138-36(0 0016) 
823-88(0-0324) 
1-4227 
1-4840 
1-4227 
2-0609 
Average natural rubber export earnings (b) 
Without stabilization 
Instability 
Value 
505-99(0-3283) 
41-40(0-0912) 
1,618-60(0-1143) 
723-08(0-0574) 
1,236-00(0-0736) 
173-67(0-0571) 
1,040-90(0-1517) 
index 
6-6490 
3-5139 
3-6049 
2-6636 
3-1290 
2-6646 
4-4199 
With stabilization 
Instability 
Value 
502-58(0-2414) 
43-06(0-0663) 
1,625-70(0-0547) 
728-93(0-0377) 
1,231-30(0-0439) 
170-21(0-0217) 
1,033-00(0-0892) 
index 
5-8599 
3-0177 
2-5331 
2 0910 
2-3856 
1-4888 
3-4681 
Notes : (a) Average natural rubber output volume is given in thousand tonnes. 
(b) Average natural rubber export earnings value is given in million U S dollars 
Figures within parentheses denote coefficients of variation of the distribution of values from which the corresponding averages are derived. 
O-i 
Country 
Table 3 (c). Summary of price stabilization effects on major consumers, 1983 —1995 
Average natural rubber consumption (a) Average natural rubber consumption expenditures (b> 
Without stabilization 
Instability 
Volume index 
With stabilization 
Instability 
Volume index 
Without stabilization 
Instability 
Value index 
With stabilization 
Instability 
Value index 
France 219 61(0 0117) 1 2990 217-86(0 0126) 1 2709 285-47(0-0874) 3 3857 283-85(0-0508) 2-6358 
Fed. Rep. of 
Germany 191-90(0-0015) 0-4667 190-73(0-0010) 0 3711 248-63(0 0558) 2 6423 250-02(0 0239) 1-7771 
Italy 190-44(0 0182) 1 5887 187-90(0 0171) 1 5051 248-91(0-1054) 3 7374 245-47(0-0600) 2-8600 
Japan 633-41(00371) 2 1855 630-97(0 0368) 2 1559 799-22(0-1620) 4 5372 791-65(0-0955) 3-5981 
United Kingdom 155-33(0-0105) 1 1863 153-97(0 0116) 1 1813 201-99(0-0858) 3 2963 200- 63(0 0495) 2-5568 
United States 978-36(0-0172) 1 5057 970 -90(0 0165) 1 4336 1,306-70(0 0982) 3 5587 1,297-20(0-0565) 2-7144 
Notes : (a) Average natural rubber output volume is given in thousand tonnes. 
(b) Average natural rubber export earnings value is given in million U S dollars. 
Figures within parentheses denote coefficients of variation of the distribution of values from which the corresponding averages are derived. 
First, the simulations show prospective BS operations to be an essential part of success­
ful partial price stabilisation. This arises from the need to consider the lagged effects 
in a dynamic system. Second, the simulations show that the stabilisation results are 
critically dependent on the appropriate choice of the reference price. Moreover, the 
asymmetry in natural rubber supply response to price change, ( 4 ) and longer distributed 
price lags on the supply side than on the demand side of the rubber market are such that 
more BS purchases than warranted for floor price support are required if the same BS 
is to be able to defend the ceiling price. The resultant stabilisation becomes, in effect, 
about non-equidistant price bands. Finally, price stabilisation reduces production and 
consumption instability for the major countries. While price stabilisation induces income 
(export earnings) stabilisation, it also provides scope for income transfer from the consu­
ming to the producing countries, albeit in an uneven way. 
From the simulation results, two caveats for successful BS operations are in order. 
These concern prospective BS operations and the INRA constraint on the admissible net 
change in BS volume within 18 months. 
The efficacy of the prospective BS purchases/sales is contingent on foresight about 
the behaviour of business cycles and the natural rubber price in future periods. As Fig. 1 
shows, the level and stability of synthetic rubber price has an important influence on 
the level and volatility of natural rubber price. This role.of synthetic rubber price could 
therefore be explicitly incorporated into the INRA clauses governing price revisions so 
as to enhance the automaticity of the price revision rules. The importance of foresight 
for successful stabilisation highlights the critical reliance on the judgement and ability, 
to take timely action, of the BS manager. As Nurkse (1958) and Benoit (1959) have 
indicated, BS operations could result in investing excessive arbitrary authority on the 
BS manager. 
In all the simulations reported, the constraint of net change in BS not exceeding 
100,000 tonnes within 18 months was binding. In the case of ex post simulations, the 
constraint was binding in 1971 and 1975 when price support was necessary. Similarly, 
the ex ante simulation indicates that the constraint will be binding because of the conse­
cutive support buying in 1987 and 1988. In practice, the constraint became binding only 
4 months after the initial BS buying began and an 1NRC meeting had to be convened 
in March 1982 to lower the reference price and stabilisation price bands. By September 
1982, only 19 months after BS began, the net change in BS had reached 200,000 tonnes 
and an INRC meeting had to be convened again. This experience, together with the 
simulations results presented in this paper, suggest relaxing the constraint on net BS 
change within 18 months to about the 200,000 tonnes level. In addition to reinforcing 
the efficacy of the BS purchases, this relaxed constraint would also reduce the need for 
frequent and costly renegotiations. 
In concluding, it is interesting to mention that the problem of the distributed lagged 
price effects for INRA is similar to that for the International Tin Agreement which Smith 
and Schink (1976) had ascertained. The simulations have shown that it is these lagged 
(4) This asymmetry stems from the nature of rubber production. When price remains too low, tapping 
of the tree and hence production can be stopped instantly. However, when price increases sharply, 
output can be increased only in so far as slaughter tapping permits, assuming that all the mature trees 
are being tapped. 
647 
price effects that primarily call for the larger BS accumulation than is permissible under 
the present INRA. This demand on INRA resources may be reduced if export controls, 
a6 is now being suggested, are introduced as an additional instrument for stabilisation 
under INRA. However, the implementation of export controls without concurrent 
production controls would mean imposing an additional burden on governments to hold 
national stockpiles and therefore reduce the gains from stabilisation to the producing 
countries. If production controls are concurrently introduced, then the problem of 
further distorting the supply response will have to be considered. 
A P P E N D I X 
Main features of the rubber model used in simulating INRA 
The rationale for the joint modelling of the natural and synthetic rubbers market 
stems from their joint derived demand. In the last two decades, about 65 — 70 percent 
of this stems from the production of tyres, tubes and automative parts for the transport 
sector. The remaining 30 — 35 percent of demand derives from the production of 
producers' goods such as conveyor belts to consumers' goods such as footwear and 
gloves. This joint demand gives rise to substitution, and hence competition, between 
the two types of rubber. 
In tyre production, there exists minimum requirements of each type of rubber input 
so that input substitution due to relative price changes occurs only in the range beyond 
these minimum requirements. However, adjustment costs and the known volatility of 
natural rubber price delimit the extent of price-induced substitution in the short run. In 
the former, adjustment costs incurred could offset savings from input substitution. This 
i6 reinforced by the natural rubber price instability which restrains prompt substitution 
whenever prices change. Consequently, substitution occurs gradually after the relative 
price change is perceived as sustainable. 
Assuming cost minimization, the input demand functions takes the following forms 
CN - fi { (PN/PS), ( P N / P S K , . . . . (PN/PS)- n , Q, T> 
CS = f2 { (PN/PS), ( P N / P S ) - I , , • • ( P N / P S K . Q> T> 
where CN and CS are the consumption of natural rubber and synthetic rubbers respectively, 
(PN/PS) is the relative price of natural rubber to synthetic rubbers, Q is the production 
of rubber goods (such as tyres and tubes in the transport sector) and T is a trend variable 
to proxy technological change in production. The presence of the distributed relative 
price lags reflect the " habit persistence " restraining prompt rubber input substitution 
upon relative price changes. 
Separate equations for rubber input demand by the transport and nontransport 
sectors were estimated for each of the main consuming countries. Where disaggregated 
data are absent, a single equation wa6 estimated for each type of rubber consumed. 
648 
The differences in their natures of production, marketing and organizational struc­
tures call for different approaches to the modelling of natural rubber and synthetic rubbers 
supplies. Natural rubber is a perennial crop produced under conditions approximating 
perfect competition. In contrast •, synthetic rubber is a manufactured product based on 
petrochemical feedstocks. Furthermore, the industry is vertically integrated to a large 
extent, backwards with the petrochemical industry and forwards with the tyre industry. 
Hence the industry operates under imperfect competition; 
The problems of modelling the supply response of perennial crops are well known 
and stems from the long gestation and productive lifespan periods which introduce long 
lagged effects into the system. To date, one of the more sophisticated models on perennial 
crop supply is that by Wickens and Greenfield (1973) which takes into account the invest­
ment feature of perennial crop supply. Adopting the Wickens and Greenfield approach, 
the supply of natural rubber was estimated in the following form from first-differencing of 
the reduced form :— 
k * 
q = S t p + (1 + X)p - X q (3) 
'it. i = o i t-i t-1 t -2 
where q is natural rubber supply, P is the price of natural rubber, and k is the number 
of lags which typically extends from 10 to 14 years. The coefficient t*j is a composite 
function of (a) the yields of trees of various ages, and (b) the price coefficients of different 
periods Which goes to determine the output response (Wickens and Greenfield, 1973). 
The equations for natural rubber stocks and spot price formation are based oh the 
hypothesis that price formation is due to disequilibrium demand for stocks in the consuming 
regions. Since the data for stocks located in the consuming regions are unsatisfactory, 
it was decided to estimate the price equation directly. The spot price equation is written 
as : 
(PNs/WPl u k ) = f { A +(SNC/CN), ASNG, (PN'/WPT*) - y (4) • 
(PNVWPT*)-, DV(1), DV(2), DV(3) > 
where PN"is the London spot price, W P I u k i s the U K Wholesale Price Index, A(SNC/CN) 
is the change in the ratio of stocks in consuming regions to total consumption, ASNG 
is the change in government stockpiles, DV(1) is a dummy variable for the racial riots 
in Malaysia in 1969, DV(2) is the dummy variable for the oil crisis in 1973/74, and DV (3) 
is the dummy variable for the 1973 recession and the peak of the Malaysian " Crash 
Programme" of production controls to prevent further price declines. 
Stocks held in the producing countries (SNP) and afloat (SNA) are assumed held 
for precautionary reasons, and are estimated together in. the form (SNP -+• SNA) >= f 
{ A (CN/QN), (SNP + SNA) - x } where. A (CN/QN) is the change in the ratio of total 
consumption to total production. Thus, as this ratio increases, more stocks will be 
shifted from the producing regions and afloat to the consuming regions. The 6tocks in 
the consuming regions are then treated in the balance equation. 
649 
The modelling of synthetic rubber supply is based on the dependence of the synthetic 
rubber industry on the oil industry for its raw material inputs and on the tyre industry 
for the bulk of its demand. The upshot o f these interrelations is the observed vertically 
integrated oligopolistic structure of the industry which facilitates intra-firm pricing and 
price discounting. Such closed trading also harbours the synthetic rubber industry to 
some extent from the general market instabilities. 
Assuming cost-plus pricing for synthetic rubber, the average price of synthetic rubbers 
(PS) is determined by the average cost of synthetic rubber production and the level of 
mark-up per unit of synthetic rubbers sold. As the general-purpose synthetic rubbers 
compete with natural rubber, the mark-up is hypothesized as being determined by the 
level of natural rubber price (PN). Since the average cost consists of average fixed costs 
(which in an industry with scale economies is determined by the production capacity 
CQRS) and average variable cost (which is dependent on the price of raw material inputs, 
chiefly oil) the average synthetic rubber price is given by : 
PS = f4 {CQSR, P o i l, P N } (6) 
where P o i l is the price of oil. 
Similarly, synthetic rubber supply is some form of 
QSR = f6 {CQSR, P o i l, P N } (7) 
The model is closed with balance equations for the stocks of natural rubber and 
synthetic rubbers. In addition', there are several price linkage equations, linking natural 
rubber prices in the primary and terminal markets to the London spot price, and linking 
various countries' synthetic rubber prices to the average synthetic rubber price that is 
estimated. 
REFERENCES 
BENOIT, EMILE (1959). Purchase guarantees as a" means of reducing instability of 
commodity export proceeds of underdeveloped countries, Kyklos, 1 2 , 300 — 306. 
N U R K S E , RAGNAR (1958). Trade fluctuations and buffer policies of low-income 
countries, Kyklos, 1 1 , 141 — 154. 
SIAMWALLA, AMMAR (1981). The analytics of competitive storage of agricultural 
commodities with rational expectations." (mimeograph). 
SMITH, GORDON W. AND GEORGE R. SCHINK (1976). The international tin agreement : 
A reassessment," Economic Journal, 86, 715 — 728. 
T A N , C SUAN (1984). World rubber market structure and stabilisation, Commodity 
Working Paper No . 10, The World Bank. 
U N C T A D (1979). Preparation of an International Natural Rubber Agreement — 
Note by the U N C T A D Secretariat, TD/RUBBER/R.5, 27 April, 1979. 
WICKENS, M. R. AND J. N . GREENFIELD (1973). The Econometrics of Agricultural Sup­
ply : An. Application to the World Coffee Market, Review of Economics and Statistics, 
55, 4 : 433 — 4 4 0 . 
650 
COMPLETING THE CENTURY: LONG RUN PRODUCTION 
CYCLES AND THE SRI LANKA RUBBER INDUSTRY OVER 
THE NEXT 25 YEARS 
By 
D. M . ETHERINGTON* AND W. M . PREMACHANDRA** 
(* Australian National University, Australia 
**National Institute of Plantation Management, Sri Lanka) 
ABSTRACT 
Data collected for the Rubber Industry Master Plan Study are used to examine the 
implications of alternative replanting strategies for the productive capacity of the rubber 
industry over the next quarter of a century. Five strategies are simulated, including one 
based on farmers declared intentions. The starting year for the simulations is 1978 so that 
it is possible to begin to trace the beginnings of the most likely path. It is concluded that 
Sri Lanka's rubber output will decline until about 1998 before commencing to increase. The 
study also suggests that there is a substantial gap between productive capacity and actual 
output levels. 
INTRODUCTION 
All tree crops have certain features in common, irrespective of whether their commer­
cial product is fruit, fodder, building material, firewood or an industrial raw material. 
Like man, trees take time to mature and usually live from one to many decades. The 
rubber tree is unique among, the important tree crops in that the current techniques used 
to obtain the main product results in the gradual destruction of the tree. We might 
say then that the process of rubber production involves the deliberate killing of the tree 
by the necessity of ring barking it. 
For an individual tree of a particular clone, in a given environment on a given soil, 
the current flow of production is determined by the age of the tree, how well it has been 
maintained, the rate at which its bark has been removed and the rate at which it is currently 
being removed. For a particular farmer (estate, country), the current flow of rubber 
production is determined by the existing stock of trees of each clone of different ages 
multiplied by the clone/age specific yields. Thus current output is constrained by past 
planting and exploitation decisions. In the same way future output will be constrained 
by present planting and exploitation decisions. 
. A baby boom in a human population results in a cohort' bulge ' that gradually works 
its way up the demographic pyramid. . In so doing it has ramifications on demand patterns, 
incomes and dependency ratios. In parallel fashion planting booms in tree crops work 
their way up the age profile with implications for production and processing. To change 
the analogy, we might say that past .planting decisions throw their ' waves ' into the future 
and it is important for marketing and national planning policy to be able to forecast 
such long run cyclical patterns. . 
The purpose of this paper is to demonstrate the methodology of forecasting and to 
examine the future ' w a v e s ' of the past planting decisions of the Sri Lankan rubber 
651 
industry- In particular we shall focus on the implications of alternative rates of replanting 
and derive their impact on future rubber production. 
In undertaking an exercise of this nature it must be understood that we are concen­
trating on the long run implications of past activities. Annual deviations from these 
trends are expected due to fluctuations in climatic conditions and in the response of pro­
ducers to current market prices. The crucial policy perspective is that the long run trends 
themselves cannot be avoided without radical changes in the technology of exploiting 
the rubber tree. 
After describing the sources of. data in the next section, the following sections first 
examine the rubber area and intensity of exploitation then suggest five alternative re­
planting strategies. Projections based on these are given in Section V and production 
scenarios are presented in Section VI. The final section compares the results with other 
studies and presents some overall conclusions. 
Data sources 
The^data used in this analysis were collected from two sources. The first was the 
Annual Administrative Report of the Rubber Controller from which data on area, pro­
duction and prices were obtained. Data on the immature and mature bearing areas 
were not presented in a manner that allows one to make projections of the future area. 
The second source of data was the Rubber Industry Master Plan (RIMP) Study 
(1979) of Sri Lanka. Unlike the Rubber Controller's data sources, this is a cross-sectional 
survey carried out in 1978, to investigate age -specific area, bark consumption, the re­
maining life of the trees and their general condition. With.the encouragement of the 
Rubber Research Institute, the Commonwealth Development Corporation, who conducted 
the survey, released to us 1305 questionnaries relating to over 13000 individual holdings 
selected on the basis of a stratified random sample using a set of aerial photographs as 
the sample frame. 
The major problem encountered in the RIMP Survey data was in using sample para­
meters for estimating population parameters. The lack of clarity in the text of the final 
Survey report and the fact that the ground survey and area estimates were not dii'ectly 
linked with each other created difficulties for further analysis. As a result in some areas 
sample strata had to be added using ' best guess ' weights, based on the significant rubber 
areas. The other main limitation the RIMP Survey was the lack of information on stock 
depletion which occurs as a consequence of extraneous weather conditions, disease, and 
deliberate decisions made by farmers to uproot or abandon the old rubber trees. 
In addition to these two main sources of data, data given by the Rubber Research 
Institute of Sri Lanka was also examined. Since these data are also based on the Rubber 
Controller's data, no significant difference could, be observed between them. 
Rubber area and exploitation 
At the end of the 1970s, the total rubber area in nine main rubber growing districts 
of Sri Lanka was estimated by the aerial photographic analysis of the Rubber Industry 
Master Plan Study (1979) to be 210,465 hectares + / - 4 per cent. This figure is below 
652 
the figure given by the Rubber Controller (226,400 hectares) but is close to the estimates 
provided by the Agricultural Census (1972/73) and Agricultural Productivity Committee 
Survey (1973), with figures of 205,800 ha and 209,800 ha respectively. The total rubber 
area, estimated from photographic analysis by the Rubber Industry Master Plan Study, 
is broken down by districts and given in Table 1. 
Table 1. Total rubber area 
Districts Estimated rubber Percentage 
area (ha) of Total 
Kalutara 42,400 20-3 
Kegalle 61,400 29-4 
Ratnapura 38,100 18-3 
Colombo 31,400 15 0 
Galle 15,700 7-5 
Matara 7,600 3 6 
Kurunegala 5,900 2-8 
Matale 2,800 1 3 
Kandy 3,400 1-6 
Total 208,700 100 0 
Source: Report on the Rubber Industry Master Plan Study — Volume V. Economic 
Studies — September 1979 — Page (ii) of the Summary. 
It has been estimated that large areas of marginal land were withdrawn from rubber 
production during the 1960s and 70s. Due to the high prices of rubber in the immediate 
post war period, some of the rubber plantations were exploited very intensively. On 
the other hand, rubber areas were abandoned in the following period of lower prices. 
The substantial reductions in 1962 and 1981 of the estimate of the total rubber area sug­
gests an underlying downward trend that has been obscured in the annual data (see 
Table 2). 
If senile rubber stands are not abandoned or withdrawn from rubber production, 
they must be replanted. Replanting of rubber is a major investment, especially for small­
holders. In addition to the finance required to clear and replant the land, there is also 
the loss of income during the immature period of the new cycle. Although intercropping 
can bring additional income during this period, it has not been practised on any great 
scale. This perhaps is largely due to lack of capital for initial investment in intercrops, 
coupled with factors such as uncertainty of.markets, labour constraints and the prevalence 
of theft. Furthermore, there has not been any active encouragement of intercropping 
and demonstrations have not been widespread. 
An accelerated replanting scheme was started in 1953. The success of the replanting 
subsidy program in exceeding the target of 75,000 acres (30,350ha) for the first five years 
is evident in Table 2. 93,978 acres (38,038 ha) were replanted or planted in this period. 
However from 1963 until the end of the 70s the rate of replanting fell to unsatisfactory 
levels, in the sense of being well below the target of 6,070 ha a year. 
