Livestock Research for Rural Development 13 (2) 2001

Citation of this paper


Evaluation of nutrients of rubber seed meal in Mong Cai pigs


J Ly, Chhay Ty and Chiev Phiny


University of Tropical Agriculture Foundation

Royal University of Agriculture

Chamcar Daung, PO Box 2423, Phnom Penh 3, Cambodia and





Three experiments were conducted to evaluate the nutritive value of Cambodian rubber (Hevea brasiliensis) seed meal either in vivo with young Mong Cai pigs or in vitro by the pepsin/pancreatin incubation procedure. The average weight of the seed was 3.29 ± 0.78 g and the proportion kernel/husk was 1.17 ± 0.40 g. The average composition of rubber seed meal was ash 3.10, NDF 53.8, crude fat 28.4 and  N 2.39% in dry basis. The cyanide content of the seeds decreased from 82.5 to 29.3 mg/kg DM in approximately 45 days of storage.


In vivo digestibility of nutrients of the feed (sugar palm syrup and dried fish) was not impaired by the introduction of either 17.8 or 40.5% of rubber seed meal in the diet. The balance of N was significantly better with 40.5% of the seeds in the diet than in rations with no rubber seed meal. Digestibility indices of rubber seed meal as estimated by difference were DM 79.4, NDF 75.0, N 83.6 and organic matter 85.5 %.


In vitro digestibility studies indicated significantly (P<0.001) higher values for N utilization in the rubber seed kernel (89.7%) than in fish meal or dried fish (81.3 and 77.2%, respectively). In vitro DM and organic matter digestibility coefficients in the husk were comparatively low, probably due to the fibrous nature of this material.


Key words: Rubber seed meal, Hevea brasiliensis, sugar palm syrup, Borassus flabellifer, digestibility, in vivo, in vitro, nitrogen, organic matter, cyanide,  Mong Cai pigs



It is well known that protein sources are the main constraint for the improvement of pig production in many tropical regions of the World. In this connection, the use to some extent of non-conventional, locally available feeds is indicated. Rubber seeds have a great potential of utilization for feeding pigs either by individual farmers or if included in compound feeds.


In Cambodia, the plantations of rubber trees are not so great as compared to other countries of South East Asia (see for example FAO 1999). However, in the northeastern Cambodian provinces there are some 39,700 ha of rubber trees (MAFF 2001), from which it could be obtained a substantial amount of rubber seed meal, since the annual yield of seeds per ha has been estimated to be 800 to 1200 kg (Siriwardene and Nugara 1972).


Several studies have been devoted to study the potential value as feed of different types of rubber seed meal in monogastric animal species (Fetuga et al 1977; Ong and Yeong 1977; Rajaguru and Ravindran 1979; Nwokolo 1987). From the point of view of the nutritive value of the seeds for pigs, very little information is available (Siriwanathanukul 1982; Babatunde et al1990; Nutdechanun 1991; Agumbiade et al 1996).

In this communication a first approach to estimating the nutritive value of Cambodian rubber seed meal is made by examining several indices either in vivo with young pigs or in vitro, simulating the mouth to ileum digestive processes.


Materials and methods


Three experiments were conducted to study the nutritive value of Cambodian rubber seeds in pigs fed sugar palm syrup as the energy source. In experiments 1 and 2, in vivo total digestibility of nutrients and N balance was undertaken with graded levels or rubber seed meal in the diet. The third experiment consisted of an in vitro digestibility study.

Characteristics of the rubber seeds

The rubber seeds were from Cambodian plantations in the province of Kampong Cham. A representative sample from the batch used in the experiments was examined for several seed characteristics. A total of 20 seeds was selected at random and weighed entire or decorticated for the determination of the kernel/husk ratio. The seeds were on average 2.37 ± 0.27 cm long with a diameter of  1.98 ± cm. The weight of the entire seed was 3.29 ± 0.78 g whereas the kernel/husk ratio in natura was 1.17 ± 0.40.


The chemical composition of the husk and kernel of the rubber seeds is shown in Table 1. 