653 
table 2. Area replanted 1945 — 1978 (ha.) i 
Total Replanted area 
Year area Total S. H. M.H. Est 
1945 267126 38 na na na­
1946 267025 666 na . na na 
1947 266794 857 na na na 
1948 266538 592 na na na 
1949 265226 890 na na na 
1950 265273 1565 na na na 
1951 265385 1387 na na na 
1952 265943 1700 na na na 
1953 266165 2348 188 400 1780 
1954 266886 7472 1588 2075 3809 
1955 267605 8464 2270 2484 3710 
1956 267500 9809 2757 3029 4024 
1958 269184 8355 2650 2357 3348 
1959 2705J7 7503 3150 1848 2504 
1960 270829 7244 2911 1837 2495 
1961 271591 7565 3106 1674 2807 
1962 229684 7269 2831 1413 3004 
1963 229952 6442 2489 1090 2863 
1964 230123 5487 1985 1559 1942 
1965 230384 5061 1956 1284 1821 
1966 230540 4689 1968 863 1866 
1967 230413 4083 1337 858 1888 
1968 230448 5155 1734 953 • 2470 
1969 230228 4892 1665 707 2520 
1970 230178 4145 1430 691 2023 
1971 229953 3431 1207 553 1672 
1972 229655 3539 951 448 2139 
1973 228673 2946 930 374 1641 
1974 228100 2865 986 338 1540 
1975 227730 3230 1128 520 1582 
1976 277074 2550 1132 317 1102 
1977 226660 2617 1196 313 1107 
1978 226400 3225 1415 320 1490 
1979 226300* 5381 1548 3833 
1980 227300* 5333 2240 3193 
1981 205600* 6442 3579 2863 
1982 205700* 6376 5224 1 152 
Source : Department of Rubber Control, Sri Lanka 
and 1982* 
New planted Replanted area 
area as a percentage 
of total area 
1104 0 02 
613 0 25 
183 0 33 
119 0 23 
232 0 34 
272 0 59 
570 0 53 
553 0 64 
348 0 89 
619 2 80 
502 3 17 
1130 3 67 
935 3 11 
813 2 78 
786 2 68 
572 2 79 
277 3 17 
268 2 81 
171 2 39 
260 2 20 
157 2 04 
55 1 78 
237 2 24 
126 2 13 
112 1 80 
238 1 50 
180 1 55 
186 1 29 
34 1 26 
142 1 42 
56 1 12 
45 1 15 
378 1 42 
582 2 37 
977 2 39 
1055 3 13 
1650 3 10 
Central Bank of Ceylon, 1978 
654 
The rubber replanting subsidy scheme is operated with the aim of achieving a re­
planting cycle of 33 years. This would require that about 3 percent of the total area 
under rubber be replanted annually. For fifteen years ('64 — 7 8 ) this target was not 
achieved, suggesting an ageing national rubber plantation. In fact the RIMP Study 
showed that 44 percent of trees had between 0 and 9 years' life remaining (RIMP Study, 
1979 : p. 7 ) . The registers of the Rubber Controller for the same year showed about 
89,070 ha of overage rubber, representing about 40 percent of the crop area. 
In order to remedy this situation and on the recommendation and with the funding 
of the FAO and World Bank (FAO — World Bank, 1 9 7 9 ) , the government of Sri Lanka 
in 1981 implemented a smallholder rubber replanting project. The resultant sharp in­
crease in replanting since then is shown in Table 2. The Table also indicates the' fluctua­
tions in the replanting rates. These fluctuations cause the ' waves' that are the main 
focus of this study. 
The rate of rubber replanting is to a large extent a reflection of the economic deci­
sions taken by the thousands of farm households. The long term investment decision 
of rubber replanting cannot be separated from a muliitude of other decisions, both short-
term and long-term, made by farmers. The decision to replant a particular rubber stand, 
plant up a new area or choose another crop takes place in the normal course of attempting 
to maximize some form of a multi-period utility function. While the variables in such 
a function include non-economic factors, economic considerations are likely to be crucial 
for rubber is, after all, grown as a cash crop. The theory of optimal replacement was 
set out by us at the last International Rubber Conference in this venue-(Eth.erington and 
Jayasuriya, 1976) and does not need repeating here. In practical terms the replacement 
date is decided upon by the past rate of bark consumption and the anticipated average 
annuity of returns from the new stand of trees. 
Using sample data collected by the RIMP Survey on the rate of bark consumption 
and the estimated remaining life of the trees we calculate a simple index of the intensity 
of exploitation. Although the remaining life of the trees is a highly subjective judgement, 
it does give an important assessment on the future life of the national rubber plantation. 
On the basis of the supposed standard replanting cycle of 33 years, the index of the inten­
sity of exploitation (INTEX) is simple : 
INTEX = 33/(A + R) X 100 
where A = Age of trees 
R = Remaining life of trees 
Table 3 gives the frequency distribution of this index for the 12 thousand hectares 
of rubber in the RIMP sample. The Table shows that the intensity of exploitation is 
equal to or below 100 percent only in 15 3 percent of the rubber area. This area is a 
reflection not of undeveloped trees, but of areas of proper bark consumption which have 
a reasonable age and remaining life. 
The estimate of the remaining life of the trees was calculated with reference to the 
rate of bark consumption. High consumption of bark while a tree is young reduces the 
.655 
remaining life. If this happens frequently, the ' ago ' plus the ' remaining l i fe' is below 
33. This is reflected in the Table 3 where intensity of exploitation is greater than 100 per­
cent. Thus 84 -7 percent of the total area falls into this category. However it is im­
portant to note that there are some areas of rubber which go out of production or pre­
maturely ' d ie ' due to pests and diseases and climatic factors. Such areas are not very 
significant, but it could be a minor reason for the intensity of exploitation being above 
100 percent. The age distribution of the national rubber estate is based on the relative 
frequency distribution presented in Table 3. In the RIMP Survey this frequency distri­
bution distinguishes between smallholders and estates. Unfortunately, however, in­
consistencies found in the survey data prevented the undertaking of regional analyses and 
it was only possible to build up a national^age distribution accounting for 171 thousand 
hectares (85%) out of the survey's estimated total of 208 thousand hectares. Again, 
while the impression given by Table 2 is one of basic stability in the size of the area under 
rubber, the need for major periodic downward adjustments suggests that the official 
records are not entirely satisfactory and belie an underlying downward trend of between 
0 5 and 1 percent per annum. In spite of these difficulties with the data, it seems to us 
valid to proceed with an analysis of the cyclical fluctuations in area and output as an 
illustration of this method of forecasting. 
Replanting strategies 
Given the, admittedly weak, estimates of the size and age distribution of the national 
rubber ' estate ' a set of five purely mechanistic ' demographic ' replacement models are 
applied to the data. The models are mechanistic in the sense that three of them do not 
allow divertions from deterministic cohort or vintage replacement cycles. This changes 
in the relative price of rubber are assumed to have no impact, yet we know that historically 
the Sri Lankan rubber industry has responded to high prices (Herath, 1975) by increasing 
the intensity of exploitation, raising output and, consequently, altering the length of the 
replacement cycle. 
Using the year of the RIMP Survey as the fulcrum, five replacement strategics are 
applied to the national estate : ' stepping back' through time as a means of validating 
the procedure and then 'stepping forward' in a predictive manner. The five strategies 
examined are : 
(a) a 26 years replanting cycle ; 
(b) a 30 years replanting cycle ; 
(c) a 33 years replanting cycle ; 
(d) a 3 percent of the area replanting per annum ; 
and (e) a replanting policy based on farmer's expressed intentions to replant. 
The first three strategies are based purely on the age of the plantation, that is every 
26, 30, or 33 years the rubber trees are all replanted. The first two cycle lengths were 
selected on the basis of the analysis done by Jayasuriya (1973) on the optimal replacement 
age under varying assumptions regarding tapping systems, discount rates, technological 
change, subsidy, and prices. These studies and the evidence of Table 3 show that optimal 
cycle lengths are typically less than 33 but more than 25 years. Nevertheless, the selection 
of the 33 years policy is particularly important since the current official rubber replanting 
scheme operates within an implied replacement cycle of this length. This has been the 
accepted ' rule of thumb ' of the rubber industry. 
6 5 6 
Table 3. Intensity of exploitation (INTEX) of rubber 
Length of Absolute Relative Cumulative 
replanting cycle INTEX frequency frequency frequency 
(Years) % (Sample ha.) % % 
78 46 24 0-2 0 2 
53 62 1 0 0 0 2 
51 65 15 0 1 0-3 
47 70 13 0 1 0 4 
46 72 15 0 1 0 6 
44 75 35 0-3 0 9 
43 77 25 * 0 2 1 1 
42 79 50 0 4 1-5 
39 85 100 0 8 2 3 
38 87 54 0 5 2 8 
37 89 221 1 8 4 6 
36 92 137 1-1 5 8 
35 94 432 3 6 9 4 
34 97 306 2 5 1 1 9 
33 100 406 3 4 15 3 
32 103 583 4 9 20 1 
31 106 2532 21 1 41 2 
30 110 1586 13 2 54 4 
29 114 2511 20 9 75-4 
28 118 576 4 8 80 2 
27 122 439 3 7 83 8 
26 127 317 2 6 86 5 
25 132 1310 10 9 97-4 
24 138 107 0 9 98 3 
23 143 135 1 1 99 4 
22 150 37 0 3 99 7 
21 157 9 0 1 99 8 
20 165 25 0 2 100 0 
19 174 1 0 0 100 0 
12 276 1 0 0 100 0 
Total 12003 100 0 
While a crop cycle of 33 years implies 3 percent annual replacement, a major dis­
tinction must be drawn between a cycle based on age and one based on a percentage of 
the area planted. In the former case the annual replanted area depends critically on the 
cohort structure of past planting. In the latter case, the annual replanting depends only 
on the total area under rubber and not on the age structure of the trees. The rubber 
industry planners in Sri Lanka have not faced up to this fundamental distinction since 
alarm is expressed whenever the replanting falls below 3 percent with no discussion as 
to whether this is simply because 33 years earlier fewer trees had been planted and/or 
replanted. An age-specific replanting policy will only match a corresponding' percentage-
of-area ' policy, if the latter policy is in fact implemented every year. 
657 
The final strategy under investigation is probably the most realistic since it is based 
on responses to a RIMP Survey question regarding the replanting intentions of the rubber 
farmers. Clearly the intentions of farmers are crucial in replanting policy decisions. 
The most significant aspect of this procedure is its capability to capture both the ' infant' 
and ' juvenile' mortality of the trees. In reality whatever the replanting strategy adopted, 
it is also important to consider the premature ' death' of trees due to wind, drought, 
excessive rains, disease or pests. 
Table 4, which is based on the survey results, bears out the contention that replace­
ment should be viewed as a probability distribution rather than a deterministic fact (Ethe-
rington 1977). The cumulative distribution shows that 14 1 percent of replacement is 
premature in the sense that removal has taken place at ages less than that normally con­
sidered to be economic (i.e. 25 years). A further 6 9 percent are allowed to continue 
into ' old age ' (over 37 years) and 3 9 percent into ' senility' (over 48 years). 
One simple method of validation is comparison of each model output with the available 
historical records. The logic behind the procedure is to go back to the past in the same 
way and under the same policy that future values are simulated. This has a dual purpose. 
First, it validates the model by comparing it with past records. Second, it allows the 
selection of the most accurate replanting strategy for future prediction purposes. It is 
presumed that the correct policy for future prediction is the one which gives past values 
closest to the actual records. 
In general the results of the validity tests of the models over the period 1950 — 1978 
were not very good. The best regression result was using Strategy 5 with a coefficient 
of determination (R**2) of 0-583. (The other strategies ranged from 0 247 for strategy 
2 to 0 315 for strategy 1). Rather better results were obtained comparing actual and 
predicted past output of strategy 5 with a 3 year moving average to smooth out annual 
fluctuations due to weather and other random influences. In order to obtain the pre­
dicted output figures, predicted areas were multiplied by the yield coefficients to be presented 
later. The two distributions are presented in Table 5. The coefficient of determination 
between these time series is 0 • 949 with a slope coefficient of 1 097 and an intercept of 
- 12505-7, indicating almost a 1 :1 correspondence. 
Model experimentation and projections 
In a real sense it is not particularly important whether any of the five strategies pre­
sented in the previous section accurately ' predict' the past if they are in fact applied in 
the future. Whatever strategy or group of strategies applied in the past, if from the pivot 
date of 1978 everybody actually adopted a 33 year cycle, then future replantings are com­
pletely predictable. Thus, given an initial (1978) age distribution of the total area, any 
strategy can be mechanistically applied to generate successive cohorts or vintages while 
retaining a constant total area. 
The computer simulations of the age distribution of the total rubber area under each 
of the five strategies are presented in Appendix Tables A. 1 to A. 5. With the first 3 stra­
tegies it is assumed that the area older than the strategy cycle length is ' senile' and is 
replanted in the first year. In the case of strategy 4 the issue does not arise while for 
658 
Table 4. Frequencies of intended removal age 
Sample Relative Cumulative 
ge frequency frequency frequency 
(ha.) 
0 81 0 586 0 586 
1 0 0 000 0 586 
2 1 0 007 0-593 
3 27 0 195 0-688 
4 I 0 007 0-695 
5 0 0 000 0 695 
6 61 0 441 1 1 3 6 
7 1 0 007 1-143 
8 1 0 007 1-150 
9 12 0 087 1-237 
13 3 0 021 1 258 
15 19 0 137 1-395 
16 1 0 007 1 -402 
17 0 0 000 1-402 
18 5 0 036 1-438 
19 1 0 007 1 -445 
20 40 0 289 1-734 
21 11 0 079 1 8 1 3 
22 52 0 376 2-189 
23 135 0 976 3-165 
24 129 0 932 4-097 
25 1385 10 014 14-111 
26 320 2 314 16-425 
27 488 3 528 19-953 
28 587 4 244 24 197 
29 2517 18 200 42-397 
30 1657 11 981 54-378 
31 3118 22 545 76-923 
32 664 4 801 81-724 
33 446 3 225 84 949 
34 306 2 213 87-162 
35 469 3 391 90-553 
36 137 0 991 91-544 
37 221 1 598 93 142 
38 215 1 555 94 697 
39 120 0 867 95-564 
40 68 0 492 96-056 
41 & above 533 3 944 
TOTAL 13830 100 000 100-00 
Source : RIMP Survey Data. 
659 
Table 5. Comparison of three year moving averages of 
model output and actual production 
Year Actual Model 
(metric tonnes) 
1953 107000 104795 -
1954 102200 109340 
1955 98000 103679 
1956 96000 105970 
1957 98000 107145 
1958 100000 110403 
1959 99000 102310 
I960 98000 100887 
1961 97000 102953 
1962 101000 104663 
1963 103000 98387 
1964 107000 106896 
1965 121000 108206 
1966 131000 127186 
1967 141000 144094 
1968 148000 142524 
1969 151000 105370 
1970 151000 154961 
1971 147000 138247 
1972 145000 140075 
1973 143000 142779 
1974 146000 144193 
1975. 145000 145270 
1976 149000 146928 
1977 151400 148388 
Source . RIMP Study Data and Model output. 
660 
Table 6. Annual replanting area (ha.) 
Year Strategy 1 Strategy 2 Strategy 3 Strategy 4 Strategy 5 
1978 39528 22333 16800 5141 7911 
1979 10726 1499 18 4141 4695 
1980 2037 5994 2374 5141 3142 
1981 4544 4174 3142 5141 6116 
1982 10660 5528 1499 5141 4126 
1983 10743 10726 5994 5141 6596 
1984 4704 2037 4174 5141 5263 
1985 11806 4544 5528 5141 4866 
1986 16695 10660 10726 5141 7164 
1987 9142 10743 2037 5141 6279 
1988 7845 4704 4544 5141 5325 
1989 2718 11806 10660 5141 6777 
1990 5078 16695 10743 5141 7209 
1991 2518 9142 4704 5141 6168 
1992 2206 7845 11806 5141 7892 
1993 1008 2718 16695 5141 5372 
1994 4623 5078 9142 5141 5596 
1995 10265 2519 7845 5141 71717 
1996 1289 2206 2718 5141 322 
1997 4059 1008 5078 5141. 3862 
1998 1350 4623 2519 5141 6121 
1999 1993 10265 2206 5141 6790 
2000 1381 1289 1008 5141 4103 
2001 249 4059 4623 5141 6895 
2002 2804 1350 10265 5141 7761 
2003 1400 1993 1289 5141 5002 
2004 39528 1381 4059 5141 4744 
2005 10726 249 1350 5141 2625 
2006 2037 2804 1993 5141 3527 
2007 4544 1400 1381 5141 2850 
2008 10660 22333 249 5141 3544 
2009 10743 1499 2803 5141 3212 
2010 4704 5994 1400 5141 4810 
2011 11806 4174 16800 5141 6816 
2012 16695 5528 18 5141 3505 
2013 9142 10726 2374 5141 5007 
2014 7845 2073 3142 5141 3835 
2015 2718 4544 1499 5141 4355 
2016 5078 10660 5994 5141 4216 
2017 2518 10743 4174 5141 3826 
2018 2206 4704 5528 5141 4952 
2019 1008 11806 10726 5141 4410 
2020 4623 16695 2037 5141 6765 
Source : RIMP Study Data and Model output. 
661 
strategy 5 all trees 41 years of age and older are pooled and the relative frequency of in­
tended removals from Table 4 (3.944 %) is applied to this area. The results in the Appendix 
are given for 1978, 1980 and then in 10 year intervals up until the year 2020. 
More interesting from a policy point of view are the annual projections of the re­
planting area presented in Table 6. The data given in Table 6 for the strategy of a 26 
years cycle (Strategy 1) seems to be impractical for immediate implementation, since the 
replanting of 39,068 ha in the first year is impossible. Additionally this strategy implies 
very low rubber production levels during the immature period of this area. Furthermore, 
this strategy would necessitate heavy capital investment in one large burst in order to 
replant such a large area. 
The second strategy (30 years cycle) presents a similar problem of impracticability. 
The initial block of replanting would be over 22,000 ha and this also could only be imple­
mented with a massive replanting programme. However, the one thing in favour of 
these two strategies is the rapid elimination of the senile area. 
The third strategy investigated the implication of the ' official' 33 years replanting 
cycle. Thus this projection should give a better approximation to reality. The initial 
replanting of over 16,000 ha is massive but maybe not as impossible as the two previous 
policies. 
The fourth strategy is, according to the rubber policy formulators, meant to be similar 
to the third. However, it can be seen from Table 6 and the Appendix that it has quite 
different implications for the rate of replanting. This strategy selects the oldest 3 per 
cent of the area and replants this amount every year. This replanting target has been 
set in order to secure the 33 years cycle. Thus, on an assumed total rubber area of 200,000 
hectares, policy makers have attempted to relate this policy to, the replanting of 6,000 ha. 
per annum (3 percent of 200,000). However, in reality, the area to be replanted annually 
is not uniformly equal to 6,000 ha. If the implementation of this policy commences in 
1978 and only takes the oldest 3 percent of the area each year it would only match the 
intended 33 years cycle by 2010 when the total senile area will have been eliminated. In 
the projections we make, the constant annual replanting of 3 percent is 5,141 hectares 
per annum. 
. . Table 6 makes it clear that a 3 percent per annum replanting policy will only equate 
with a 33 year cycle provided the 3 percent policy is adhered to every year. In general 
"a"33" years 'exploitation cycle (or any age specific cycle) will always reflect the cohort/ 
vintage history of previous plantings and replantings. 
The area replanted under Strategy five is also presented in Table 6. Whatever govern­
ment policy is, the real situation will deviate from it according to the farmer's behaviour. 
The advantage of this strategy is its likely closeness to reality, which was demonstrated 
by the validation procedure. 
Total production under different strategies 
The rubber production potential is to a large extent predetermined by the yield profile 
of the trees that are already being exploited. Standard yield curves have been constructed 
by the Rubber Research Institute of Sri Lanka (RRISL) and the Association of National 
662 
Rubber Producing Countries (ANRPC) for the area above and below 40-5 hectares. 
The yield curve of the ANRPC is largely based on the yield pattern of PB 86 taking the 
average of various clones into consideration. • The yield curve of RRISL is also based 
on the yield of PB 86 over 16 years of tapping in large scale experiments. 
Both curves are based on a 33 year replanting cycle. In this study these yield curves 
have to be modified to fit the length of each replanting cycle. The yield of a particular 
age in a shorter cycle is of course higher than in an longer cycle. Further, yields are 
assumed to differ between estates and smallholdings with the former having higher yields. 
In addition, for the future, it is also assumed that the new plantings will be high yielding 
varieties and together with the application of fertilizer and stimulants will give better 
yields. The yield curves related to the five strategies and the one based on a hypothetical 
yield assuming future improvements in technology are presented in Appendix Tables 
Bl and B2. 
Rubber production is determined by the mature area and the yield curve. The area 
prediction made previously and the yield curves mentioned in this section are used together 
to work out the production capacity. The following equation gives formal expression 
to the approach adopted : 
Q(t) = E a (i). X(i, t) + E b (i) -2 (i, t) 
where Q(t) is the production capacity in year ' t ' 
a(i) and b(i) are the yield levels of one hectare of rubber in the ' i ' the year of. 
tapping for smallholders and estates respectively. 
X(i,t) and Z(i, t) are the areas, in hectares, in their ' i ' the year of tapping in 
year ' t ' . 
The results are summarised in Fig. 1. The overall pattern clearly demonstrates 
the long run cyclical fluctuations caused by the cohort structure and the age/yield rela­
tionship. 
Rubber trees planted during the first half of the 1978's will come into bearing towards 
the end of the same decade and will reach maximum yield at the end of the next decade. 
Therefore the action that will be taken with respect to replanting and new planting today 
will have an impact, in terms of incremental production, in the'mid-1990's. This' pattern 
can be recognized from the projected production given in Fig. l a n d Appendix Table CI. 
Under Strategy one, the replanted area will only come into bearing the latter part of the 
1980's and full impact will not be felt until the mid 1990's. Gradually production begins 
to decrease again at the beginning of the 2000's. Generally with a shorter replanting 
cycle, high yields are obtained in the early years hut the trees also go out of production 
early. Since the cycle is a little longer in strategy two, the impact of the newly planted 
area will come later. The full impact on increased production of the initiation of this 
policy will be felt in the latter part of the 1990's. However, in strategy 3 the full impact 
of the strategy only starts from about the year 2000. Strategy four results in a constant 
amount to be replanted each year since it is based on 3 percent of a total area which is 
663 

itself fixed by assumption. Although the area replanted each year is constant the age 
structure of the total area at the beginning of the strategy is variable so that recovery to 
constant production levels of about 190,000 tons is not achieved until the next century. 
The production ' strategy' five is expected to be closest to reality. 
The projections of production presented here are simply aimed at quantifying the 
output norms or the productive capacity up to the year 2020. A comparison of Table 5 
and the projection of productive capacity in Fig. 1 (& Appendix CI) show a major 
discontinuity. This implies that all the projections grossly overestimate current and 
possibly future yields. However, even if the levels of output capacity are optimistic, the 
cyclical pattern will remain. To provide direct comparability to actual production levels, 
we present an alternative scale on the left hand side of Fig. 1. This scale represents 
a shift downwards of 30 percent and is an informed guesstimate on our part. Irrespec­
tive of the model chosen a major decline in Sri Lankan rubber production is predicted to 
occur between the starting date of 1978 and 1990. Using Model 5 as the most likely 
scenario, the prediction is for increasing output from about 1995 for about 15 years 
(to 2010) when the long run cycle will start to repeat itself. 
Comparisons and conclusions 
The comparison of our projections of output capacity with other projections of out­
put is informative and is given in Table 7. It can be seen that the projections give quite 
different estimates. A major difference between our projection and the others is the decline 
in production up to about 1995. 
Table 7. Comparison of results of different sources 
Year Model output RRlSL(l) ANRPC (2) World Bank (3) Smit (4) 
(thousand metric tonnes) 
1980 210 173 180 180 165 
1985 168 169 193 185 165 
1990 156 172 174 195 180 
1995 164 201 
— 195 
2000 191 245 
— — 210 
Source : 1. Rubber Research Institute of Sri Lanka, 1979. 
2. The Association of Natural Rubber Producing Countries, 1976. 
3. World Bank/FAO, 1978. 
4. Smit, 1981. 
Our predicted values from Model 5 are based on the expressed intentions and expecta­
tions of the rubber farmers/planters in 1978, prior to any recommendations or proposals 
under the Rubber Industry Master Plan. It is therefore interesting to compare declared 
intentions with actual replanting in the period 1978 to 1982. Actual replantings have 
been 26,857 hectares as against a declared intention of replanting 25,990 hectares. This 
suggests that the scaled output projections should be taken seiiously. In particular, 
the relatively low levels of output during the next ten years should be noted. A recent 
study of the National Planning Division of the Ministry of Finance and Planning (Public 
Investment Report 1983 — 87, May 1983) comes to a similar conclusion regarding these 
trends when they say that ' if the replanting programme continues . . . production will 
665 
decline to a low of 100 (thousand metric tons) by 1986 and then commence increasing 
rapidly as the replanted acreages mature, regaining the 1978 peak of 158 (thousand metric 
tons) by 1995, (para. 4.48). Our estimate is for a slightly longer trough with a new peak 
occurring well after the turn of the century. Actual output figures since 1978 and our 
scaled projections read as follows : 
Year Actual output Scaled projection 
(thousands of metric tonnes) 
1978 155 157 
1979 152 151 
1980 133 147 
1981 124 140 
1982 123 134 
1983 126* 129 
1984 
— 
123 
1985 — i 118 
Source: Central Bank of Ceylon (1983) *Estimate 
Two conclusions stem from this analysis : first, many of the projections of Sri Lanka's 
rubber output do not take adequate account of the long run cyclical fluctuations caused 
by past bursts of planting and replanting activity ; secondly, there is a very considerable 
gap between output projections based on idealised yield curves and the yield levels actually 
achieved. Our 30 percent adjustment is admittedly somewhat arbitrary but it highlights 
the major discrepancy between actual rubber yields and those suggested as likely by the 
RRISL. The RRISL yield estimates are modest while the hypothetical yields we give 
are not unreasonable. The International Rice Research Institute (IRRI) has made de­
tailed investigations into the 'constraints' which prevent farmers achieving either ex­
perimental yields or those of better farmers. Clearly a ' yield g a p ' of 30 percent is 
well worth exploring and it suggests that there may be very substantial returns to Sri 
Lanka from improvements in the effectiveness of the extension service — especially to 
the smallholders who, because of their current low yield levels provide the greatest po­
tential source of increasing output above historic levels. 
This paper has used data on the age distribution of the Sri Lanka rubber plantings 
collected by the RIMP Survey (1978) and has used this to generate forecasts of rubber 
production based on alternative replanting strategies. The mechanistic application of 
these replanting strategies result in the long-term cyclical recurrence of planting (replanting) 
booms. The results suggest that rubber output levels will decrease steadily over the next 
ten years before starting to recover. Such projections do not take account for short 
term fluctuations caused by weather or responses to changes in producer prices. This is 
the main drawback of this simulation procedure. It simply assumes that the natural 
rubber prices will remain high enough to continue rubber production and consistent re­
planting. Unless the future is very different from the past, prices will deviate around the 
trend and cause variations in the intensity of tapping and in the use of the other inputs 
and, therefore, actual output levels are likely to deviate from these projections. 
Note: References were not included in the original manuscript. 
666 
Table A l . 
1978 1980 
39528 809 2037 469 
1399 600 10726 425 
2803-714 39528-809 
248 729 1399 600 
1381-548 2803 714 
1993 340 248 729 
1350 456 1381 548 
4059 393 1993 340 
1289 277 1350 456 
10265 073 4059 393 
4623 044 1289 277 
1007 953 10265 073 
2206 464 4623 044 
2518 880 1007 953 
5077 877 2206 464 
2717 963 2518880 
7845 485 5077 877 
9141-782 2717 963 
16695 412 7845 485 
11806 088 9141 782 
4703 781 16695 412 
10743 476 11806 088 
10660 231 4703 781 
4543 812 10743-476 
2037-469 10660 231 
10726 425 4543 812 
RIMP Survey data. 
APPENDIX 
Total rubber area under Strategy 1 (ha.) 
YEAR 
1990 2000 2010 2020 
5077-877 1381-548 4703-781 4623 044 
2717-963 1993 340 10743 476 1007-953 
7845 485 1350 456 10660 231 2206 464 
9141-782 4059 393 4543 812 2518 880 
16695 412 1289 277 2037 469 5077-877 
11806 088 10265 073 10726 425 2717 963 
4703 781 4623 044 39528 809 7845-485 
10743 476 1007 953 1399 600 9141-782 
10660 231 2206-464 2803 714 16695 412 
4543 812 2518 880 248-729 11806 088 
2037 469 5077-877 1381 548 4703-781 
10726 425 2717 963 1993 340 10743 476 
39528 809 7845 485 1350 456 10660 231 
1399 600 9141 782 4059-393 4543-812 
2803 714 16695 412 1289 277 2037-469 
248 729 11806 088 10265 073 10726 425 
1381 548 4703 781 4623 044 39528 809 
1993 340 10743-476 1007 953 1399 600 
1350 456 10660 231 2206-464 2803 714 
4059 393 4543-812 2518 880 248-729 
1289-277 2037 469 5077-877 1381-548 
10265 073 10726 425 2717 963 1993 340 
4623 044 39528 809 7845 485 1350 456 
1007 953 1399 600 9141-782 4059 393 
2206 464 2803 714 16695 412 1289 277 
2518-880 248-729 11806 088 10265 073 
667 
Table A 2. Total rubber area under Strategy 2 (ha.) 
YEAR 
A G E 1978 1980 1990 2000 2010 2020 
0 22333-430 5994 061 16695 412 1289-277 5994 061 16695 412 
1 1399 600 1499-393 11806 088 10265 073 1499 393 11806 088 
2 2803 714 22333-430 4703 781 4623 044 22333 430 4703 781 
3 248 729 1399-600 10743 476 1007-953 1399 600 10743 476 
4 1381-548 2803-714 10660-231 2206 464 2803 714 10660 231 
5 1993 340 248 729 4543 812 2518 880 248 729 4543 812 
6 1350 456 1381548 2037 469 5077 877 1381-548 2037 469 
7 4059 393 1993 340 10726 425 2717 963 1993 340 10726 425 
8 1289 277 1350 456 5528 196 7845-485 1350 456 5528 196 
9 10265 073 4059 393 4173 728 9141-782 4059 393 4173 728 
10 4623 044 1289 277 5994 061 16695 412 1289 277 5994 061 
11 1007 953 10265 073 1499 393 11806 088 10265 073 1499 393 
12 2206 464 4623 044 22333 430 4703 781 4263 044 22333 430 
13 2518 880 1007 953 1399 600 10743-476 1007 953 1399 600 
14 5077 877 2206 464 2803 714 10660 231 2206 464 2803-714 
15 2717 963 2518 880 248 729 4543 812 2518 880 248-729 
16 7845 485 5077-877 1381-548 2037 469 5077 877 1381 548 
17 9141-782 2717 963 1993 340 10726-425 2717 963 1993 340 
18 16695 412 7845-485 1350 456 5528-196 7845 485 1350 456 
19 11806 088 9141-782 4059-393 4173-728 9141-782 4059 393 
20 4703 781 16695 412 1289 277 5994 061 16695 412 1289-277 
21 10743 476 11806 088 10265 073 1499 393 11806 088 10265 073 
22 10660 231 4703 781 4623 044 22333 430 4703-781 4623 044 
23 4543 812 10743 476 1007-953 1399 600 10743 476 1007 953 
24 2037 469 10660 231 2206 464 2803 714 10660 231 2206 464 
25 10726 425 4543 812 2518 880 248•729 4543 812 2518 880 
26 5528 196 2037 469 5077 877 1381-548 2037 469 5077 877 
27 4173-728 10726-425 2717 963 1993 340 10726 425 2717 963 
28 5994 061 5528 196 7845 485 1350-456 5528 196 7845 485 
29 1499 393 4173-728 9141 782 4059 393 4173 728 9141 782 
Source : RIMP Survey data. 
668 
Table A 3. Total rubber area under Strategy 3 (ha.) 
* Y E A R 
# 
AGE 1978 1980 1990 2000 2010 2020 
0 16800-220 2373-955 10743 476 1007-953 1399 600 2037-469 
1 1399 600 17 551 10660 231 2206 464 2803 714 10726 •425 
2 2803 714 16800 220 4543 812 2518 880 248•729 5528 196 
3 248 729 1399 600 2037 469 5077 877 1381 548 4173 728 
4 1381 548 2803 714 10726 425 2717 963 1993 340 5984 061 
5- 1993 340 248-729 5528 196 7845 485 1350 456 1499 393 
6 1350 456 1381-548 4173 728 9141 782 4059 393 3141 704 
7 4059 393 1993 340 5994 061 16695 412 1289 277 2373 955 
8 1289 277 1350 456 1499 393 11806 088 10265 073 17 551 
9 10265 073 4059 393 3141 704 4703 781 4623 044 16800 220 
10 4623 044 1289 277 2373 955 10743-476 1007 953 1399 600 
11 1007 953 10265 073 17 551 10660 231 2206 464 2803 714 
12 2206 464 4623 044 16800 220 4543 812 2518 880 248 729 
13 2518 880 1007 953 1399 600 2037 469 5077 877 1381 548 
14 5077 877 2206 464 2803 714 10726-425 2717 963 1993 340 
15 2717 963 2518 880 248 729 5528 196 7845-485 1350 456 
16 7845 485 5077 877 1381 548 4173-728 • 9141 782 4059 393 
17 9141 782 2717 963 1993 340 5994 061 16695-412 1289 277 
18 16695 412 7845 485 1350 456 1499 393 11806 088 10265 073 
19 11806 088 9141 782 4059 393 3141 704 4703-781 4623 044 
20 4703 781 16695 412 1289 277 2373 955 10743-476 1007 953 
21 10743 476 11806 088 10265 073 17 551 10660 231 2206 464 
22 10660 231 4703 781 4623 044 16800 220 4543 812 2518 880 
23 4543 812 10743 476 1007 953 1399 600 2037 469 5077 877 
24 2037 469 10660 231 2206 464 2803 714 10726 425 2717 963 
25 10726 425 4543-812 2518 880 248•729 • 5528 196 7845 485 
26 5528 196 2037 469 5077 877 1381 548 4173 728 9141 782 
27 4173-728 10726 425 2717 963 1993 340 5994 061 16695 412 
28 5994 061 5528 196 7845 485 1350 456 1499 393 11806 088 
29 1499 393 4173 728 9141 782 4059-393 3141 704 4703 781 
30 3141 704 5994 061 16695 412 1289 277 2373 955 10743 476 
31 2373 955 . 1499 393 11806 088 10263 073 17 551 10660 231 
32 17-551 3141-704 4703 781 4623 044 16800 220 4543 812 
Source ; R I M P Survey data 
669 
Tabic A 4 . Total rubber area under Strategy 4 (ha.) 
YEAR 
A G E 1 9 7 8 1 9 8 0 1 9 9 0 2 0 0 0 2 0 1 0 2 0 2 0 
0 5 1 4 1 2 8 2 5 1 4 1 2 8 2 5 1 4 1 2 8 2 5 1 4 1 2 8 2 5 1 4 1 2 8 2 5 1 4 1 2 8 2 
1 1 3 9 9 6 0 0 5 1 4 1 2 8 2 5 1 4 1 2 8 2 5 1 4 1 2 8 2 5 1 4 1 2 8 2 5 1 4 1 2 8 2 
2 2 8 0 3 7 1 4 5 1 4 1 2 8 2 5 1 4 1 2 8 2 5 1 4 1 2 8 2 5 1 4 1 2 8 2 5 1 4 1 2 8 2 
3 2 4 8 7 2 9 1 3 9 9 6 0 0 5 1 4 1 2 8 2 5 1 4 1 2 8 2 5 1 4 1 2 8 2 5 1 4 1 2 8 2 
4 1 3 8 1 5 4 8 2 8 0 3 7 1 4 5 1 4 1 2 8 2 5 1 4 1 2 8 2 5 1 4 1 2 8 2 5 1 4 1 2 8 2 
5 1 9 9 3 3 4 0 2 4 8 7 2 9 5 1 4 1 2 8 2 5 1 4 1 2 8 2 5 1 4 1 2 8 2 5 1 4 1 2 8 2 
6 1 3 5 0 4 5 6 1 3 8 1 5 4 8 5 1 4 1 2 8 2 5 1 4 1 2 8 2 5 1 4 1 2 8 2 5 1 4 1 2 8 2 
7 4 0 5 9 3 9 3 1 9 9 3 3 4 0 5 1 4 1 2 8 2 5 1 4 1 2 8 2 5 1 4 1 2 8 2 5 1 4 1 2 8 2 
8 1 2 8 9 2 7 7 1 3 5 0 4 5 6 5 1 4 1 2 8 2 5 1 4 1 2 8 2 5 1 4 1 2 8 2 5 1 4 1 2 8 2 
9 1 0 2 6 5 0 7 4 4 0 5 9 3 9 3 5 1 4 1 2 8 2 5 1 4 1 2 8 2 5 1 4 1 2 8 2 5 1 4 1 2 8 2 -
1 0 4 6 2 3 0 4 4 1 2 8 9 2 7 7 5 1 4 1 2 8 2 5 1 4 1 2 8 2 5 1 4 1 2 8 2 5 1 4 1 2 8 2 
11 1 0 0 7 9 5 3 1 0 2 6 5 0 7 4 5 1 4 1 2 8 2 5 1 4 1 2 8 2 5 1 4 1 2 8 2 5 1 4 1 2 8 2 
1 2 2 2 0 6 4 6 4 4 6 2 3 0 4 4 5 1 4 1 2 8 2 5 1 4 1 2 8 2 5 1 4 1 2 8 2 5 1 4 1 2 8 2 
1 3 2 5 1 8 8 8 0 1 0 0 7 9 5 3 1 3 9 9 6 0 0 5 1 4 1 2 8 2 5 1 4 1 2 8 2 5 1 4 1 2 8 2 
1 4 5 0 7 7 8 7 7 2 2 0 6 4 6 4 2 8 0 3 7 1 4 5 1 4 1 2 8 2 5 1 4 1 2 8 2 5 1 4 1 2 8 2 
1 5 2 7 1 7 9 6 3 2 5 1 8 8 8 0 2 4 8 7 2 9 5 1 4 1 2 8 2 5 1 4 1 2 8 2 5 1 4 1 2 8 2 
1 6 7 8 4 5 4 8 5 5 0 7 7 8 7 7 1381 5 4 8 5 1 4 1 2 8 2 5 1 4 1 2 8 2 5 1 4 1 2 8 2 
1 7 9 1 4 1 7 8 3 2 7 1 7 9 6 3 1 9 9 3 3 4 0 5 1 4 1 2 8 2 5 1 4 1 2 8 2 5 1 4 1 2 8 2 
1 8 1 6 6 9 5 4 1 2 7 8 4 5 4 8 5 1 3 5 0 4 5 6 5 1 4 1 2 8 2 5 1 4 1 2 8 2 5 1 4 1 2 8 2 
1 9 1 1 8 0 6 0 8 8 9 1 4 1 7 8 3 4 0 5 9 3 9 3 5 1 4 1 2 8 2 5 1 4 1 2 8 2 5 1 4 1 2 8 2 
2 0 4 7 0 3 7 8 1 1 6 6 9 5 4 1 2 1 2 8 9 2 7 7 5 1 4 1 2 8 2 5 1 4 1 2 8 2 5 1 4 1 2 8 2 
2 1 1 0 7 4 3 4 7 5 1 1 8 0 6 0 8 8 1 0 2 6 5 0 7 4 5 1 4 1 2 8 2 5 1 4 1 2 8 2 5 1 4 1 2 8 2 
2 2 1 0 6 6 0 2 3 1 4 7 0 3 7 8 1 4 6 2 3 0 4 4 5 1 4 1 2 8 2 5 1 4 1 2 8 2 5 1 4 1 2 8 2 
2 3 4 5 4 3 8 1 2 1 0 7 4 3 4 7 5 1 0 0 7 9 5 3 1 3 9 9 6 0 0 5 1 4 1 • 7 8 2 5 1 4 1 • 2 8 2 
2 4 2 0 3 7 4 6 9 1 0 6 6 0 2 3 1 2 2 0 6 4 6 4 2 8 0 3 7 1 4 5 1 4 1 2 8 2 5 1 4 1 2 8 2 
2 5 1 0 7 2 6 4 2 5 4 5 4 3 8 1 2 2 5 1 8 8 8 0 2 4 8 7 2 9 5 1 4 1 2 8 2 5 1 4 1 2 8 2 
2 6 5 5 7 8 • 1 9 6 2 0 3 7 4 6 9 5 0 7 7 8 7 7 1381 5 4 8 5 1 4 1 2 8 2 5 1 4 1 2 8 2 
2 7 4 1 7 3 7 2 8 1 0 7 2 6 4 2 5 2 7 1 7 9 6 3 1 9 9 3 3 4 0 5 1 4 1 2 8 2 5 1 4 1 2 8 2 
2 8 5 9 9 4 0 6 1 5 5 2 8 1 9 6 7 8 4 5 4 8 5 1 3 5 0 4 5 6 5 1 4 1 2 8 2 5 1 4 1 2 8 2 
2 9 1 4 9 9 3 9 3 4 1 7 3 7 2 8 9 1 4 1 7 8 3 4 0 5 9 3 9 3 5 1 4 1 2 8 2 5 1 4 1 • 2 8 2 
3 0 3 1 4 1 • 7 0 4 5 9 9 4 0 6 1 1 6 6 9 5 4 1 2 1 2 8 9 2 7 7 5 1 4 1 2 8 2 5 1 4 1 • 2 8 2 
3 1 2 3 7 3 9 5 5 1 4 9 9 3 9 3 1 1 8 0 6 0 8 8 1 0 2 6 5 0 7 4 5 1 4 1 • 2 8 2 5 1 4 1 2 8 2 
3 2 1 7 5 5 1 3 1 4 1 7 0 4 4 7 0 3 7 8 1 4 6 2 3 0 4 4 5 1 4 1 • 2 8 2 5 1 4 1 2 8 2 
3 3 1 2 0 3 5 2 6 2 7 7 3 9 5 5 1 0 7 4 3 4 7 5 1 0 0 7 9 5 3 1 3 9 9 6 0 0 1 7 1 3 7 6 1 
3 4 1 0 2 4 5 0 1 1 7 5 5 1 6 6 0 0 7 4 2 2 0 6 4 6 4 3 1 4 161 0 0 0 0 
3 5 0 0 0 0 1 2 0 3 5 2 6 0 0 0 0 2 5 1 8 8 8 0 0 0 0 0 0 0 0 0 
3 6 1 6 9 7 4 7 3 1 7 2 8 4 6 0 0 0 0 5 0 7 7 8 7 7 0 0 0 0 . 0 0 0 0 
3 7 2 0 0 8 3 8 4 0 0 0 0 0 0 0 0 2 7 1 7 9 6 3 0 0 0 0 0 0 0 0 
3 8 0 0 0 0 0 0 0 0 0 0 0 0 7 8 4 5 4 8 5 0 0 0 0 0 0 0 0 
3 9 3 1 9 4 3 5 9 0 0 0 0 0 0 0 0 . 2 3 3 7 • 7 8 9 0 0 0 0 0 0 0 0 
4 0 2 4 0 7 0 5 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 
4 1 1 7 6 2 1 6 3 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 
6 7 0 
tab le A 5. Total rubber area under Strategy 5 (ha.) 