Table 1. Chemical composition of the kernel and husk of rubber seeds (dry basis)




Analysis, %1



Dry matter












Organic matter



1 Analysis conducted in quadruplicate

The entire batch of seeds was ground in a hammer mill and a representative sample was used for cyanide determination and nutrient composition (vide infra).


In vivo digestibility trials (Experiments 1 and 2)

Two experiments were carried out according to a change over design to study the influence of either low or high (Experiments 1 and 2 respectively) levels of rubber seeds on the in vivo, total digestibility of nutrients and the N balance of young pigs fed sugar palm (Borassus flabellifer L.) syrup as the only source of energy. 


A basal, simple diet, composed of sugar palm syrup, fish meal, vitamins and minerals, was formulated to contain approximately 18% crude protein (CP, Nx6.25). The diet was partially substituted by graded levels of full-fat, untreated rubber seed meal (Table 2). The sugar palm (Borassus flabellifer L.) syrup was prepared daily from palm sugar. The DM content of 5 composite, weekly samples was 82.7 ± 4.85%. The palm sugar was purchased in the local market. The sugar palm syrup was transparent and light brown in appearance and very sweet. A study of some of the characteristics of the sugar palm syrup revealed that the total carbohydrate content, as determined by refractometry, was 94.06 ± 0.34% in dry basis. Organic matter (100 - % ash) and N values in dry basis estimated in duplicate from a composite sample were 98.7 and 0.27% respectively, whereas the DM content of the prepared sugar palm syrup as fed to the animals (1:1 diluted with water, by weight in fresh basis), was 41.4 ± 2.43%.


Table 2. Characteristics of the diets (in dry basis)


Rubber seed meal, %





Ingredients, %




Sugar palm syrup




Fish meal




Rubber seed meal








Vitamins and minerals4




Analysis, %




Dry matter
















Organic matter




1 Average values from Experiments 1 and 2;  2 Experiment 1;  3 Experiment 2

 4 According to NRC (1998) requirements for vitamins and minerals


A total of eight Mong Cai, castrated young male pigs were used according to a balanced change over design (Gill and Magee 1976; Gill 1978). The average initial weight of the animals was 14.0 ± 0.9 and 21.0 ± 1.5 kg in experiments 1 and 2 respectively. The animals were allocated to metabolism cages constructed with rattan, allowing the quantitative separation of faeces and urine. The description of the metabolism cages has been reported previously by Chiev Phiny and Rodríguez (2001).


The pigs were alternatively fed with one diet during five days, followed by another five days of collection of faeces, urine and feed refusals. Then, a second 10 day period followed, where the other diet was offered to the animals. There were two diets in every in vivo experiment, the control diet or this diet partially replaced by either 18.7 or 40.5% of rubber seed meal.


 Method of calculation of in vivo digestibility of rubber seed meal


The digestibility of the different nutrients of the rubber seed meal was estimated by difference, according to the method outlined by Crampton and Harris (1969).

In vitro digestibility trials (Experiment 3)

The method of two stage in vitro digestibility determination with pepsin and pancreatin was that utilized by Dierick et al (1989). Samples of rubber seed kernel and husk were incubated by quadruplicate in pig pepsin dissolved in HCl for four hours at 39oC, followed by another four hours with pig pancreatin dissolved in a phosphate buffer (pH = 7.5). In vitro digestibility of DM, organic matter and N of the samples was compared with that of standard casein as a control sample. The in vitro digestibility values for rubber seed fractions were compared with those of a representative sample of the fish meal employed in the in vivo trials. In addition, a sample of dried fish (N 8.36; ash 20.5,  dry basis) was used in this study. Dry fish, average weight 1.2 g (DM 92.9%) is very common in local markets of Cambodia, and its origin is from family fisheries in inland waters (MAFF 2001). 

Chemical analyses

The determination of DM in feeds and faeces was conducted by drying to constant weight by microwave radiation (Undersander et al 1993). The analyses of ash, crude fat and N in feeds and faeces were carried out according to AOAC (1990) recommendations. Urinary N was determined by the macroKjeldahl procedure (AOAC 1990). The NDF in all samples studied was estimated following Van Soest et al (1991). Faeces and urine analyses were conducted in a composite, representative aliquot of every five day period collection for each pig.