Y E A R 
AGE 1978 1980 1990 2000 2010 2020 
0 7910 3142 7208 4102 4809 6765 
1 1391 4670 6728 6750 3193 4384 
2 2803 7864 5293 6085 . 3523 - 4923 
3 249 1391 6241 3839 2833 3803 
4 1379 2798 6125 3197 3499 4183 
5 1993 248 4827 7114 2604 4320 
6 1350 1378 5220 5552 4706 3840 
7 4041 1984 6515 5305 4940 4944 
8 1289 1344 4705 7793 7664 3461 
9 10264 4040 6040 6091 6809 6730 
10 4619 1287 3100 7112 4048 4745 
11 1008 10255 4635 6687 6669 3169 
12 2206 4619 7805 5253 6039 3497 
13 2519 1008 1318 6195 3810 2812 
14 5077 2203 2782 6090 3179 3479 
15 2717 2518 246 4800 7073 2859 
16 7834 5069 1369 5184 5513 4673 
17 9141 2714 1979 6497 5291 4927 
18 16695 7834 1340 4064 7773 7645 
19 11801 9138 4029 6022 6073 6789 
20 4703 16688 1285 3093 7097 4039 
21 10712 11767 10204 4611 6653 6666 
22 10651 4686 4592 7760 5223 6005 
23 4526 10663 998 1368 6136 3775 
24 2017 10508 2164 2729 5975 3118 
25
 r 
10626 4440 2447 240 4665 6875 
26 ' 4974 1798 4440 1199 4540 4825 
27 4077 9340 2322 1693 5559 4527 
28 5782 4688 6466 1106 5559 4527 
29 1435 3766 7224 3185 4761 4802 
30 2569 4529 10793 831 2000 4590 
31 2090 1034 6718 5826 2633 3799 
32 15 1752 2073 2032 3434 2311 
33 1145 1540 4509 422 578 2595 
34 991 13 4543 894 1128 2470 
35 — 1084 1811 998 98 1903 
36 1640 937 787 1945 525 1988 
37 1988 
— 
4147 1030 751 2467 
38 
— 
— 
2122 2928 501 1519 
39 3145 1598 1753 3363 1483 2216 
40 238 1927 2555 6089 469 1128 
41 & above 1753 3102 659 4285 3716 1679 
Source : R I M P Survey data. 
671 
APPENDIX B 
Table B 1. Yield curve tinder diffen 
Year of Strategy 1 Strategy 2 Strategy 
tapping 
0 420 560 560 
2 672 700 700 
3 840 805 805 
4 980 910 910 
5 1120 980 980 
6 1120 1050 1050 
7 1120 1120 1120 
8 1120 1170 1170 
9 1232 1190 1190 
10 1232 1210 1230 
11 1232 1210 1230 
12 1232 1190 1200 
13 1232 1170 1180 
14 1232 1140 1160 
15 1064 1110 1120 
16 1064 1010 1010 
17 1064 960 . 950 
18 980 860 840 
19 960 770 750 
20 960 630 650 
21 700 700 
22 900 900 
23 800 800 
24 700 700 
25 700 
26 650 
27 600 
28 
29 
30 
31 
32 
33 
34 
35 
•nt Strategies area below 40-5 ha. 
3 Strategy 4 Strategy 5 Hypothetical 
curve 
450 450 800 
630 650 1000 
900 750 1200 
1035 850 1350 
1053 870 1450 
1080 900 1550 
1134 980 1654 
1215 1020 1700 
1180 1040 1750 
1220 1150 1800 
1210 1150 1800 
1210 1140 1650 
1200 1060 1700 
1180 1040 1650 
1140 1020 1550 
1010 1010 1500 
940 950 1350 
840 820 1250 
750 730 1150 
730 630 900 
700 640 650 
850 720 650 
800 720 600 
700 650 600 
650 640 580 
650 640 580 
650 640 560 
640 630 560 
640 620 560 
600 620 560 
550 580 580 
540 540 • 540 
540 540 540 
530 530 
520 520 
510 510 
672 
36 j 
Table B 2. Yield curve under different Strategies area above 40 5 ha. 
Year of Strategy 1 Strategy 2 Strategy 3 Strategy 4 Strategy 5 Hypothetica 
tapping yield curve 
1 420 640 640 640 640 1000 
2 728 800 800 800 800 1200 
3 924 920 920 920 920 1450 
4 1092 1040 1040 1020 1020 1600 
5 1176 1120 1120 1120 1100 1700 
6 1232 1200 1200 1200 1180 1800 
7 1288 1280 1280 1280 1240 1900 
8 1344 1330 1300 1270 1280 1950 
9 1300 1360 1340 1320 1310 2000 
10 1400 1380 1360 1360 1340 2050 
11 1400 1380 1380 1380 1340 2050 
12 1400 1360 1340 1320 1320 2000 
13 1400 1340 1320 1320 1300 1950 
14 1400 1300 1300 1280 1260 1900 
. 15 1400 1250 1250 1250 1250 1800 
16 1400 1160 1160 1140 1120 1750 
17 1400 1100 1100 1100 1000 1650 
18 1300 1100 1100 1100 1000 1500 
19 1200 1000 1000 1000 980 1400 
20 1150 980 960 800 820 1200 
21 40 820 800 780 1000 
22 1000 980 980 840 1000 
23 960 940 900 860 1020 
24 940 920 800 720 950 
25 1000 1000 860 940 
26 1000 ' 900 880 920 
27 980 960 860 900 
28 910 800 900 
29 900 860 880 
30 840 860 820 
31 820 860 820 
32 820 840 820 
33 820 810 780 
34 780 760 
35 760 730 
36 720 720 
673 
APPENDIX C 
Total production of rubber under 5 Strategies — Sri Lanka 
Year Strategy 1 Strategy 2 Strategy 3 Strategy 4 Strategy 5 
1978 199874 490 213895 800 218076 780 225541 460 223698 730 
1979 187171 770 208607 950 213836 750 218267 680 215911 590 
1980 182234 480 200477 180 208450 650 211147 700 209739 890 
1981 172511 600 190913 910 199676 050 200966 910 198957 350 
1982 157740 340 180247 010 191954 390 190586 670 189156 990 
1983 143475 160 166392 740 181737 790 180992 600 178502 390 
1984 168928 310 178352 700 187001 690 175057 140 174968 640 
1985 168636 800 172002 100 177393 600 168011 070 168079 360 
1986 160044 570 166321 880 167099 240 162685 940 160622 000 
1987 159957 970 160912 200 165353 530 160431 530 158171 460 
1988 165846 700 162058 160 161014 300 159559 080 155995 470 
1989 178117 540 163526 360 156869 640 159615 610 156179 030 
1990 185272 510 155599 710 152416 910 160665 390 155919 460 
1991 198905 080 154659 270 153349 540 161861 990 156172 520 
1992 218282 180 160669 260 155082 450 163926 140 157342 810 
1993 234143 020 171556 400 146800 160 166048 260 160609 760 
1994 243618 640 174761 880 144309 430 166879 680 161777 150 
1995 241609 810 185981 490 148278 890 168438 870 163991 020 
1996 248766 260 203637 220 158766 380 171649 440 170766 790 
1997 251056 980 217591 440 163673 630 174528 970 176395 560 
1998 252563 070 227376 260 176142 460 177494 060 182237 380 
1999 251990 320 228117 810 194488 360 180329 290 185628 580 
2000 251730 190 234502 700 208126 •570 183155 860 190534 830 
2001 255980 •980 235996 790 218355 320 185401 310 194289 540 
2002 250892 970 237897 570 218832 220 187270 330 194068 130 
2003 243631 160 233621 230 224088 410 188459 020 194501 660 
2004 199874 490 231968 110 224928 250 188888 330 195821 030 
2005 187171 770 239085 650 229130 580 189430 990 199896 770 
2006 182234 480 235811 890 228721 770 189650 500 200867 900 
2007 172511 600 234841 860 230857 440 189987 040 204017 210 
2008 157740 340 213895 800 237747 010 190239 210 207257 800 
2009 143475 160 208607 950 234656 930 190500 660 208363 040 
2010 168928 310 200477 180 234049 800 190585 910 208516 650 
2011 168636 800 190913 910 218076 780 190589 050 204896 070 
2012 160044 570 180247 010 213836 750 190589 050 202980 400 
2013 159957 970 166392 740 208450 650 190589 050 199536 300 
2014 165846 700 178352 700 199676 050 190589 050 197084 200 
2015 178117 540 172002 100 191954 390 190589 050 193202 660 
2016 185272 510 166321 880 181737 790 190589 050 191034 060 
2017 198905 080 160912 200 187001 690 190589 050 190985 800 
2018 218282 180 162058 160 177393 600 190589 050 187829 640 
2019 234143 020 163526 360 167099 240 190589 050 186837 850 
2020 243618 640 155599 710 165353 530 190589 050 183138 470 
Source : RIMP Survey data. 
674 
THE IMPACT OF GOVERNMENT POLICIES ON INVESTMENT 
IN THE RUBBER INDUSTRY OF SRI LANKA 
By 
S. K . JAYASURIYA AND C. BARLOW 
(Australian National University) 
Throughout the 1970's the Sri Lankan rubber industry stagnated in terms of output, 
cultivated area and share of gross domestic product (GDP) sec Table 1. Such a decline, 
in itself, h no cause for concern ; indeed, it is to be expected that as economic develop­
ment proceeds, the role and significance of labour intensive primary product industries 
would diminish after a certain stage. However, it is difficult to attribute the decline of 
the Sri Lankan rubber industry to the process of economic growth. A review of (he 
past period suggests that the content and implementation of a series of government policies 
contributed far more than general economic development to the decline of the industry. 
And this occurred during a period in which the industry could potentially have made an 
enhanced contribution to the national economy, facilitating a higher rate of growth. 
Tabic 1. Some basic statistics of the Sri Lankan rubber industry : 1970 —1983 
Year 
Area 
(hectares) 
Production 
(millions of 
kilogrammes) 
Share of 
gross domestic 
product (%) 
Share of 
exports 
(%) 
Average yield 
(kilogrammes 
per hectare) 
1970 230,324 159 1 7 21 6 792 
1971 229,957 142 1 3 15 0 736 
1972 229,579 140 1-2 13 3 737 
1973 228,744 155 1 7 22-6 778 
1974 228,100 132 1 3 21 1 710 
1975 227,730 149 1 5 16 5 774 
1976 227,074 152 1 5 18 4 790 
1977 226,660 146 1-3 14 0 773 
1978 226,420 156 1-3 15 3 845 
1979 226,419 153 1-2 16 3 820 
1980 229,335 ' 133 1 0 14-7 718 
1981 205,605 124 0 9 13 7 705 
1982 205,690 125 0-8 10 8 729 
1983 205,649 135 0 9 1 1 0 823 
Source : Annual Review of the Economy, Central Bank of Ceylon, Colombo. 
675 
After the 1950 -—51 Korean War related price boom, rubber prices followed a secular 
decline for some two decades ; a clear falling trend was most apparent during 1960 — 1972. 
Other things remaining constant, this would have automatically led to lower profitability 
and, where attractive alternatives were present, to lower levels of investment. However, 
other things were not, or at least were not necessarily, constant. The national rubber 
industry had developed the potential for major technological changes which could drasti­
cally reduce the cost of production. The most important of available new technologies 
were embodied in new cultivars, which had a yield potential many times that of the trees 
planted a few decades ago. Replacement of older trees with new high yielding cultivars, 
together with associated new technologies, could have raised per hectare yields substan­
tially and reduoed the cost of production which could have offset the downward pressure 
on profitability due to falling prices. This was the path broadly traversed by the Malay­
sian rubber industry (Barlow, 1978); as a result, when rubber prices were more favourable 
in the 1970's, it was in a position to register high levels of output and profits. Despite 
substantial replantings the Sri Lankan industry proved unable to take advantage of the 
favourable period in the 1970's. and yields and exports stagnated. 
Sri Lankan government policies, or rather policy statements, have always stressed 
the importance of the plantation crop sector and the need to enhance its output and pro­
ductivity. Indeed, a number of direct policy measures were aimed at achieving these 
objectives. These included replanting grants, provision of research and extension ser­
vices and subsidised fertiliser inputs. There is no doubt that these had beneficial effects. 
Little new planting was undertaken after the early 1950's, partly because most suitable 
lands were already under rubber cultivation ; new investment refers most importantly to 
replanting. The replanting grants, availability of high yielding cultivars and an effective 
extension service contributed to the substantial replanting with new cultivars which 
occurred during the 1950's (and to a lesser extent in the 1960's). These replantings were 
the major factor behind rising yield levels from the mid 1950's, a trend which continued 
till the end of the 1960's. By 1970, average yields had nearly doubled over those in 1953, 
an impressive achievement in a long lived crop industry. These yield gains, however, 
masked parallel processes which were to result in the stagnation of yields and the decline 
of the industry during the 1970's. 
In Tables 2 and 3 we present some figures on rubber replantings, most of which 
involved high yielding cultivars (due to limited access to Sri Lankan data, we are unable 
to present replanting performances by sub-groups after 1973). The rapid decline of the 
large estate replantings after the early period of the replanting subsidy scheme preceded 
the very low prices of the 1960's^ However, till the late 1960's, the other sub-groups 
showed a respectable replanting performance. A detailed discussion of the replanting 
performance up to 1970, in the context of policies directly affecting the rubber sector, has 
been presented by Wijepala and Senanayaka (1970). An econometric analysis of replant­
ing activities (Herath, 1975) found that profitability considerations were dominant in 
explaining the behaviour of the three producer groups. Further econometric analysis 
which puts particular emphasis on the major role of replantings in supply response of 
Sri Lankan rubber is currently under way (Nerlove, M : personal communication), and 
preliminary results indicate that the real level of the replanting grant is of major impor­
tance. 
676 
Table 2. Area replanted by size group : 1953 — 1973 
Year .' , . Replanted area (acres) 
Estates over Estates Smallholdings 
100 acres 10 -100 acres under 10 acres 
1953 4,346 989 465 
1954 9,408 5,126 3,923 
1955 9,164 6,135 .; 5,608 
1956 10,803 7,012 6,770 
1957 9,939 7,481 6,809 
1958 8,270 5,821 . 6,547 
1959 6,185 4,566 ., 7,781 
1960 - 6,164; 4,538 < 7,191 
1961 7,617 . 4,135 6,933 
1962 7,741 • 3,490 6,993 
1963 '7,071 2,694 ; . 6,147 
1964 4,797 3,851 •'. 4,904 
1965 4,498 3,171 4,832 
1966 4,608 2,131 4,842 
1967 4,664 2,120 , 3,302 
1968 6,100 : 2,352 4,282 
1969 6,225 .' 1,747 4,112 
1970 4,998 1,708 3,533 
1971 . 4,130 1,365 • 2,981 
1972 • 5,283. 1,109 2,350 
1973 4,053 . 925 2,298 
Cumulative Total 135,794 72,468 • 102,604 
Source : Administration Report of the Rubber Controller for 1973; 
Tabic 3 . Total annual replanted areas after 1973 
. Year Area (acres) 
1974 7,076 
1975 7,980 
1976 ' 6,299 
1977 6,464 
1978 7,968 
1979 10,294 
1980 13,422 
1981 15,912 
1982 14,489 
1983 12,643 
Source : Report of the Central Bank of Ceylon, 1983. 
• > 
677 
The replanting performance declined during the 1960's and was quite low until the 
late 1980's, when substantially higher replanting grants and favourable price expectation 
lifted the rate up ; this was also due to the implementations of the rubber rehabilitation 
programme under World Bank assistance. However, due to the below target performance 
of replanting in the past (and as areas replanted in the early 1950's are again coming due 
for replanting) higher levels need to be achieved if productivity levels are to be improved. 
Most discussions of the policies of the rubber industry in Sri Lanka have paid atten­
tion to the direct effects of government policies which affected profitability. These in­
clude the political environment created by the nationalisation, land reform issues and the 
export tax. However, to appreciate the full extent of the impact of government policies 
these should be assessed in the context of overall government policy affecting all sectors 
of the economy. Within the substantial data and time constraints, in this paper we 
hope to make an initial contribution towards such a study. 
The nationalisation issue 
The coming into power of a government pledged to nationalise foreign owned plan­
tations drastically changed the investment climate for foreigners from 1956 ; they were 
not to know that two decades would pass before the government would move in to acquire 
foreign plantations with appropriate compensation. The policy of proclaiming the in­
tention to nationalise but postponing its actual implementation ensured, as Kaldor (1958) 
pointed out, that the foreign estates industry had the worst of both worlds. Investment 
in technical change, particularly in the crucial replantings, would yield benefits only in 
the long run. Given the uncertainty regarding the date of nationalisation and the distinct 
possibility of lower than market value compensation levels, privately optimal policy for 
foreign plantations was clearly to reduce long term investments and to increase the rate 
if exploitation of current fixed capital stock. The " running down " of trees and equip­
ment was inevitable. Some decline in replantings in the large estate sector towards the 
end of the 1950's was to be expected anyway, since a backlog of replantings were under­
taken in the early years of the replanting subsidy scheme initiated in 1953 ; however, 
there is no doubt that the continuing low levels of replanting reflected the desire to reduce 
investment in the Sri Lankan rubber industry by foreign interests. Where possible foreign 
companies reduced their ownership of rubber plantations ; by 1974 they owned less than 
one third of the estates (though admittedly these included some of the most productive 
rubber area). 
The trade policies 
The nationalisation threat directly affected only the foreign owned sub-sector. The 
changes in the trade policies, particularly from the late 1950's, affected the entire industry. 
However, the main consequence of these changes was to progressively reduce, both ab­
solutely and relatively, the incentives for investment in rubber. The trade regime embodied 
an anti-agricultural and 1 anti-export bias which particularly affected rubber, in an envi­
ronment of declining prices. 
1. Jayasuriya (1981) has argued that government policies became relatively more favourable towards the 
domestic food crop sector during the late 1950's and thereafter ; however, this only highlights the 
discrimination against export agriculture. 
6 7 8 
The deterioration of rubber and tea prices and rising domestic inflation during the 
late 1950's were reflected in growing balance of payments difficulties. These, together 
with the emphasis on import substituting industrialisation as the strategy for economic 
development led to increasing trade restrictions on imports. By 1962 such restrictions, 
operating both through high tariffs and more importantly through quantitive restrictions 
had become quite pervasive. This is not the place to go into an exhaustive discussion of 
the experience of the Sri Lankan experience with import substituting industrialisation 
behind high levels of effective protection. It should be noted, however, that while the 
rate of growth of the industrial sector was fairly rapid and higher than that of gross national 
product (GNP), the rate of value added has been relatively low and the import content 
of many new industries has been quite high. Industrialisation was quite " thin ", con­
fined generally to light consumer goods and catered mostly to the small domestic market. 
By the end of the 1960's much of the scope for import substitution in major sectors was 
already exhausted. 
The shift towards this particular kind of industrialisation necessarily involved dis­
crimination against agriculture, particularly against export oriented agriculture. Changes 
in domestic producer incentives are influenced by relative prices. Tariff and non-tariff 
barriers to imports raise their domestic prices, increase profitability of investment in their 
production, and distort the relative prices in their favour vis a vis other commodities. 
In the Sri Lankan context, import substitution was most significant in the industrial sector 
and the export agricultural sector was the most adversely affected. Fernando (1979) 
has reported on levels of protection (and their consequences) in some industries. The 
effective rates of protection 2 were often over 100 percent, and rates as high as 400 per 
cent were not uncommon. Tariffs were low on imported raw materials and intermediate 
goods ; these contributed to the import intensive character of the industrialisation by 
discouraging backward linkages. We do not have precise estimates of the import content 
of rubber production in Sri Lanka ; the artificial cheapening of imports may have lowered 
the costs of these, but it is unlikely that this would have had any appreciable favourable 
impact as the import content was low. In sharp contrast to these industrial protectionist 
measures, export taxes on the traditional plantation crops were quite high (except when 
prices were extremely low). These export taxes were the main source of government re­
venue. Table 4 presents data on export taxes in rubber during the 1970's. From very 
low levels during the early 1970's when prices were highly depressed, they have risen to 
over fifty percent in recent years. 
Impact of export taxes on profitability 
Tables 5 and 6 present some indications of the impact of export taxes on profitability. 
They are presented to illustrate the general trends, and are not meant to be accurate esti­
mates of actual rates of profits. The cost of production figures are averages and their 
basis is not known to us. Despite these limitations, they do provide interesting infor­
mation for the period 1970— 1981. 
2. These refer to the EPR estimates given in Fernando's paper. 
679 
Table 4. Export taxes (including levies, cesses etc. on rubber) 
Total tax Value of exports Tax as percentage 
Year Rs.-million. Rs.-million of export value 
1970 70 439-7 16 
1971 14 306-8 ' 5 , 
1972 . •12 265:1 . 5 ' 
1973 141 5 591 5 24 
1974 . 284 3 738-5 38 
1975 139 6 653-7 21 
1976 197 4 889-6 22 ' 
1977 ' 260 6 930-6 '. 28 
1978 1001.0 2020 0 50 . 
1979 1239 0 2491-4 50 , 
1980 1368-5 2590-4 54-
1981 : . : 1432-5 ; 2889 0 50 .. 
Sources : (i) Annual'Review of the Economy, Central Bank of Ceylon, various issues 
(ii) Government Financial Statistics (IMF) — various issues. 
Table 5. Profitability of f.o.b. prices 
Average • Average cost of Rate of Rate of profit 
Year f.o.b. price production ' profit '* in terms of 
(Rs./Kg) -: (COP) (Rs./Kg) . 1970 figure 
1970 ... 2-73 1 52. 80 100 
1971 . 2 38 1 67 30 . 38 
1972 - • . 2 0 5 1 65 24 30 
1973 . 3 68 1 87 97 121 
1974 " • 5-75 2-46 134 . . 168 
1975 4 0 5 2 75 ' .47 59 
1976 6 50 3 28 . 98. 123 > 
1977, 6;85 3-74 83 104 
1978 14-90 4 85 .' 207 259 
1979 19-44 6 86 183 229 
1980 21-42 8-20 - 161 202 
1981 .•• 21-80 8 92 '. 144 180 
. f.o.b. price — cost of production 
*Rate o f ' profit' == : - - - - - - - • — X 100 
cost of production 
^Source : Annual. Review of the Economy, Central Bank of Ceylon, various issues. 
680 
Tabic 6. Profitability at domestic producer prices 
Average* Average cost Rate of Rate of pre 
Year domestic price of production ' profit '+ in terms of 
(Rs./Kg) (Rs./Kg) figure 
1970 2 29 I • 52 51 100 
1971 2-26 1 67 35 69 
1972 1 95 1 65 18 35 
1973 2 8 0 J 87 50 98 
1974 3 5 7 2-46 45 88 
1975 3 20 2-75 16 31 
1976 5 07 3 2 8 54 106 
1977 4 93 3'74 31 61 
1978 7'45 4 85 53 104 
1979 9 72 6 86 41 80 
1980 9 85 8 20 20 40 
1981 10 90 8 92 22 43 
Average domestic price was computed as : 
Total export revenue — Export taxes, levies etc. 
Total volume of exports 
Average domestic price —• cost of production 
-i-Rate of ' profit' = • • — X 100 
Cost of production 
Source : Annual Review of the Economy, Central Bank of Ceylon, various issues.* 
Clearly, at the aggregate level, the profitability of the rubber industry is substantial 
from a national view point 3 ; the difference between domestic cost of production (COP) 
and the border (f.o.b.) price has been very significant throughout the I970's. Even before 
the exchange rate reforms and the associated devaluation in 1977, the rate of ' profit' 
was never below 30 percent and ranged as high as 134 percent in later years ; it has been 
always higher than 100 percent. However, producer profitability shows a different picture 
due to the export taxes which sharply reduce domestic prices below the border price. 
The level of the total tax (including levies and cesses) as a proportion of export revenues 
reached 50 percent in 1978 and was maintained in subsequent years. The resulting 
impact on producer profits has been quite severe (Table 6)'1. Further, rubber estates 
were subject to the usual company taxes, while many new industrial and non-traditional 
enterprises were granted generous tax holidays ; this tax would have affected relative 
profitability substantially. 
3. More accurate estimates of the social profitability could be obtained by computing domestic resource 
costs ; however, data constraints did not permit this. 
4. As a small fraction of the levy is transferred back to the industry in the form of replanting grants, re­
search, expenditure etc, this figure over estimates the impact of the tax on producer profits. On the 
other hand, marketing and storage costs may not have been included in the COP. 
681 
An examination of the profitability rates show that when the border prices are con 
sidered the industry has become substantially more profitable during the end of the de­
cade compared with the early years ; this is due to better prices and very importantly to 
the more favourable exchange rate since 1977. However, the producer profitability shows 
a different situation. Once export taxes are taken into account, the profitability in later 
years is markedly lower. Even in 1978, the exchange rate adjustment only arrested the 
declining profitability in previous years, restoring it to the 1970 level. In subsequent 
years the decline has been sharp ; by 1981 it was only 43 percent of the 1970 level. Clearly, 
the impact of all the policy changes in 1970's did not halt the declining trend in producer 
profits. This point will be taken up again later. 
The exchange rate policies 
The exchange rate policies further aggravated the bias. Adverse movements in the 
terms of trade (which deteriorated by an average annual compound rate of 5 5 percent 
during the 1960's) resulted in a continuing trade imbalance. This occurred despite import 
restrictions and even a slight upward movement in import volumes. Though there wa6 
a widespread perception that the currency was overvalued, no significant change in the 
(fixed) exchange rate was undertaken until November 1967 when the Rupee wa6 devalued 
by 20 percent in the aftermath of the devaluation of the British pound. In May, 1968 
a dual exchange rate system, the Foreign Exchange Entitlement Certificates (FEEC'S) 
system, was introduced and was maintained until 1977. 
According to the FEEC'S system, certain foreign exchange receipts and payments 
were subject to a premium rate which rose from 44 percent to 55 percent in 1969 and 
to 65 percent in 1972. While some investment and raw material imports were brought 
under the scheme, many export goods, particularly non-traditional commodities were 
permitted to obtain the higher rate (Kappagoda & Paine, 1981). However, tea, rubber 
and coconut were excluded from this category. The exclusion of the plantation crops 
from the FEEC'S scheme, which gradually broadened to include a large proportion of 
the foreign trade sector, was a heavy implicit tax on these crop sectors ; it exacerbated 
the effects of the already substantial export tax and the positive protection granted to 
industry. 
Employment and efficiency consequences 
Thus, both the trade regime and the exchange rate policies operated to reduce pro­
ducer incentives in the export crop sector ; these adverse effects were not balanced by the 
research, extension and fertiliser subsidies which returned a small proportion of the export 
taxes back to the industry. 
These policies had important consequences. They encouraged the development 
of a large group of inefficient industries. In the state sector, these took the form of a 
larger number of ' white elephants' which needed continual injections of new funds for 
survival despite the high levels of protection and other implicit subsidies. Many of the 
new private sector industries, too, never managed to outgrow their infancy, catering to 
a limited home market with little prospects of surviving without protection. However, 
these constituted more profitable investment alternatives, for private investors pulled 
682 
investment capital away from plantation crops. Being generally more capital intensive, 
the protection granted to these industries would have raised returns to capital and shifted 
relative prices to the disadvantage of labour. 6 Further, their capital intensive nature 
would have inevitably led to employment levels below what could have been achieved 
otherwise. The overvalued exchange rate which reduced the price of imported capital 
goods strengthened this anti-employment effect. 
In addition to the protection granted to import substituting industries, there was an 
active policy of providing added incentives to the so called non-traditional agricultural 
crops. Planting subsidies, subsidised credit and FEEC'S incentives were made available 
to these crops. Again, the effect on relative profitability was similar, though there was 
little growth in the production and export of these crops in the 1960's, though some crops 
registered substantial expansion during the 1970's. The industrial protection sometimes 
acted to reduce exports of some of these minor crops (such as cocoa) by diverting them 
into home market oriented production. Also, despite the incentives of the FEEC'S, 
the relatively high export taxes discouraged official exports (Wijewardana, 1973). Further 
incentives had to be granted to obtain significant growth, and later these export taxes were 
drastically reduced for most such crops. 
The marked decline in replanting in the medium estates sector during the 1960's 
(Table 2) was probably a clear reflection of the movement of domestic capital away from 
the rubber industry into these new, more privately profitable investment outlets. The 
comparatively better replanting rate achieved by the smallholders can be understood if 
we take into account the fact that rubber replanting was a substantial labour investment 
for them rather than a capital investment; even the larger smallholders probably found 
investment in land to be sufficiently attractive, given the land market imperfections and 
the often lumpy investments needed to enter into the new industries. Cost pressures, 
too, were perhaps lowest in the smallholding sector which had greater flexibility in terms 
of labour use. 
The rationale for discriminatory policies 
The discriminatory policies towards export agriculture, or rather the favourable 
policies towards industrial protection and non-traditional crops were justified on a number 
of grounds. The most commonly advanced reasons included the desirability of industrial 
development as a necessary pre-condition for sustained economic growth and the need 
for diversification of the economy to reduce dependence on a small number of primary 
goods whose prices fluctuated violently around a declining trend. These arguments 
have been advanced elsewhere and in different contexts ; both economic theory and actual 
experience suggest that the proponents of these arguments often failed to realise the nature 
of the premises on which they were based or the consequences which almost always followed. 
Of course, groups who gained economic rents from protectionist policies exercised much 
political pressure to enforce and maintain these. 
5. This follows from the Stopef-Samueison theorem ; for a discussion, see Chipman (1966). 
683 
Here, we shall lake up some of the issues raised ; for a more comprehensive review 
of the theoretical issues the reader is referred to Corden (1974). The empirical experience 
of import substituting industrialisation has been analysed extensively (see, for example, 
Krueger, 1978). 
The need for special treatment, for tariff protection, tax holidays and similar fa­
vourable policies for industries has been often based on the so-called " infant industry 
argument ". While presented in many forms, it usually involves the proposition that 
new industries cannot be expected to originate and develop if they are immediately subject 
to international competition ; they require time for "learning" how to produce and 
market their products efficiently. Further, capital market imperfections in many less 
developed countries cause under investment in socially desirable industrial development ; 
lack of relevant information denies potential entrepreneurs the knowledge needed for 
assessing and choosing profitable new industrial investments. None of these considera­
tions, however, necessarily justifies specially favourable policies for industry at the ex­
pense of agriculture. As Corden (1974) points out : " In less developed countries private 
agriculture usually finds capital harder to obtain and more costly than industry, and 
surely a lack of information or foresight is a greater weakness of farmers than of urban 
industrialists. Furthermore, the scope for learning in agriculture is often immense ". 
The latter point is particularly strong where new technologies are becoming available in 
agriculture. In any case, the first best policies to overcome capital and information mar­
ket imperfections would be the removal of barriers to access needed capital and the im­
provement of information supply. The industrial protectionist policies impose both 
static and dynamic efficiency losses on the economy. In the Sri Lankan case, as else­
where, the diversion of scarce capital and skilled manpower to the establishment and 
perpetuation of inefficient industries in both private and state sectors has certainly im­
posed heavy costs by stifling economic activity and restraining the development of those 
sectors where the country had a real comparative advantage. 
There are few strong arguments for government intervention to facilitate diversifi­
cation in agriculture. Lack of information about future price movements may hold back 
farmers from shifting to more profitable crops. This argument appears quite weak in 
the case of annual crops, as adjustments can be made relatively rapidly once the price 
changes takes place. However, there may be some grounds for it to apply in the case of 
perennial crops as factors other than information alone (such as time preference rates 
which differ from the social discount rate) may prevent farmers from responding in the 
desired manner. The second argument is that greater stability of incomes is desirable 
and this would necessitate diversification. 
Here it must be stressed that diversification for income stability is not costless ; its 
costs must be weighed against potential benefits as income stability may be brought about 
at the expense of higher average incomes if the changes require a shift away from the most 
profitable crops. On the basis of the Sri Lankan experience, we would argue that there 
is no compelling evidence that average incomes may have improved during the past decade 
as a result of crop diversification brought about by artificially distortingproducer.incentivcs. 
684 
It is a fact that the export composition exhibited greater diversity at the end of the 
1970's. However, to a considerable extent the reduction of the share of the traditional 
plantation crops in export revenues was brought about by the stagnant or even declining 
output of tea and rubber and the greatly reduced coconut exports. The sectors which 
grew more were the highly import intensive petroleum refinery operations and the made 
up garment industry and these contributed the bulk of increased industrial exports. In 
Table 7, we present the behaviour of selected minor export crops prices relative to rubber 
price ; note that the 1970 rubber price was low. Clearly, the relative price trends show 
no uniform pattern. While cardamom and cinnamon relative prices moved up in general, 
cocoa, pepper and cloves prices moved down. In terms of potential technological change, 
rubber was best placed at the beginning of the period ; hence it could be argued that high 
yields resulting from replanting could have compensated even where relative price move­
ments were somewhat unfavourable. In any case, without a much more disaggregated 
and careful study there is no room for asserting that the distortion of incentives to favour 
minor export crops has helped increase national income during the 1970's. It is quite 
likely that the gains from greater diversification were substantially offset by efficiency 
losses arising from price distortions. 0 
Devaluation and profitability 
The change in policies in 1977 was expected to substantially improve the prospects 
for the export crops. The 1977 devaluation (and exchange rate reform) immediately 
raised producer profits, despite the progressive tax structure which siphoned off half of 
the export proceeds as government revenue (Table 4). With the new exchange rate 
policy which signifies an end to an overvalued currency, it was hoped that a new, much 
more favourable period for exports, including rubber, had dawned. This was expected 
to improve long term profitability expectations and to result in higher levels of private 
investment in rubber. On closer examination, however, this appears not to be the case 
and raises the issue of whether devaluations stimulate growth in perennial crop industries. 
The major effect of a devaluation on the export sector is the change of domestic relative 
prices in favour of (internationally) traded goods vis a vis non-traded goods. Unless 
the country has market power in international trade, which is not the case with Sri Lankan 
rubber, there will be no effect on international prices. Recent studies of the experiences 
of devaluation have shown that much of the effect on relative prices of traded and non-
traded goods is eroded quite rapidly (usually within 2 to 3 years) as the non-traded 
goods prices rise (for a discussion of some of these issues, see Branson, 1983); this then re­
moves the original improvement in profitability. This is confirmed in the case of Sri 
Lankan rubber after the 1977 devaluation. 
From Tables 5 and 6 we can see that within a comparatively short period of time 
the improved profitability resulting from the 1977 devaluation was lost. Without the 
export tax effect, the "profit" rate even in 1981 was significantly higher than in 1977. 
However, once the tax effects are considered, producer profitability was significantly 
lower in 1980 compared with 1977. This decline was not due to a decline in rubber prices, 
6. This discussion does not imply that rubber should not be replaced by other crops when the agro­
climatic conditions are clearly unsuitable for rubber. However, the overall incentive structure is 
baised against rubber, the private investors' choices will lead to replacement of rubber by other 
crops even where rubber may be more socially profitable. 
685 
as average f.o.b. as well as domestic prices rose during this period. Indeed, if there was 
no price rise the effects of the devaluation on profitability would have been negated even 
sooner. In Table 8, we show how rapidly profitability would have declined, if 1978 prices 
continued ; the producer profitability levels would have been completely eliminated bv 
1980. 
Table 7. Prices of selected minor export crops relative to rubber prices 
Year Cocoa Cardamoms Pepper Cloves Cinnamon* 
1970 100 100 100 100 100 
1971 103 109 27 128 101 
1972 110 154 42 144 93 
1973 121 165 101 117 138 
1974 96 262 98 127 150 
1975 90 211 56 70 92 
1976 96 172 49 89 132 
1977 41 101 56 73 131 
1978 54 110 88 114 129 
1979 65 126 128 85 103 
1980 106 159 126 70 191 
1981 113 177 148 61 118 
* All relative prices have been scaled such that 
1970 price of minor export crop 
, . = 100 
1970 price of rubber 
@ (a) 1970 — 1972 — average price of all cinnamon products 
(b) 1972 — 1981 — price of cinnamon quills 
Source: (a) 1970 — 1972— and 1980 — 1981 Annual Review of Economy, Central 
Bank of Ceylon 
(b) 1973 — 1979 — Economic and Social Statistics of Sri Lanka, Central Bank 
of Ceylon. 
Table 8. Changes in profitability after 1977 devaluation at 1978 price 
Year 
1978 
1979 
1980 
1981 
* Data not adjusted for variations in exchange rate. 
Rate of ' profit' Rate of ' profit' 
at f.o.b. price at domestic price 
207 53 
147 9 
103 _ ve 
686 
Clearly, rising costs of production due to domestic inflation were eroding profitability 
at a pace that restricted the positive effects of the devaluation to a very brief period. Thus 
if industry profit levels are to be maintained, a substantial revision of the tax structure 
appears to be of crucial importance. Such a revision should link tax rates to the price 
as well as to domestic cost increases. In a general sense, devaluations appear to do little 
to stimulate long run growth in the tree crop sector (or in any industry with a long gesta­
tion period). In rubber, tapping begins after 5 — 6 years and economic yields, even later. 
Such investments will not be made in response to short lived profitability improvements ; 
producers' response will be to maximise short run gains by intensive tapping (unless they 
expect a continuous series of devaluations). 
Conclusions 
From the available data and information, wc would conclude that unless quite 
drastic measures are taken to increase and maintain profitability in the industry, the future 
outlook may be bleak. We understand that a substantial rise in replanting grants is 
being implemented. As replanting has responded quite positively to increased real levels 
of replanting grants in the past, this may hold hope for increasing the replanting rate, 
which has already shown some improvement at the end of the 1970's. Increased expen­
diture on replanting grants, financed from export taxes, will return a greater proportion 
of tax revenues to the industry. It will also raise expected profits for those contemplating 
replantings now ; however, it is unlikely that higher replanting grants alone can maintain 
producer profits in the industry. Both the export tax structure and the exchange rate 
policy would need to be revised and the progressive tax structure should be adjusted to 
take into account not only the woild price movements but also the changes in domestic 
cost structure. 
Future performance of the rubber industry will be affected by the distinction bet­
ween the state owned large estate sector and the privately owned smallholding sector, 
whose cost structures are likely to be substantially different. If the state owned estates 
are not expected to act as profit making enterprises, then plan targets can be set and imple­
mented without reference to export taxes etc, provided government funds are channelled 
to meet the needed expenditures. Such injections of funds will be essential if trends ob­
served in the study period continue. There are major problems that may arise in imple­
menting such a policy, however. It could easily breed inefficiency, corruption and mis­
management and more white elephants may be created in the state sector. 
On the other hand, if they are to operate as profit making enterprises, the same in­
fluences which determine profitability in the private sector will operate. The fortunes of 
the entire industry will then depend on market forces which are heavily distorted by govern­
ment policies. Due to the aggregated nature of the data available to us, we were unable 
to address the issue of variations in efficiency across and within subsectors. In formulating 
appropriate policies for the future, an analysis of the efficiency of the different subsectors 
should complement the other studies on tax, exchange rate and diversification policies 
impacting on the rubber sector. 
687 
REFERENCES 
BARLOW, C. (1978). The Natural Rubber Industry. Oxford University Press. 
BRANSON, W . H. (1983). "Economic Structure and Policy for External Balance". 
IMF Staff Papers 30, No. 1. 
CHIPMAN, J. S. (1966). " A Survey of the Theory of International Trade : Part 3, The 
Modern Theory ". Econotnetrica 34, 30, No. 1. 
CORDEN, W . M. (1974). Trade Policy and Economic Welfare. Clarendon Press, Oxford. 
FERNANDO, T. S. N. (1979). " Effects of Protection on Backward Linkage in Industry ". 
Staff Studies, Central Bank of Ceylon. 9, No. 1 & 2 
HERAT ii, H. M. G. (1975). " A study of Supply Response of Rubber in Sri Lanka". 
Unpublished Masters Thesis, Australian National University. 
JAYASURIYA,. K. (1981). " Politics, Class and the Internal Terms of Trade : An Explora­
tion in the Political Economy of Sri Lanka". Unpublished B.A. (Hons) Thesis, 
University of Western Australia, Perth. 