Faecal pH was determined with a glass electrode. A faecal extract was prepared after straining a slurry made from fresh faeces and distilled water (1:4 partsby weight), through two layers of cotton cloth. In the faecal extract, the concentration of total short chain fatty acids (SCFA) and ammonia was determined by separation from the sample in a distillation unit, after treating the faecal extracts with either saturated MgSO4 in 4N H2SO4 or 40% NaOH, and titatring the resulting aqueous extract by the acid-base technique.


The cyanide content in the rubber seed meal was estimated according to AOAC (1990). 

Statistical analyses

Data resulting from the in vivo and in vitro experiments were subjected to analysis of variance according to Steel and Torrie (1980). The Minitab statistical package (Ryan et al 1985) was used to test for significance of treatment differences in the experimental measures. Duncan’s multiple range test was used for identifying differences among means in the required cases, when the analysis of variance was significant (P<0.05). 



In vivo experiments

There were no symptoms of cyanide intoxication in the pigs fed different levels of rubber seed meal. At the end of the experiments, approximately 45 days after the preparation of the rubber seed meal, the cyanide concentration in samples of the remaining material was relatively low (29.3 mg/kg DM), as compared to the initial value (82.5 mg/kg DM. The animals were always in positive N balance.


The experiments were conducted during February and March 2001 in a hot environment. Air temperature in the metabolism cage area, measured daily at 12:00 during 45 days, was 31.8 ± 1.7oC. No rains were recorded during this period in the Farm.


The analysis of variance showed no significant effect (P<0.05) of the period.


Table 3 lists some faecal characteristics related to the large intestine fermentation activity. 


Table 3. Faecal fermentative indices in Mong Cai pigs fed rubber seed meal


Rubber seed meal, %






Faecal pH





Experiment 1





Experiment 2





SCFA, mmol/100 g DM



Experiment 1





Experiment 2





NH3, mmol/100 g DM



Experiment 1





Experiment 2





** P<0.01; *** P<0.001


The fermentative indices determined in the faeces of the Mong Cai pigs appeared to be similar in both experiments. On the other hand, an inclination of the data to be influenced by the inclusion of rubber seed meal in the diet was apparent. This was evidently so in Experiment 2, where 40.2% of the seeds in the diet gave rise to lower pH values and higher SCFA concentration in faeces, as contrasted with the control diet (P<0.001). High values of faecal ammonia were also noted with the higher levels of dietary rubber seed , as compared with the control diet (P<0.01).


There was no significant differences between diets in in vivo DM and OM digestibility when rubber seed meal accounted for 18.7% of the diet as compared to the control treatment (Table 4). However, a lower in vivo OM digestibility was found (P<0.05) when the treatment consisting of 40.5% rubber seed meal was compared to the control treatment. NDF digestibility was high when rubber seed meal was introduced in the diet. 


Table 4. In vivo nutrient digestibility in Mong Cai pigs fed rubber seed meal


Rubber seed meal, %






DM digestibility, %





Experiment 1





Experiment 2





OM digestibility, %





Experiment 1





Experiment 2





NDF digestibility, %





Experiment 1





Experiment 2





1 There was no NDF in the diet and in the faeces of pigs fed the control diet

* P<0.05

Table 5. N balance in Mong Cai pigs fed rubber seed meal


Rubber seed meal, %


SE ±




N digestibility, %





Experiment 1





Experiment 2





N retention, % intake





Experiment 1





Experiment 2





N retention, % digestion





Experiment 1





Experiment 2





.** P<0.01


There were no significant differences between treatments in the digestibility of N. Nevertheless, a trend was observed for the balance of N to be improved with increasing levels of rubber seed meal in the diet. In fact, the retention of N as a proportion of that consumed or digested, was higher (P<0.01) in the diet with 40.5% rubber seed meal as compared to the control diet.