KALDOR, N. (1958). As cited in Wijcsinghc, M.E. (1976). The Economy of Sri Lanka 
1948— 1975. Ranco Printers and Publishers Ltd., Colombo. 
KAI'I'AGODA, N. AND. PAINE, S. (1981), The Balance of Payment Adjustment Process: 
The Experience of Sri Lanka. Marga Institute, Colombo. 
KRUEGER, A. O. ((1978). Foreign Trade Regimes and Economic Development : Liberaliza­
tion attempts and consequences. National Bureau of Economic Research, New-
York. 
WUEI'ALA, G. AND SENANAYAKA, Y. D. A. (1970). "A. study of Ceylon's Rubber Re­
planting Subsidy Scheme " 77/e Journal of the National Agricultural Society of Ceylon 
7. 
WIJEWARDANA, D. (1973). "The Problems of a Multiple Currency System in a Deve­
loping Economy with special Reference to Sri Lanka". Staff Studies Central Bank 
of Ceylon. 3, No. 1. 
688 
SESSION 12. P L A N T E R S ' V I E W S 
COMPARATIVE VIEW OF MANAGEMENT IN THE 
PLANTATION INDUSTRY IN SRI LANKA AND MALAYSIA 
By 
M. P. G. S. JAYAWARDENA 
(Blenheim Estate, Telok Arisen, Malaysia) 
Effective and efficient management is the key to maximising productivity ; even 
Socialist Nations have recognised this fact and are known to be introducing relevant 
Management Training Programmes. ' 
This paper will endeavour to deal with Contemporary Management in the Plantation 
Industry with particular reference to my experience gaired in Sri Lanka and in Malaysia. 
During the last 30 years or so, there has been a very fundamental change in the Plan­
tation Industry of practically all Countries having Plantation based economies. 
This fundamental change that has taken place has been the transfer of vast assets 
consisting of Plantations from our ex-colonial masters back to Government Controlled 
Agencies of the Countries in which the land is actually situated. This change in owner­
ship of the land and in other cases, the redomiciling of Companies owing the land has 
inturn inevitably resulted in structural changes in Plantation Management in Countries 
like Malaysia, Indonesia and Sri Lanka. 
The method and style of change in ownership has varied from Country to Country. 
In some Countries like Indonesia, the changes have been dramatic to the extreme. In 
other Countries like Malaysia and Liberia, they have been more gradual but whether the 
change in ownership has been dramatic or gradual these changes have inevitably resulted 
in varying degrees, in the style and method of Plantation Management. 
Method of change in land ownership 
In Sri Lanka, both foreign owned and Sri Lankan owned Plantations were nationalised 
in 1974 and 1975 in 2 stages. Stage 1 being Sri Lankan owned Companies and private 
owned lands with certain limitations and Stage 2 being foreign owned Companies. 
Indonesia also had the same experience as Sri Lanka where the Plantations were 
nationalised immediately after their Independence. Recently, foreign investors are 
being encouraged to return and experienced Planters, even after retirement, are being 
recruited to manage and train their personnel, to make the Agricultural Sector more 
effective and productive. 
In Malaysia, the changes in ownership of Plantations was part of the New Economic 
Policy which provided for a gradual reduction in the foreign equity over a period of 20 
years, commencing 1970. These changes provided for a reduction in foreign ownership 
of Plantation assets from 100% to a maximum of 30%. The acquisition of equity in 
foreign based Plantation Companies, resulted in the redomiciling of those foreign owned 
689 
Companies by purchasing shares in the open market in Malaysia and abroad. There 
has been no effect on the Malaysian owned Plantation Companies. 
Change in management 
Although there were changes in ownership of Companies in Malaysia, the Manage­
ment Personnel were retained intact and continued with the policies and systems adopted 
by the respective Companies. The Government as far back as 1967, discussed with the 
Foreign Companies and due notice and guidelines were given to reduce the expatriate 
executive staff by the process of natural wastage over a period of 18 years. Thereafter, 
only key personnel would be treated as an exception. 
However, in Sri Lanka the established Companies managing plantations became 
totally ineffective by Plantations being vested in the state and upto October 1980, the 
Management of Estates and its Managing Agents have been put to severe strain by frequent 
convulsive changes. These changes indicate that policies affected by the change in owner­
ship had not been properly planned prior to implementation. 
Contemporary management 
Morale — Key factor to effective management 
For efficient and effective management, the morale of the Management Team is a 
key factor. The retention of the same Management and the continuity of the systems 
practised prior to change in the ownership greatly aided in maintaining the morale of 
Malaysian Planters and indeed the entire workforce. On the other hand, the frequent 
changes in the Management structure have caused a sense of insecurity as a consequence 
of which morale of the Sri Lankan Planter has declined. This will inevitably affect pro­
ductivity of plantations, very much managed by the Planter — " The Man On the Ground," 
responsible for safeguarding and even enhancing the largest source of Foreign Exchange 
Earning for Sri Lanka. 
Some of the key factors which should not be overlooked in maintaining or even 
raising the morale of the Planter are :— 
(I) Authority 
The Staff and the Workforce on the Estates must be aware that the decisions 
taken within policies and practices laid down and given out are those of the 
Planter. He must be seen to be the authority. The Planter's effective control 
over his Staff and Workforce is essential as the industry is labour intensive. 
Approximately 75 % of the work is still done manually. Tc this end, tbe Planter 
should have the ultimate authority in respect of recruitment, disciplinary 
action, including dismissal of non-executives at least. 
(II) Political Influence 
Political influence should never be allowed to creep in to Plantation Manage­
ment. Staff and Workforce should be aware of this. Planters themselves 
69*0 
must be confident that they are only responsible to their senior management 
and Board of Directors. A sense of security of tenure would necessarily moti­
vate Planters to treat their Estates as their ' Home '. Any feeling of being a 
' Modern Day N o m a d ' must equally necessarily destroy Planteis' morale and 
productivity is bound to suffer as a result. 
(III) Bureaucratic control 
Chain of command should be clearly defined and unnecessary red-tape should 
be eradicated to ensure effective management. Planters should be given au­
thority to decide within given bounds. 
It is important for Directors and Senior Management not to give the impression 
that they are ' Hard to G e t ' for discussion but to make themselves available 
to conscientious Planters enthusiastic with proposals for better achieving their 
targets. Building up of confidence between Planters and Directors would 
enhance morale of Planters. As we all are aware, " Management through 
discussion " has been proved to be successful over the years in meeting the set 
objectives and targets. 
(IV) Discipline 
As stated earlier, Staff and Workforce must be aware that the Manager is the 
final authority on the Estate, and that all decisions pertaining to disciplinary 
matters are within his sole purview. This must be included amongst policies 
laid down. This is an absolute necessity in any labour intensive Industry. 
Procedures regarding the handling of disputes must continue on established 
principles and practices and should, if possible, be made uniform throughout the 
Industry. 
(V) Recognition 
Recognition should be shown by way of adequate remuneration and fringe 
benefits, in keeping with not only the round-the clock responsibilities a Planter 
shoulders, but the way of life he is required to lead to maintain his position and 
authority. His position cannot be generalised with those of most other sectors 
of the economy. A Planter mainly lives in isolation and away from the com­
forts of urban living, in many instances necessitating living away from his family 
so as to provide urban education for his children. 
A well-motivated Planter would be one who is confident that besides his overall 
performance, including his initiative and drive, will be appreciated and recog­
nised when it comes to transfers and promotions. He should be able to look 
forward to a achieving greater heights of recognition ia the Industry. 
691 
(VI) Incentives 
Incentive payments should be related to productivity but bonus payments 
should not be generalised and should be related to the earnings of an individual. 
In this regard, performance should be judged not merely on productivity, as a 
particular estate may not have the potential to maximise yield any further. 
Management System 
The system of Plantation Management in many parts of South East Asia was derived 
from Sri Lanka, where the Plantation Industry has a history of more than 150 years. 
However, some Countries, including Malaysia, have attained sophistication. For 
instance, apart from Visiting Directors, Planting Advisors, Consultant Engineers etc., 
larger Companies in Malaysia have set up their own Research Departments, in addition 
to the National Research Institutes covering various crops. Many of these private 
Research Departments have developed their own high yielding clones/varieties. Some 
of these high yielding rubber and oil palm are presently available in Sri Lanka. 
An important aspect in this regard, is that Senior Planters who are considered specia­
lists in a field are invited by the Plantation Companies concerned to serve on the planting 
policy committee of the research Department to advise on such matters as the practicality 
of implementation. 
The full potential and the commercial value of a high yielding done or variety ob­
tained after years of research depends on the correct agronomic inputs and proper Manage­
ment practices, implemented by the Planter. 
Supporting Agencies 
A Planter today is over-burdened with a multiplicity of duties and responsibilities. 
He is often termed, " Jack of all trades". In order to afford him to spend as much t ; ms 
as possible in the field to maximise productivity, a number of commonly pooled supporting 
agencies are available in Malaysia, viz. The Malayan Agricultural Producers Association 
(M.A.P.A.), United Planting Association of Malaysia (U.P.A.M.), East Malaysian Plan­
ters' Association (E.M.P.A.), Malaysian Estate Owners' Association (M.E.O.A.) and 
Rubber Growers' Association Bhd. (R.G.A.B.). There are a number of technical 
committees of these Associations, Planters are invited to serve on them. This system 
is one of the main reasons for the success story of the Malaysian Plantation Industry. 
The functions of M.A.P.A. are restricted to trade union matters, wages and condi­
tions of service. The aim of the other Organisations are to protect, foster and promote 
the interests of the Plantation Industry. 
The U.P.A.M. is the main liaison body with Government Authorities on non-
labour matters. The Planters' Associations of States/Districts are represented at the 
U.P.A.M. Council at national level. The owners or proprietors are also represented 
in the Council. Similarly representatives of the local M.A.P.A. Advisory Panels of 
States/Districts attend M.A.P.A. Council Meetings. 
692 
An Estate security scheme established by R.G.A.B. with the blessings of the Police 
Authorities is in operation under the guidance of District Area Security Advisors. Through 
the Area Security Advisor in the District who is generally a retired police Officer, it is 
possible to maintain a close liaison with the Police Authorities in the District. Further 
the watchmen have been replaced by Auxiliary Policemen who have the authority to 
arrest even by hot pursuit, detain and question prior to handing over the culprits to the 
Police for action. This scheme also assists to keep a track of unauthorised persons in 
the area, thus giving a sense of safety to the residents on the Estate, particularly in the 
remote areas. .What is not required is a show of strength but an intelligensia to comple­
ment and assist the Law Enforcement Authorities. The introduction of such a system 
in Sri Lanka, would in the piesent context effectively meet the demand for a para-military 
force, made by the Trade Unions in the Plantation Sector. 
The functions of the above Organisations were covered in Sri Lanka by the Planters' 
Association (P.A.) and the Ceylon Estates Employers' Federation (C.E.E.F.). In fact, 
the last Executive Vice-President of the Malayan Planting Industries Employers Associa­
tion (M.P.I.E.A.) and the 1st Director of M.A.P.A. (successor to M.P.I.E.A.) was a for­
mer Secretary of the C,E,E,F. In Sri Lanka, currently there are no commonly pooled 
supporting agencies. 
Trade union legislation 
Malaysia has evolved effective labour and industrial relations laws. This is one 
of the main factors that led to the success of the Plantation Industry. Legislation res­
tricts organisation of trade unions to a particular trade occupation or industry or similar 
trades, occupations or industries. Hence, there are two main Unions in the Plantation 
Industry, viz. National Union of Plantation Workers (N.U.P.W.) and All Malayan Estates 
Staff Union (A.M.E.S.U.) for the workers and staff respectively. 
M.A.P.A. has succeeded in bringing about uniformity in wages and terms and con­
ditions of employment throughout the Industry. . It has further minimised areas of fric­
tion at Estate level and between Companies and even between the Planters and the workers. 
The Regional Industrial Relations Officer (I.R.O.) and Unions sort out most of the labour 
and industrial relations problems. Wages and terms and conditions of employment 
are decided at National level. All Agreements are incentive orientated with a profit 
sharing element included. In the case of rubber there is a fixed basic daily wage with 
two other main components ; one being incentive payment for latex brought in beyond 
a stipulated minimum per day and the other a price bonus which varies with the pre­
vailing price of rubber. The incentive payment also varies with the prevailing price of 
rubber. Some of the more important Agreements currently in force are as follows :— 
(i) Rubber Tappers Agreement — applicable to tappers. 
(ii) Oil Palm Harvesters Agreement — applicable to oil palm harvesters and 
loaders. 
(iii) Cocoa Harvesters Agreement — applicable to cocoa harvesters. 
693 
(iv) Coconut Harvesters Agreement — applicable to coconut harvesters. 
(v) Palm Oil Mill Agreement — applicable to palm oil mill workers. 
(vi) Central Installations Agreement — applicable to workers in rubber factories. 
(vii) Fringe Benefits Agreement — applicable to all workers, 
(viii) Field and General Workers Agreement — applicable to such workers in all 
crops. 
(ix) Retirement Benefits Agreement — applicable to all workers. 
These Agreements do away with the situation of Planters acting on their own an 
setting precedents. Further, they have ample time to concentrate on increasing pro­
ductivity. 
Staff terms and conditions of employment are covered by an Industrial Court Award. 
Productivity 
It is a well documented fact that for the achievement of high productivity, close 
co-operation and understanding between the Employer, Employee and Government is 
important. In Malaysia, this is a reality to a large extent. 
At this juncture, it will be prudent to have a comparative look at the development 
and productivity of the Plantation Industry in Malaysia and Sri Lanka. 
The major crops in Malaysia are rubber, oil palm, coconut and cocoa whilst in Sri 
Lanka, they are tea, rubber and coconut. The overall hectarage and yield per hectare 
in Malaysia in respect of the crops have been on the increase whilst in Sri Lanka, there 
had been a small decrease in the area developed and a decline in yield profile. The area 
under some of these crops, total yield and yield per hectare, between 1971 and 1983 are 
presented in Figs. 1, 2 and 3 respectively. 
Though in Malaysia, (Fig. 1) the hectarage under rubber has been stagnant, the 
hectarage under oil palm and cocoa have increased tremendously. The low preference 
for rubber against oil palm is due to the larger profit margin per unit area and the quicker 
returns. Furthermore most of the areas marginally suitable for rubber have been replaced 
with oil palm. The same cannot be said for Sri Lanka (Fig. 1) as both the hectarage 
under rubber and tea have remained unchanged. Similar trends can be seen in respect 
of total yield form the various crops (Fig. 2). Another important aspect to note would 
be that, the difference in yield per hectare of rubber between Malaysia and Sri Lanka 
(Fig. 3) is widening. 
Privatisation of management 
Although the basic concept of nationalisation was to benefit the Country, it may 
be timely to assess the results so far achieved. Low productivity in any area of economic 
694 
1,755,COO 
i ,625 ,0L}0 
i,:oo,ooo 
975,000 
650,000 
125,000 
i RUBBER (M) 
»OIL PSUN <M)V 
- / 
a. 
/ 
- o - - - 0 
COCONUT (:•!/ 
TEA i S / U 
. COCOA (M) 
-II- * ^ - t RUFCTER (S/T) 
75 7* 
Year 15 ' 
Fig . ' l . Source of data for robber, oil ralm and coconut for Malaysia - United 
Planting Association of Malaysia Annual Report 1983. Pg. 54. 
695 
•TCO'Al PKODOCTJON OF AGR1 CUt.TUPAf. 
CC-HTCiri' IN METRIC TON FROM 
1971-1983 (MALAYSIA a SRI LANKA) 
1,300 f-
Fig. 2. Source of data for cocoa for Malaysia - Malaysian Cocoa Growers Council Annual 
Report 1983. Pg. 20. 
696 
tektCULTUJ&L O K C U l W E S Y'IELB 
PER HECTARE P 3 P YEAR IN METRIC TCN 
J 9 7 1 - J . 9 8 3 (MALAYSIA < SRI LANKA) 
" - s r 19 ' 
Source of data for tea, rubber and coconut for Sri Lanka -
( Pers. Com.) Forbes & Walker Ltd., Sri Lanka. 
697 
activity is unhealthy for any Country, particularly so for a developing Country. Sri 
Lanka should re-assess the position to ensure that the Plantation Industry is set on the 
right path, so as to ensure that Agricultural Commodities the base of the National Economy 
can remain competitive in the world market. 
One of the most essential elements to maximise productivity is effective, independent 
and efficient Management. The Prime Minister of Malaysia has this to say on Manage­
ment :— 
" There is hardly an Industry in which costs cannot be lowered in one way or another. 
If careful analysis is made, wastage and inefficiency will certainly be discovered. 
From then on, it is a question of good Management", 
It is heartening to learn that in recent years (Pers comm), there appears to be new 
thinking with regard to Plantation Management. This augurs well for the future of 
Plantation Management in Sri Lanka. 
Nevertheless, it is prudent to consider re-privatisation of Plantation Management 
retaining a Central Board devoid of political intervention. Though the Plantations will 
be managed by private Companies or Agency Houses, the Central Board would be vested 
with the authority to ensure that the policies adopted by these Companies or Agency 
Houses are in line with Government Policies, such as :— 
(i) Decision on the most appropriate crops, based on the suitability of the land, 
agroclimate and social requirement of the region. 
(ii) Decision on the most appropriate intercrop based on the suitability of the land, 
agroclimate, micro-climate and social requirement of the region. 
(iii) Marketing policies to ensure maximum profitability. 
(iv) Legislation providing welfare facilities like housing, education, medical, etc. 
to the workers. 
(v) Ensuring that there is sufficient interaction between Estates and the neigh­
bouring villages, 
Privatisation of ownership 
In Malaysia, privatisation is being positively undertaken. Some Government De­
partments such as Television Stations, Railways, Telecommunications and Public Works 
are being contemplated for privatisation to improve efficiency and productivity. The 
Government has given an assurance that employees' interests will not be adversely affected 
in the process. Their future prospects would depend on their performance under priva­
tised Companies; Many other Countries are also actively encouraging the privatisation 
concept. This may be food for thought for the accelerated development of Sri Lanka. 
698 
In this context as the agricultural investment possibilities could be low on account 
of land ownership limitation, with the establishment of a Stock Exchange the Government 
could encourage equity participation from the private sector and the public as a whole. 
The funds derived could be used for further Plantation and other related Developments. 
Planter tomorrow 
In the Plantation Industry, another important element is the Planter ' The Man on 
the Ground' who is vital to ensure the achievement of set targets and to maximise pro­
ductivity. Thus, as part of any reorganisation the following aspects could be considered :-
Harnessing of Expertise 
In Sri Lanka, there is no shortage of experienced Planters, some of whom Bre in 
retirement. They are or may be operating on a very low key due to lack of recognition 
or for various other reasons best known to them. These sources could be harnessed to 
build up a core of Planters of good quality and also to undertake training programmes 
for Cadets as well. 
Selection of Personnel 
Selection of key personnel is of paramount importance, particularly in the Planta­
tion Industry. This need not be over emphasised. Tertiary academic-qualification 
should not be taken as the sole criterion for selection of Plantation Executives although 
this would be an added advantage, particularly with rapid advancement in science 
and technology. Some of the essential qualities to be considered are :— 
(i) Readiness and suitability to lead a life in a secluded environment away from 
urban comforts. 
(ii) Communicational ability. 
(iii) Alert and quick thinking quality. 
(iv) Ability to organise. 
(v) Qualities of leadership by example. 
Motivation 
An important aspect of management is motivation. Due recognition motivates a 
person to give of his best to maximise productivity. Remuneration alone, 
although important, does not provide job satisfaction. In this regard, regular visits by 
Directors from the Head Office to the Estates provide an opportunity to discuss 
problems and innovations in situ. A word of encouragement at such visits will no 
doubt be a morale booster to the Planters' initiative and drive. The recognition of 
a job well done can be no better booster of morale to a planter. Above all, such 
visits would also enable the Hierarchy of Management to judge the Planters perfor 
mance on the ground rather than solely through reporting arms. 
699 
Exposure 
A forum for discussion and exchange of views is an aspect that should be encouraged. 
This would allow Planters to learn and understand each other's problems. Government 
officials in the District and personnel from research and technical bodies connected with 
the Industry could be invited to such a forum which would be useful in bringing about 
better mutual understanding and would solve problems expeditiously. 
Such exposure of young Planters would assist in formulating their thinking on ad­
ministrative matters. Besides building up their self confidence, such fora will also give 
an opportunity to the Management Hierarchy to identify potential Leaders for their future 
requirements. 
Concluding remarks 
A good Planter is not only required to know the technical knowhow of planting and 
increasing productivity but he should also be required to know, among other things :— 
(i) Principles of Management. 
(ii) Labour Relations (including a working knowledge of Labour Laws, employee 
development and training). 
(iii) Financial Procedures and Physical Resources. 
(iv) Productivity. 
(v) Public Responsibility and Community Relations. 
(vi) Security of Crops and Property. 
(vii) Knowledge of other relevant Laws relating to land, taxation, safety, etc. 
To carry out these responsibilities well, he must be able to have a sound mental and 
physical state, as Oomen Mathew (1984) defines " Management combines man's mental 
efforts to identify, organise, decide, act and evaluate ". Such a group of Planters can 
be called as a group of people " who know, because they know" (Taylor, 1984), from the 
experience gained on the field. There is no substitute for experience. 
Every action, they say, has a reaction. Improve " The quality of life " to quote the 
words of His Excellency the President of Sri Lanka — whether it be that of the Planting 
Executive or that of the worker at the lowest level — and the quality of the Plantation 
must necessarily improve. 
Foreign financial aid will be useful to improve facilities on the Plantations but Ex­
patriate Consultants are not required to improve Management, as Managerial Expertise 
of the former Agency Houses is available among the Sri Lankans. 
700 
Finally, let it not be said that Sri Lanka which mothered and nurtured the Plantation 
Management in South East Asia has now been left behind by its siblings. It is never 
too late to regain that eminent position. It is with will and determination of those in 
power that this can be achieved. 
I thank Kuril Plantations Sendirian Berhad for the facilities afforded me to present 
this paper, and state that the views expressed in this paper are my own and does not reflect 
that of my Company. 
REFERENCES 
D B MEL, H . G . R. (1984). Pers. Com., Rubber Research Board, Sri Lanka. 
MALAYSIAN COCOA GROWERS COUNCIL, Annual Report 1983, pp. 20. 
OOMEN MATHEW (1984). Plantation Management — A perspective. Rubber Research 
Bulletin 1 9 (3) 16- 18. 
TAYLOR, J. G . (1984). Cocoa Estate Management — A personal view, Planter 60, 274 — 
286. 
UNITED PLANTING ASSOCIATION OF MALAYSIA, Annual Report 1983, pp. 54. 
WIJERATNE, D . H . T. (1984). Pers. Com., Forbes & Walker Ltd., Sri Lanka. 
701 
RESEARCH REQUIREMENTS FOR THE RUBBER 
INDUSTRY 
By 
C. N. M. RODRIGO 
( Janatha Estates Development Board I, Avissavella, Sri Lanka- ) 
Dr Peries — Director Rubber Research Institute of Sri Lanka, Scientists and Heads 
of Divisions of this Research Institute, Technologists and Research Personnel from the 
premier rubber growing countries of the world, Chairmen and Directors from the Na­
tionalised Plantation Organisations, Superintendents and Assistant Superintendents of 
estates administered by the Sri Lanka State Plantation Corporation (SLSPC) and 
JEDB distinguished visitors, ladies and gentlemen. 
It gives me great pleasure as a Chairman of a Regional Board in one of the State 
Organisations to address you today on the subject of " Research requirements for the 
rubber industry ". Prior to my isolating and evaluating the research requirements 
I should pay a tribute to the Director and the Board of the Rubber Research Institute 
of Sri Lanka for giving me this opportunity to express the views of the producer —• and 
also to extol the first class job done by the team from the RRISL for the organisation 
of this most successful Symposium —- which brings together under one roof the hierarchy 
that guides the destinies of the rubber industry in the International arena. The impor­
tance and significance of Symposiums of this nature cannot be expressed in words as 
unless such International Conferences become more frequent and conclusive — this 
massive Plantation Industry can be submerged ere long by the competition from techno­
logical development in the fast expanding synthetic field — which now appears to be 
over shadowing and substituting all that is natural. I am certain that most of those ga­
thered here are aware of the conclusions drawn during the past few years that rubber 
is the crop for the future and the financial profitability ' Boom " from this crop is just 
round the corner. But my contention is that possibly due to inadequate agitation and 
interest by those who constitute us from the natural Rubber World — this anticipated 
boom and prosperity is still to come and will it ever come is the question. In consequence, 
natural rubber growing is definitely much less profitable today than growing the parallel 
plantation crop tea. 
Hence, if this trend continues — and we are still lethargic and complacent as we 
have been — then crops like tea and oil palm will takeover from this great natural rubber 
industry in the world which has gone on uninterrupted for decades. Prior to my talking 
on the " Research requirements for the rubber industry " let me in brief evaluate and 
commend the RRISL and its affable Director — Dr Peries for the very successful 
manner in which he conducts his activities at the Institute — and also for the aspect of 
' working together' evinced and projected by him in all deliberations and decisions bet­
ween the Research Scientist in the Institute and the practical Rubber Planter on the Planta­
tion. This aspect is very important when only scientific technology and research can 
be effectively disbursed through the channels of the working Planters to the media of 
propagation — the Plantation itself. 
703 
RRIC iOO series clones. 
To talk in brief of the effective break-through in research already obtained — t 
-think the foremost should be the selection and isolation of the clone RRIC 121 which I 
am told is resistant to the dreaded ' South American' leaf disease — generally when a 
clone is isolated for a certain individual characteristic the aspect of yield depression on 
same is accepted. But the break-through i n this instance is all the more commendable 
as I am told that RRIC 121 is also high yielding as the other isolated clones of the RRI 
— 100 series. Sri Lanka still holds the majority of its plantations in the PB86 clone 
— which is susceptible to this South American leaf blight. 
Advantages — coming to bearing early 5 — 6 years —- high yielding and tree 
itself is generally robust and sturdy and generally resistant to disease. 
Soil foliar analysis 
Fertilizer applications, curtail expenditure on fertilizer applications, correct nutrient 
level maintenances— whilst maintaining high yields as well. Approximately 30% cost 
saving on fertilizer. For large organisations this is a substantial saving. 
Effective Chemicals 
Nexobleach — RRISL — water soluble bleaching agent, no smell and non pollution 
prone in consequence — Chemical Engineering Division should be commended. 
Collar protectant for Fomes. Effective advisory and research discovery resulted in 
Fomes now not being a serious problem from what it was — control and incidence 
reduced. Also back rot — Phytophthora eatlier problematic now no problem. Same 
with Oidium, Phytophthora leaf disease etc. 
Green budding and polybags 
Problem of no seed for 1986 planting season. 1983 planted runts and 85 seed — 
green bud for 1986 planting. Polybags—Plant and bud with clearing plants and use 
for resupplying of vacancies in the field about 10 — 15%—•uniformity of clearings by 
gaining the 1 year in the clearing. 
Tissue culture 
RRISL has practical problem of inadequate finances to progress in expensive tissue 
culture for plant propagation. Government should release more finances on an isolated 
programme of priority to help keep pace with Research and Technological Development 
as evident in other countries. 
Recommended research requirements 
1. Marketing of products manufactured, presently no market for sole crepe. Biggest 
producer in World of sole crepe. Marketing Institute for rubber and rubber 
products, diversification into other products — block rubber, no subsidy for Alfa 
Laval Centrifuge Separators and bowls — departments relevant should help and 
encourage diversification to new end products and not discourage those that desire 
to divresify to new products. 
704 
2. 65% Smallholders and 35% of Nationalised Plantations JEDB and SLSPC. 
Rubber smallholders products have to be perfected by better systems of manufacture 
and technology to eventuate in a quality export product in the consumer accepted 
form. Presently bulk, manufactured to only RSS. Encourage to give latex to 
large Government Plantations RSS — who should like in tea be advised to pay a 
remunerative rate for latex given on a predetermined formula. 
3. Sri Lanka rubber sent abroad and we buy rubber based products. This is a pity 
—open and encourage greater rubber based product manufacture in Sri Lanka— 
Free trade zone (FTZ) etc. Offer worthwhile subsidy for import of relevant 
machinery Alfa Lavals etc. 
4. Exchange and sharing of Rubber Technology and Research for advantage of the 
International rubber growing countries and not frog in well attitude. Exchange 
RRIC 100 series clones with say latest PB Clones from Malaysia. 3,000 Kg/ha 
100, 102, 103, 104, 117, 121, 131. (International Rubber Research and Development 
Board—IRRDB) 
5. Further reduce the period of immaturity. 
6. Stimulants and stimulation to derive weaker yields in shorter period and thereafter 
uproot and replant advantageous today with increased wages. 
7. Further development on rubber machinery which have in effect remained unchanged. 
8. Indonesians use Michie Golledge knife and tap 500 trees on S/2 d/2 system. Malay­
sia uses the Gouge knife & taps 500 trees, Sri Lankans use the Michie Golledge knife 
& taps on 250 trees. Why ? RRISL and all others should investigate and move 
towards achieving higher task; will result in more latex whereby harder work for 
greater prosperity of the workforce like in Malaysia cars—motorcycles etc. 
eventuate in worker welfare will then not need subsidising as is done now. 
7Q5 
E F F E C T I V I T Y O F R E S E A R C H A N D I T S I M P L E M E N T A T I O N 
I N T H E I N D U S T R Y 
By 
D. LAMB 
(SLSPC IV, Horana, SriLanka) 
This, the 75th year of the foundation of the Rubber Research Institute of Sri Lanka 
is a opportune moment to review the effectiveness of Research done by the Institute. 
Broadly the research undertaken could be classified as being of a purely scientific 
nature and research undertaken to solve specific problems which arise from time to time. 
The first i.e purely scientific research from a practical point of view could be said 
to be of limited use in an industry such as the production of rubber. But, such a state­
ment, it could be said is not wholly acceptable. Any research undertaken which results 
in expanding the knowledge of a product could be said to be useful. Such research may 
at the moment, it is done, be purely of scientific interest but in the course of time such 
research can become the spring board on which more practical research could be done. 
We then come to research undertaken to deal with specific problems and here the record 
of the RRISL is indeed enviable. In plant breeding, tremendous strides have 
been made in developing high yielding clones. The RRIC 100 series, is second to none 
when compared with clones bred in other countries. The IRRDB plant breeders have 
endorsed the suitability of the RRIC 100 series for large scale planting in Sri Lanka. 
On various tapping systems progress has been made which confirm the superiority 
of the s/2 d/2 system which is in wide use. Also experiments are in progress on puncture 
tapping and stimulation using Ethrel. Coupled to the tapping experiments research is 
being done on sensitivity of various clones to Brown Bast. These experiments have given 
a rich yield of results and the proposal to continue with them would be of immense benefit. 
This year has been particularly wet and the experiments on the use of rain guards should 
give us useful information. Research on the stock-scion relationship, we are informed 
have not given any significant result. On the other hand crown budding, was observed 
to have resulted in significant yield increases. 
Raising of plants in poly bags has indicated significant growth gains, but general 
implementation of raising plants has been rather slow. Green budding coupled with 
poly bagging has achieved near 100% success, when compared with bare root stump 
planting. Another advantage is the evenness of growth obtained with poly bag plants. 
The Plant Pathology Department continues investigations into Oidium and Phytoph­
thora leaf fall, barkrot and the susceptibility of various clones to these diseases. General 
findings have not shown significant variation. A cause for concern has been Bird's eye 
spot caused by Drechslera heveae in seedling nurseries especially during periods of drought. 
Research is continuing on control of these diseases. 
707 
What has so far been mentioned are just a few avenues of research conducted 
by the Plant Pathology Department. 
Significant advancement has been made on plant nutrition. Research programmes 
are being conducted into response of rubber trees to fertilizer, economics of fertilizer 
application, foliar analysis and the influence of leaf nutritional status on latex yield. The 
importance of cover crops together with developing analytical methods for the study of 
mineral nutrition. Fertilizer application is now based on soil and foliar analysis and 
significant savings have been possible on in-puts of fertilizer. The advantages of legu­
minous covers on the growth of the rubber plant are being evaluated. In this connection, 
mention should be made of a variety of winged bean which has been developed and could 
be grown as a cover crop. This is a significant break through as in developing countries 
where protein is always in short supply, this could prove to be a cheap alternate source. 
Further research into this would be of tremendous benefit not only to the rubber industry 
but to the people of this country as a whole. 
Today, we apply 100% urea based mixtures. The Planter himself both tea and 
rubber viewed this trend with misgiving but experimental evidence indicates that such 
fears are unfounded. Economically, even if we apply an additional 20% on the recom­
mended dosages, there would be still a tremendous saving. However, experiments are 
being continued on the use of urea together with the effects of P, K and Mg on growth. 
.Special.study is also being conducted on the application of NPK to ground covers 
and in most instances it has been established that in general growth of trees in legumes 
is superior to that in naturals. 
Recently, a major break through was made in the development of a water miscible 
bleaching agent commercially marketed as New Nexobleach. The research conducted 
by the Rubber Chemistry Department on liquid natural rubber, CV rubber, skim and 
cyclised rubber have proved to be of immense benefit. The various commercial uses 
of N R are being evaluated and expanded. The Biochemistry Department we are given 
to understand is investigating into the causes of brown bast, which incidence planters view 
with concern. 
Inter-cropping is a subject which continues receiving close attention. Bananas, 
passion fruit, pineapple are some of the crops being inter planted. The inter-cropping 
of papaw has not been'quite satisfactory. Planting of pepper between immature rubber 
is in experimental stage and it is still too early to fully evaluate. I have already touched 
upon the growing of winged bean (Dambala) as a crop. Cocoa in rubber is nothing 
new and further experimentation would be of tremendous financial benefit. 
The fore-going are just a few examples of research which has been done. The impact 
of such research has been significant in that by developing high yielding clones, produc­
tion should increase. Also with such increases cost should come down. The discovery 
of suitable bleaching agents has kept the crepe industry viable. 
708 
For any research to be truly effective, there should be early and wide spread dissemi­
nation of information. Here perhaps it could be said there have been delays or a break­
down in communication. It is normal when anything new is promulgated that such find­
ings are slow to be accepted. Immediately following nationalisation of estates and the 
creation of multifarious state organisations to manage these estates there was a break­
down in communication between the Research Institutes and the management. How­
ever, since 1977 with most State managed plantations coming under the umbrella of the 
JEDB and the SLSPC, once again useful dialogue has been established between the Re­
search Institutes and the planters. It is not unusual to now see teams of Scientists visiting 
factories and fields, attending meetings of planters, conducting work shops for members 
o f the field and factory staff, in order to give of their knowledge to those who would ulti­
mately put it into practical use at the very grass root level. This dissemination of in­
formation has already had an impact in bringing about a transformation in nursery tech­
niques in the field and in improving the general standard of manufacture. Development 
programmes are also being drawn, harnessing the invaluable wealth of knowledge and 
experience available in our Research Institutes. 
There is one other matter which I would like to touch upon, namely Market Research 
and Intelligence. This may be outside the purview of research conducted by an Institute 
such as RRI but market intelligence is an essential ingredient for future research and 
without such information it is not possible to plan a strategy for the future development 
of the crop in Sri Lanka. Today we produce the best crepe rubber in the world, which 
is the highest grade of N R available. Little information is available on its end uses. 
Why does Russia suddenly pull out of the market and why have the USA and U K which 
were the main consumers of crepe rubber reduced on their purchases. It would be rele­
vant to mention that a country such as Malaysia has 15 Bureaus in the world, which 
feed back to their producers and Research Institutes the demand for various types of 
N R and even go to the extent of arranging direct sales, between the producer and the 
manufacturer of the end product. The greatest competitor of N R is SR. It is up to 
us to take the battle into areas which has been solely dominated by synthetic rubber. 
This can be done with correct market evaluation and with new methods of processing 
N R to meet specifications required. You will therefore see the link which could be es­
tablished between market research and intelligence and the expanding use of N R into 
avenues which have been the sole preserve of synthetic rubber. 
Recently we had one of the major buyers from the continent visiting factories in Sri 
Lanka and following discussions with him a rubber was produced which he felt could 
be used exclusively for medical purposes. Certain specifications and guarantees had to 
be given and with the assistance of the RRI this particular type of rubber was manufac­
tured which gave us a premium price of Rs. 1/- per kg. over and above the top IX price. 
In concluding I quote from Dr O. S. Peries's Annual Review for 1980. Therefore 
the policy makers the top management on estates the advisory services to small holders 
and the Research Staff of this Institute have a duty by this country to develop NR to be 
the top money earner for Sri Lanka. This can be done. It only needs clear thinking, 
bold planning and the will at the top to ensure that the plans are implemented. 
13th September, 1984. 
709 
THE PROBLEMS AND PROSPECTS OF THE RUBBER 
INDUSTRY IN SRI LANKA 
By 
M. R. C. PBIRIS 
(Janatha Estate Development Board J, JEDB, Panawatte Estate, Sri Lanka) 
Sri Lanka was the fourth largest producer of Natural Rubber for a long time, next 
to Malaysia, Indonesia and Thailand and in 1970 produced the maximum ever at 156,644 
MT. Production gradually declined thereafter and in 1982, we produced only 128,144 MT 
while India with 165,000 MT and China with 140,000 MT, pushed Sri Lanka into the 
sixth place in the world Rubber Production Arena. The decline in production i6 attri­
buted to numerous reasons, such as nationalisation, backlog in replanting under the 
Subsidy Scheme, decrease in extent and also poor prices for RSS, which discourages 
the smallholder from maximum exploitation of his trees. 
70% of the extent under rubber in the island is still under private ownership and 
of this 54% consists of smallholdings. Therefore, the State Organisations control only 
30% of the extent under rubber. 11 % of the Country's export earnings is from rubber, 
which is equivalent to 4% of the Government's budgetory revenue. 87% of the pro­
duction is exported and only 13% is utilized by local industries. 
Forecasts by experts indicate that global output of rubber would increase annually 
at the rate of 3-6%, while the demand would grow at 4 8%, thereby assuring a steady 
demand for natural rubber. 
Sri Lanka will also endeavour to take advantage of this increasing demand for natural 
rubber. New development programmes have been formulated under the Medium 
Term Investment Programme for State owned estates during 1985/89, when it is envi­
saged that 12,500 ha would be replanted annually by the Sri Lanka State Plantation 
Corporation (SLSPC) and JEDB, with high yielding clones, such as the RRIC 100 series 
evolved by the Rubber Research Institute of Sri Lanka. The private sector will also step 
up the rate of replanting assisted with World Bank Aid. 
The Development Programme also envisages the modernisation of factories and 
grouping estates to facilitate the operation of central factories, which would concentrate 
on production of high quality sole crepes and blanket crepes. The demand for sole 
crepe fluctuates and producers of this grade of rubber are aware of the difficulties they 
have to face, when suddenly there is no demand or contracts forthcoming. Why should 
we face such uncertainty ? Is it because we have no control over the marketing strategy V 
Should we producers not search for and establish contracts with consumers of this product 
and ensure a steady supply and demand position at a reasonable price, which would be 
beneficial to the producer as well as the consumer ? Long Term Contracts, say for a 
period of 6 months, when 50 % of the sole crepe production of an organisation could 
be offered, could also tend to stabilize price6 and assure a steady demand. With the 
increase in production envisaged, this important factor should receive immediate and 
serious consideration by the state ; if not, the millions spent on factory development 
would not achieve the desired result. 
711 
Technically specified rubber (TSR) could very well erode into the latex crepe market 
and here too, long term contractual sales would support the demand for latex crepe. 
Latex based products for export 
The need to encourage the conversion of raw rubber into value added finished goods 
for export cannot be over emphasised in order to utilize the increase in production anti­
cipated, as a result of the accelerated replanting programme. 
This can be achieved by attracting foreign firms with expeiience in latex product 
manufacture and with access to ready markets abroad, to set up industries in Sri Lanka, 
in collaboration with the JEDB and SLSPC. 
Joint ventures have many advantages since they eliminate marketing problems and 
the need for specialised technical training and also, after a period of time, such joint ven­
tures could become 100% Sri Lankan owned. The basic raw material for such industries 
is centrifuged latex and this raw material of acceptable quality, should be made freely 
available to any industrialist, at a fair and competitive price, in comparison to the cost 
of such raw material in other rubber producing countries. 
The present maximum production capacity of centrifuged latex is aiound 4,000 mt 
annually, while actual production is 2,500 mt. This should gradually be increased to 
10,000 mt since centrifuged latex has an expanding market and is very profitable too. 
The JEDB is at the moment the largest producer of centrifuged latex and recently have 
taken steps to double their production. The SLSPC is also exploring the feasibility of 
setting up a centrifuging plant and it is hoped that their efforts would be successful, in 
the national interest. 
Centrifuged latex should always be produced in amounts greater than immediate 
local needs, to ensure ready availability of this raw material and in order to continue to 
attract potential investors into this area. 
Supply of centrifuged latex 
The raw material for the manufacture of centrifuged latex, as you know, is field 
latex, which can be obtained from any rubber plantation. 
In Sri Lanka, almost all the field latex is processed into RSS, latex crepe, sole 
crepe and block rubber and at present, only a minute quantity is centrifuged. 
The crisis that has now affected the latex crepe industry necessitates the planned 
diversion of latex normally converted to latex crepe, into centrifuged latex manufac­
ture, thereby controlling the availability of latex crepes in the market, in order to stabilize 
prices. 
As I mentioned earlier, producers of rubber such as the JEDB and SLSPC should 
earnestly explore possibilities of setting up industries to convert raw rubber into value 
added finished goods and there should be no difficulty in identifying foreign collaborators. 