The in vivo digestibility values of several nutrients, as calculated by difference are presented in Table 6. There were no differences in these indices between diets with 18.7 or 40.5% rubber seed meal. Therefore, the presented values correspond to eight observations. 

Table 6. Nutritive value of rubber seed meal


Composition, %1

Digestibility, %2

Dry matter


79.4 ± 12.1






75.0 ± 8.64

Crude fat





83.6 ± 12.2

Organic matter


85.5 ± 7.80

1 In dry basis; 2 Mean and SD of eight animals; # Not determined


In vitro experiments

A highly significant (P<0.001) decrease in in vitro DM, organic matter and N digestibility for rubber seed husk was observed (Table7). On the other hand, in vitro organic matter digestibility of rubber seed kernel was not different from that of fish meal. This index was the highest (P<0.001) for dry fish. The in vitro N digestibility was significantly higher (P<0.001) in rubber seed kernel, as compared to either that of fish meal or dry fish.

Table 7. In vitro (pepsin/pancreatin) digestibility indices in rubber seed fractions1


In vitro digestibility, %





Fish meal




Dried fish




Rubber seed kernel




Rubber seed husk








1  In vitro digestibility of pure casein was 100% ; *** P<0.001;
abcd  Means in the same row with different superscripts differ at P<0.05




Digestion of the organic matter fraction 

In the in vivo studies, sucrose constituted one of the main components of the organic matter fraction of the diet, due to the fact that the animals were fed with substantial amounts of sugar palm syrup. In fact, some 40 to 50 % of the diet was sucrose, taking into account that this carbohydrate is the main component of sugar palm syrup (Romera 1968). Therefore, a high organic matter digestibility value for the diet could suggest that the sucrose contribution to the above mentioned results could be of significance, since this disaccharide is practically digested in the precaecal areas of the pig gastrointestinal tract (Ly 1992). On the other hand, the high digestibility values found for the NDF fraction could be a consequence of the ability of the Mong Cai pigs, even young animals of this genotype, to utilize efficiently fibrous materials. Substantial amounts of SCFA and low faecal pH in diets with high levels of rubber seed meal could be an indication of a very active microflora in the large intestine thus degrading the dietary cell wall constituents. Moreover, low in vitro pepsin/pancreatin digestibility of organic matter in the rubber seed husks could indicate that most of the fibrous fraction of the husk must disappear in the caecum and colon of the pig, such as it does occur with the majority of this type of feeds (Close 1993).


The high values found for in vitro digestibility of organic matter in the rubber seed kernels may suggest an extensive disappearance of fatty acids in the small intestine of   pigs. As it is known (Nwokolo 1987), undefatted rubber seed kernels could contain about 48% of ether extract. No attempt was made in the experiments described herein to determine the extent of utilization of the fatty acids from rubber seeds by the pigs, but the data of Agumbiade et al (1996) concerning the utilization of energy of African rubber seed meal did not suggest an impaired digestion of rubber seed fatty acids in this specie. Moreover, Siriwathananukul (1982) found in young pigs that graded levels of rubber seed meal of up to 30% in the diet, decreased DM digestibility, but no treatment effect was found on digestible or metabolizable energy values.


It has been observed that Chinese breeds of pigs are very efficient in crude fat digestion, mainly due to an increase in pancreatic lipase activity (Février et al 1988). There is no reason to believe that the Mong Cai genotype could be an exception to this assumption. On the other hand, use of full-fat rubber seed meal could compensate for the low energy value of such a feed prepared with the husk, even if an efficient microbial digestion of the cell wall contained in the husk does occur in the large intestine of pigs. 