712 
Their knowledge, expertise and ready markets could be matched with our raw material 
and inexpensive labour, to achieve a product which would be attractive cost wise. 
Smallholders 
Very nearly two thirds of the rubber extent in the Island is composed of smallholdings. 
The Government has assisted these smallholders in numerous ways but such assistance 
has not had the desired effect in most instances. Although these smallholdings are to 
be found in and around the State owned plantations, it is unfortunate that the small­
holder has not benifitted by the expertise and facilities available on these plantations. 
Smallholder latex is rarely accepted in our factories due to various constraints. Is it 
in the national interest to continue in this manner ? With the re-organisation of factories 
under the MTIP it is suggested that, in each sub-district, a small crepe factory be set 
apart to handle smallholders' latex. 
Perhaps you may be aware that latex crepe rubber production showed a remarkable 
growth in 1983. This was at the expense of sheet rubber but as a result of this increase 
of 26,162 mt of latex crepe production over 1982, the country earned an extra Rs. 56 
million, and this was achieved without any new latex crepe factories coming into opera­
tion, indicating the existence of a considerable unused installed capacity for the produc­
tion of latex crepe in the plantations. 
I would venture to state further, that the smallholder should also be assisted in the 
cultivation of his smallholding, by the managerial staff on estates in the vicinity, and 
the local advisory officers of the Rubber Research Institute (RRI) could co-ordinate and 
assist in planning such measures and inputs. The feasibility of attracting and adopting 
smallholdings around an estate, on a permanent basis, merits consideration. This would 
induce tremendous confidence in the smallholder, to realize that perhaps his fertilizer, 
pesticides and even labour if required would be supplied by the parent estate, while his 
latex will be purchased at a fair price and the cost of inputs deducted from the income 
he is to receive. Such meaningful assistance will also help estate village integration and 
help to bring about better relations and understanding between the estate population 
and the villagers who live in close proximity. 
Scrap rubber harvested by the smallholder is sold for a pittance to the closest rubber 
dealer who in turn sells it to bigger dealers or to an agent of the owner of a scrap rubber 
manufacturing factory. Very little reaches the four block rubber factories, since they 
are working to full capacity. 
The State sector should explore the feasibility of setting up small block rubber 
factories to manufacture SLR 20 with estate scrap and SLR 50 with smallholders' 
scrap, thereby ensuring that the smallholder receives a fair price for his scrap rubber 
and also the country would earn extra foreign exchange by converting all scrap into block 
rubber instead of scrap crepe. 
Another factor which affects the smallholder and all rubber producers is the high 
level of duty that this commodity attracts. Should we not fall in line with other natural 
rubber producing countries, in this sphere, in order to ensure that our rubber is com­
petitive in world markets, and also to ensure a better income for the producers very 
specially the smallholders. 
713 
/ 
GROWING OF RUBBER WITH SPECIAL REFERENCE TO 
REPLANTING CYCLES 
By 
D . E . C. WlJESINGHE 
(SLSPC III, Padukka State Plantation, Padukka) 
This paper presents the views and observations of a Rubber grower, and sets out to 
briefly discuss a mutual appreciation of difficulties and responsibilities of the scientist 
and the grower in making the rubber industry of this country, which is a national assets, 
as well as a means of livelihood for the many people who are directly and indirectly asso­
ciated with the industry. There were gloomy forebodings for the future of the natural 
rubber industry in the 1950's and the competitive pressure exerted by the synthetic rubber 
industry left no alternative to the natural rubber producer but to embark on develop­
ment programmes laying emphasis on replanting and improvement in quality. The 
Government commenced the replanting subsidy scheme in the year 1953 and undertook 
development of factories, with substantial assistance to the producer. Much encourage­
ment was needed for the producer to undertake this massive investment on development 
and all of us who were present at the International Rubber Conference held in 1973 in 
this same venue, would recall the revitalising comments made by Doctor B. C. Sekhar 
who said " there is a future for Natural Rubber as the Aircraft of the future, the Concord 
has to land on Natural Rubber tyres". This remark was greeted with delighted applause 
and from what we see today, there is no doubt in the modern technical world, there is a 
need for N R and its has to exist. 
It is in this context with a firm belief on the future of NR, I express my views related 
to the three main categories of rubber growers in Sri Lanka. 
Table 1 
* Acreage Production 
Small Holdings (1 — 1 0 Acres) 260,000 55,000 
(10 — 5 0 Acres) 80,000 25,000 
(Over 50 Acres) 170,000 55,000 Tons 
Mostly State owned Plantations 
* Acreage in rubber is being revised 
Earnings from Rubber as proportion of total Sri Lanka export earnings is in the 
region of 14% 
715 
Table 2 . Age analysis of rubber lands held by the three categories (acres) 
Under 10 years 10—30 yrs. Over 30 yrs. 
Small holdings (under 10 acres) 
Small holdings (10 — 50 Acres) 
Over 50 acres 
60,000 
10,000 
60,000 
120,000 
30.000 
50,000 
80,000 
40,000 
60,000 
The first two categories of growers own approximately 67% of the total rubber extent 
with the balance ownership, held by the state. 
The need for replanting in general arise when the economicyields arc not forthcoming 
due to ; 
1. Total consumption of exploitable bark. 
2. Poor yields due to clonal characteristics. 
3. Poor yields resulting from substandard after care of holding/plantation, 
A smallholder by virtue of its small extent, is faced with a difficulty in the matter 
of deciding to replant and replanting itself. These holdings, due to land alienation laws 
and the connected need for housing, tend to diminish in extent progressively, with the 
result that the skilled worker employed is driven away. This situation is understandable 
when it is seen that the total earnings as profit, from a small holding of H ha has 
been in the region of Rs. 4,300/- per annum in 1983 while the same smallholder if em­
ployed as a tapper in a neighbouring plantation would earn nearly Rs. 5,200/- per annum, 
with the added security of employment and other retiring benefits. With the increase 
in wages this year, the total earnings would be in the region of Rs. 7,800/- per annum. 
Thus for him, the alternate employment of less skilled tappers, children and elderly people 
in the village, would be meaningful to increase his income, but causes a tremendous loss 
of efficiency in intake and panel care. However this situation is under fair control due 
to the limited employment opportunities available in the larger plantations. Therefore, 
when lands are given out to villagers, either from plantations or other crown land to plant 
rubber, the minimum extent of such allotment has to be determined, based on an evalua­
tion of income per ha. Unless this basic unit is able to generate sufficient income, 
for the sustenance of the owner, extension services too may prove ineffective. But, when 
land available for distribution is limited, the other alternative would be to manage rubber 
lands as large estates, and share the profits. 
Therefore, training of tappers needs priority, specially when it is known that the youn­
ger generation is reluctant to adopt this vocation, as observed at village level. Of cou­
rse, all development programmes, where factory work is provided to young people, will 
always draw to younger workers away from tapping, weeding and other forms of labour, 
which is considered menial. Malaysia, which is rapidly becoming industrialized is 
already feeling the effect of this " labour drain ", 
716 
Low yields from this large extent of land in the hands of the smallholder is causing 
concern, and a substantial increase in yield is necessary to justify the heavy expenditure 
incurred as replanting subsidy. Application of a theoretical replanting cycle to this 
category is not feasible, but a meaningful extension service as provided by the RRI is 
valuable to inculcate a sense of awareness in their minds, on important aspects of after 
care of the rubber trees, grown at immense expense to the state and the producer. Under 
the extension service programme leaders arc selected from each area and short term in­
struction classes are provided on manufacture and cultural practices, and the desired 
awareness is conveyed rapidly to the village. 
Tapping slope and bark consumption which has a direct relation to the replanting 
cycle is often forgotten. Frequency of tapping is not in keeping with the accepted inten­
sities, inclination to avoid late tapping on wet days and poor stand per acre, are the major 
causes for the low yield observed. Extension services should cover all aspect of bud-
grafting, green budding, polybag planting, tapping, application of fertilizers, and weed 
control, as the smallholder is unaware of their importance, to increase yields. The 
low stand per acre is one of the main reasons for low yields in this group, caused by their 
inability to procure and supply vacancies at the correct time. For example a smallholder 
having 12% vacancies, in his holding is deprived of nearly 12% from his only source of 
income, and will continue to do so for the next 20 years. This unfortunate situation can 
be avoided if they arc trained to bring up a few plants in polybags to supply vacancies 
during the first year of planting. 
Owners of the second category (10 — 50 acres) include those who have received their 
land entitlement from the Land Reform Commission after their statutory determination, 
form extents of 50 acres demarcated from larger plantations. These are better managed 
and placed in charge of experienced supervisory grade officers whose emoluments arc 
higher than the earnings of a daily paid worker. Skilled tappers receiving stipulated 
wages are being employed, and have their independent RSS producing factories. Some 
of these owners supply their latex to neighbouring crepe factories. Regular application 
of fertilizers is undertaken in most units, but a planned replanting programme is not 
effectively enforced. Owners in general are not residents on the estates. This is a problem. 
Land, and income from it is virtually extra pocket money to some of them. There are 
the week-end landlords to whom land ownership provides a relaxing weekend hobby. 
They should however be mindful of their responsibility to the industry. 
The third category of lands (over 50 acres) mostly State owned, well managed, with 
an average yield of over 1,050 kg per ha., (in 1983) and keeping ahead of the national 
average by a margin of over 200 kg per ha., adhere to a replanting cycle of 30 years 
Due to the low rate of replanting undertaken prior to nationalisation, in most of the estates, 
the desired programme has not been implemented. As a result, after nationalisation 
replanting has been accelerated and in some plantations, the extent in the immature phase 
is as much as 50 to 60% of the total extent of the plantation. This accelerated replanting 
programme no doubt will interfere in the implementation of a future replanting cycle 
of the unit. Perhaps, if replanting is undertaken on a region or district basis, the desired 
target could be achieved, but the concept of retaining the viability of individual units, 
may fail for some length of time. 
717 
The commonly used clone for replanting after 1953 has been the clone PB 86. It 
has been observed that most of the plantations have exceeded the theoretical 7 inch bark 
allocation for the year, and reached the peak harvest in the 18th/19th year after planting. 
In this situation, after accommodating further 6 years for intensification, uprooting is 
due on the 25th/26th year leaving a further tapping period of 4/5 years at uneconomical 
intake levels, when the present 30 year replanting cycle is practiced. 
To overcome this position, stimulation is suggested, but our past experience is that 
the majority of us have not succeeded in obtaining a satisfactory result, due to the weather 
pattern we experience, and I suppose it is difficult to get through more than two stimula­
tions per year. To meet the challenge of high costs, a shorter replanting cycle seems to 
be the other alternative, and perhaps this may not be in accordance with the economics 
of planting. But re-thinking in this direction, seems necessary, when we consider the 
fact that an income of approximately Rs. 25,000/- per ha is possible from the sale of 
rubber trees, with the buyer, uprooting and clearing the land free of charge. This is 
equivalent to an income of Rs. 5,000/- per ha., over a period of 5 years, which again is 
much more than the annual income from rubber yielding 1,200 kg/ha. With 5% of 
the total work force of the country employed in the rubber industry the achievement of a 
economic yield within the shortest possible time is a necessity. 
Research on plant breeding is a continuous process where we all know considerable 
time is necessary to identify and establish a suitable clone with the desired features, 
and the benefits of this long process cannot be achieved, if rapid replanting is not under­
taken. Our scientists have produced suitable clones to meet the situation described above. 
Selected clones from the RRIC 100 series, with an immature phase of approximately 5 
years, giving a satisfactory initial yield and a peak yield of over 2,500 kg, per ha., have 
been recommended for large scale planting. It is my belief, that a plantation on a 3 % 
rate of planting could afford an increase of 1 %, still retaining 80% of the plantation in 
the mature phase, when the recommended newer clones are used. This increase in the 
rate would place us in a replanting cycle of 25/26 years, and would greatly assist to clear 
the backlog of uneconomic rubber at least in the State owned large plantations. After 
which replanting could be undertaken on a lower note looking forward, hopefully, for 
an established economic yield for our plantations, maintaining 85% of the extent in pro­
duction. 
Sometimes I may have touched on areas of controversy without sufficient data and 
statistics. This is intended as I consider it necessary to stimulate deep investigations, 
and moreover, I cannot think of a better forum than this representative group of scientists, 
rubber growers, government officials and others connected with the industry, to present 
m y observation. 
CONCLUSION 
Management skills of the smaller units require further improvement. Formation 
of Group Management Organisations with sufficient holdings grouped to form units of 
10 to 15 ha., managed by a trained leader from the village, utilising trained labour re­
ceiving wages applicable to the trade, as a further extension to the existing group pro­
cessing centers, is advocated and placed for consideration. 
718 
A study, preceded by a socio-economic survey of the smallholders of the group up 
to 10 acres, to identify the causes relating to the low tappings undertaken, low yields, 
insufficient fertilizer use, and the low net sale average, to be undertaken early. 
Accelerated rate of replanting or a 25/26 year replanting cycle for the larger planta­
tions is also a subject for further consideration. 
Our eminent scientists who are with us today, have always been in the forefront of the 
industry, evolve the strategy, for implementation by the producer. We thank them and 
look forward to meeting the emerging challenges to the natural rubber industry, with 
equal strength. 
719 
CLOSING SESSION 
SEVENTY FIVE YEARS OF RUBBER RESEARCH IK 
SRI LANKA 
By 
O . S. PERIES 
(Director, Rubber Research Institute of Sri Lanka) 
Rubber research was started in Sri Lanka in 1909, when some rubber planters in the 
Kalutara District pooled their resources to engage the services of a British Chemist to 
study the process of latex coagulation. This was the start of systematic research for a 
vast industry that now supports millions of people in South East Asia, and is their only 
source of income. The Rubber Research Institute of Sri Lanka organized an International 
Conference to celebrate this occasion on 1 7 - 1 9 September 1984, this is a summary of 
the highlights of this Conference. 
Tan Sri Dr B. C. Sekhar, Chairman of the Malaysian Rubber Producers Research 
Association, one of the world's foremost authorities on natural rubber (NR), advised 
all rubber producing countries to gear themselves to increase rubber production as the 
demand for the product is going to increase in the next few years. N R producing coun­
tries should not miss this golden opportunity to support their economies, because the 
industrial nations will go in for substitutes, if we default on supplies. He emphasised 
that synthetic rubber (SR) is only a substitute, a poor one at that, and that it will never 
be able to replace N R fully. The natural product is a unique material, with many special 
properties. Every pound of N R that has ever been produced has always been sold, which 
cannot be said for SR. This shows that it is only when we, the N R producers, fail in our 
efforts to supply the quantity required that SR steps in to fill the gap. 
The present unattractive price for N R is only a temporary phase, caused by recessions 
in various countries, which has led to a poor demand for motor vehicles, that use up 65 
to 70 percent of all rubber, both N R and SR, produced in the world. 
When the world economy improves, the sales of cars will increase, and then N R will 
come into its own. The forecast for rubber demand is in the scale of 18 million tons in 
the 1990s, and even on the present scale of consumption, this means that at least 6 million 
tons of N R will be required to satisfy this demand and that calls for a doubling in the 
present supply position. However, the forecasts are even better, because it appears that 
from the present situation, of N R supplying 30 percent of the world demand, N R can 
claim 40 percent of the market, on its techno-economic performance. This will tilt the 
demand towards N R still further. As far as the price is concerned, on the basic properties 
of NR and its utility value, a fair price for N R at present would be 1 US dollar per pound. 
Therefore, the current price for N R is almost exactly half of what it is worth. The reason 
for this Dr Sekhar said was mainly the poor marketing system for the product. He 
urged that a new marketing arrangement be developed, with all N R producing countries 
contributing to the system so that a realistic and fair price be paid for it. The present 
marketing system is an archaic 15th century arrangement, not in keeping with the present 
development of the world. 
m 
i t is difficult to summarize the proceedings of a Conference, where 86 papers wefe 
read, in 3 days, on such varied subjects as plant breeding, tapping, fertilizer usage, 
disease control, climate, smallholdings, soil management, economics, rubber chemistry 
and technology, energy use and labour costs, in a few minutes. Therefore, let us use 
the endeavour of the 1984 Olympics : "Higher, Faster, Further" as a yardstick to 
measure the achievements of research to support the industry - how far have we been 
successful in developing clones to give higher yields and faster growth, how successful 
have we been in developing further uses for NR ? This is what I will try to do in this 
summary. 
The N R industry has been singularly successful in developing high yielding fast 
growing rubber varieties. Sri Lanka has been particularly successful here and four of 
its recently developed clones have come out at the top of international trials carried out 
in various countries. There are new clones that have produced 4,000 to 5,000 lb of 
rubber per acre under experimental conditions — a far cry from the 400 to 500 lb of 
rubber per acre, produced by the old seedling varieties. Now, these new clones must 
be adequately looked after, fertilized, protected from pests and diseases and be tapped 
according to scientific standards-clearly laid down. The NR research studies all over 
the world have made numerous advances in all these areas. The current trend is towards 
labour saving devices for tapping and this is engaging the efforts of all countries, due to 
the rapidly escalating cost of production. All fertilizers, in general, are imported into 
all N R producing countries. Therefore, there has been a concerted effort to reduce 
nutrient costs — the use of the soil and foliar analysis technique for fertilizer usage on 
rubber has reduced costs of Hevea nutrition by at least 30 percent in recent years, a very 
significant contribution to the economy of the industry. Disease control is taking a very 
definite path towards biological control of all leaf, stem and root diseases and a number 
of scientists showed their success with this technique, where only the minimum quantity 
of fungicides and insecticides are used to preserve the health of the tree. It was also 
noted with great satisfaction that Sri Lankan scientists at the RRISL were the acknow­
ledged leaders in the field of Hevea pathology. 
The papers on marketing and economics were extremely interesting, as they brought 
out the interesting aspects of the manner in which the world rubber market operates, 
its strong and weak points. It was agreed that Sri Lanka should allow forward sales of 
rubber up to at least 6 months ahead. The restriction of this facility at present, acted to 
the detriment of our country. As far as economies of rubber growing are concerned, a 
very interesting paper from the School of Economics in the Australian National University, 
clearly showed that, when countries took steps to bring in laws to control the amount o 
earnings that a grower received from his crop, the result usually was almost always opposite 
to what the country wished to achieve. For example devaluation is a case in point. 
Theoretically, devaluation should increase the local value of commodities like rubber. 
However, the real result of devaluation is generally to increase Hie price of all goods in 
the country, with the result that the smallholder is ultimately worse off than when 
the process was started. Therefore, all economic measures taken, with the very good 
intention of helping the local population, must be studied in detail to ensure that the 
legislation enforced will achieve the results expected. Economic theory alone is not 
sufficient. 
722 
The N R industry has made several advances in the field of enlarging the scope of the 
use of rubber. Some 32 papers were read on the diverse subjects of manufacture of 
commercial goods, raw rubber manufacture, and properties, chemical modification o f 
N R , N R technology and analysis, compounding and ageing properties, latex concentrates 
alternate sources of energy and environmental pollution and its control. AH these 
studies have helped to improve the properties of NR, with a view to increasing its price. 
However, this is the one area of this industrial raw material, that gives a gloomy picture 
at present. As stated at the beginning of this summary, the value of N R should be about 
1 US dollar per pound ; but it is now selling at exactly half that price. That is the tragedy 
of the world, that the industrial nations can always dictate the terms of trade to the under­
developed countries. If we take any plantation crop, the picture is the same. The 
basic problem in agricultural production in the developing world is that the price paid 
for the product is always so low that, all that the population of developing countries can 
expect is to eke out a subsistence living. Taking comparative numbers, if a person in 
the developing world could buy a tractor for 500 bags of coffee in 1950, then he cannot 
hope to buy a tractor today for less than 50,000 bags of coffee. The cost of production 
of the tractor has gone up, admittedly ; but so has the cost of production of coffee. How­
ever, there is no consideration of this in the international market. Instead, there appears 
to be a tacit acceptance amongst the people of the developed countries that the labour in 
developing countries should live on a subsistence diet, just to enable them, the haves, 
to maintain a high standard of living. This is the reason why certain labour intensive 
industries keep moving from one country to another. As soon as labour costs go up, 
then industries move to lesser and lesser developed countries. Is this fair ? All I want 
to emphasise here is that the prices of commodities will not reach equitable levels, until 
and unless all nations in the world accept that we are all a part of the world community 
of people. None of us owe the other a living ; but we all certainly owe the other a fair 
price for his labour. If this is accepted and acted upon, then we will have a fair price 
for rubber, as well as coconuts, tea, cocoa, oil palm and all the other primary products 
in the world. The Finance Minister of Sri Lanka, Mr Ronnie de Mel always points out 
that there is nothing called a " free meal ", so that all of us must be ready to work for our 
rice and curry. But equally, it must be accepted that there is nothing called " free labour ". 
All labour must be paid a fair price. We are not asking for a 3-car family ; all we are 
asking for is a 3-meaI day. I wish to leave you with that message, and all you nice people 
from abroad, I should be grateful if you carry that message back with you to your coun­
tries and your people. Thank you. 
723