Digestion and metabolism of the N fraction

The use of high levels of rubber seed meal in the diet such as those used in the experiments reported in this communication have indicated a trend to improve the N balance of the animals, whereas no treatment effect was evident in in vivo N digestibility. Babatunde et al (1990) observed a curvilinear response in pigs fed graded levels of rubber seed meal of up to 25% of the diet, with a minimum at some 11.3% of rubber seed introduction in the feed. On the other hand, Siriwathananukul (1982) did not find differences in the biological value of diets containing from 0 to 30% of rubber seed meal given to young pigs, which supports our findings.  A high value of in vivo N digestibility such as that found in the present experiments is in accordance with the high value of in vitro N digestibility of the kernel and even the husk of the rubber seeds, thus supporting the assumption that a great proportion of digested protein from the seeds could be digested and the released amino acids absorbed from the small intestine. This in turn could explain the increase in N retained by the animals, both as a percentage of that digested, and even more, as percentage of the N consumed. 


The results of the in vivo and in vitro trials suggest a high digestibility and biological value of the untreated, full-fat rubber seed meal protein fraction. This is in contrast with other observations made by Nutdechanum (1991) who found in pigs that in vivo N digestibility was 69.0 and 71.9% for untreated or alkali-treated rubber seed kernel, respectively.


It is possible that the use of an animal protein source such as fish meal could contribute to a better amino acid balance of the diet, and therefore to a more efficient utilization of the dietary N. In fact it is well known that rubber seeds do contain low levels of lysine, methionine and histidine (Siriwardene and Nugara 1972; Narahari and Kothandaraman 1984; Nwokolo 1987). It is possible that a vegetable protein source such as the soya bean meal used in the experiments of Babatunde et al (1990) and Nutdechanum (1991) would not contribute to a satisfactory amino acid combination when rubber seed meal was employed as a protein source in diets for pigs. 



According to the experimental evidence, a full-fat, untreated rubber seed meal low in cyanide can be a good source of protein for growing pigs, if it is taken into account that a deficiency in the amino acid balance can be corrected by supplementation. In Cambodian conditions, where fresh water fish is widely available (Tilchit 1981), this source of protein, together with fish meal, could be a suitable feed supplement for improving pig production with local feeds. The next step must be study the nutritive value of the Cambodian rubber seed meal by conducting performance trials with growing pigs and breeding sows.  

It has been claimed that in certain regions of South East Asia, pigs are left to scavenge for rubber tree seeds when they are shed, and that pigs learn to harvest only low cyanogenic, mature seeds (Holness 1995). This natural integrated farming system merits more investigation. 



 The authors are indebted to the Chub Rubber Company, Kampong Cham, for the supply of the rubber seeds.




Agumbiade J A, Wiseman J and Cole D J A 1996. Improving the nutritive value of Nigerian rubber kernel (Hevea brasiliensis) products through processing. II. Apparent nutrient and metabolizable energy values. Tropical Agriculture (Trinidad) 73:124-132


AOAC 1990. Official Methods of Analysis. Association of Official Analytical Chemists. 15th edition (K Helrick editor). Arlington pp 1230


Babatunde G M, Pond W G and Peo Jr E R 1990. Nutritive value of rubber seed (Hevea brasiliensis) meal: utilization by growing pigs of semipurified diets in which rubber seed meal partially replaced soybean meal. Journal of Animal Science 68:392-397


Chiev Phiny and Rodríguez L 2001. Digestibility and N retention in Mong Cai pigs fed sugar palm (Borassus flabellifer) juice and ensiled fresh water fish. Livestok Research for Rural Development 13(2):electronic version


Close W H 1993. Fibrous diets for pigs. In: Animal Production in Developing Countries (M Gill, E Owen, G E Pollot and T L J Lawrence, editors) British Society of Animal Production Occasional Publication No 16 p 107-116


Crampton E W and Harris L E 1969. Applied Animal Nutrition. The Use of Feedstuffs in the Formulation of Livestock Rations. Ed W H Freeman. San Francisco pp 753


Dierick N, Vervaeke I, Decuypere J and Henderickx 1985. Protein digestion in pig measured in vivo and in vitro. In: Proceedings of the 3rd International Seminar on Digestive Physiology in the Pig (A Just, H Jorgensen and J A Fernández editors). National Institute of Animal Science. Copenhagen p 329-332


FAO 1999. Selected Indicators of Food and Agriculture Development in Asia-Pacific Region 1988-1998. FAO Regional Office of Asia and the Pacific Publication 1999/34. Bangkok pp 206


Février C, Bourdon D, Aumaitre A, Peiniau J, Lebreton Y and Blanchard A 1988. Digestive capacity of the Chinese pig: effects of dietary fibre on digestibility and intestinal and pancreatic enzymes. In: Proceedings of the 4th Seminar on Digestive Physiology in the Pig (S Burazewski, L Burazewska, T Zebrowska and B Pastuzewska editors). Institute of Animal Physiology and Nutrition. Jablonna p172-175


Fetuga B C, Ayani T O, Olaniyan A, Balogun M A, Babatunde G M and Oyenuga V A 1977. Biological evaluation of para-rubber seed (Hevea brasiliensis). Nutrition Report International 15:497-510


Gill G L 1978. Change over design: sequence of treatments. Estimation of residual effects of treatments. In: Design and Analysis of Experiments in the Animal and Biological Sciences. Iowa State University Press. Ames 1:179


Gill G L and Magee W T 1976.  Balanced two-period change over design for several treatments. Journal of Animal Science 42:775-780


Holness D H 1995. Pigs. The Tropical Agriculturalist Series (R Coste editor). MacMillan Education Limited. London pp 150


Ly J 1992. Studies of the digestibility of pigs fed dietary sucrose, fructose or glucose. Archives of Animal Nutrition (Berlin) 42:1-9


MAFF 2001. Agricultural Statistics 1999-2000. Department of Planning, Statistics and International Cooperation. Ministry of Agriculture, Forestry and Fisheries. Phnom Penh pp95


Nutdechanun J 1991.  Effect of supplementary synthetic amino acids in dehulled and alkali treated para rubber seed meal for early-weaning pig diets. Thesis. University of Ketsart. Bangkok pp 51 (in Thai)


Narahari D and Kothandaraman P 1984.  Chemical composition and nutritional values of para-rubber seed and its products for chickens. Animal Feed Science and Technology 10:257-267


NRC 1998.  Nutrient Requirements of Swine. National Academy of Science.9th edition. National Academic Press. Wasshinton D C


Nwokolo E 1987. Biochemical and nutritional qualities of rubber-seed meal. Tropical Agriculture (Trinidad) 64:170-171


Ong H K and Yeong S W 1977.  Prospects of the use of rubber seed meal for feeding swine and poultry. In: Proceedings of the Symposium of Feeedingstuffs for Livestock in South East Asia (C Devendra and R I Hutagalung, editors). Malayasian Society of Animal Production. Kuala Lumpur p 337-344


Ragajuru A S B and Ravindran V 1979. Rubber seed meal as a protein supplement in growing swine rations. Journal of the National Science Council of Sri Lanka 7:101-104


Romera J P 1968  Le Borassus et le sucre de palme au Cambodge. L’Agronomie Tropicale (Paris) 23:801-843


Ryan B F, Joiner B L and Ryan Jr T A 1985. Minitab (2nd edition) Hilliday Lithograph


Siriwardene J A  and Nugara D 1972. Metabolizable energy of rubber seed meal in poultry diets. Ceylon Veterinary Journal 20:61-63


Siriwathananukul Y 1982. Effect of para-rubber seed meal on the performance of Large White growing pigs (15-20 kg). Thesis. University of Kaetsart. Bangkok pp 104 (in Thai)


Steel R G D and Torrie J A 1980. Principles and Procedures of Statistics: a Biometrical Approach. McGraw-Hill Book Company (second edition) Toronto pp 663


Tilchit L 1981. L’Agriculture au Cambodge. Agence de Coopération Culturelle et Technique. Paris pp 423


Undersander D, Mertens D R and Theix N 1993. Forage analysis procedures. National Forage Testing Association. Omaha pp 154


Van Soest P J, Robertson J B and Lewis B A 1991. Methods for dietary fiber, neutral detergent fiber and non starch polysaccharides in relation to animal nutrition. Journal of Dairy Science 74:3583-3593




Received 4 April 2001


Go to top