Recognizing the potential of the sugar palm tree, (Borassus flabellifer L., Arecaceae), to produce juice (sap) rich in highly digestible carbohydrates (sugars), and the success in many tropical countries of pig feeding systems using the juice from the sugar cane, a series of trials was undertaken to evaluate the feeding of pigs with the raw juice from the sugar palm tree. Demonstrations in some 14 farm households in villages that are traditional producers of palm sugar showed that pigs fed with fresh juice from the sugar palm tree, complemented with restricted amounts of boiled whole soya beans (equivalent to 150 g/day protein), grew at an average rate of 400 g/day. An economic analysis showed that pig feeding was much more profitable than making sugar. It is also much more environmentally friendly, especially when the pig excrement is processed in low-cost biodigesters to produce gas for household cooking. Juice production ranged from 2.76 to 6.37 kg/tree/day with an average sugar content of 15%. Assuming each tree occupies an area of 50 m² this is equivalent to a range of 12 to 28 tonnes sugar per hectare per year, which for rain-fed conditions indicates that the sugar palm tree is both a highly productive and efficient user of solar energy.

The palmyra, or toddy palm (Borassus flabellifer L., Arecaceae), grows wild from the Persian Gulf to the Cambodian-Vietnamese border; it is commonly cultivated in India, Southeast Asia, Malaysia and occasionally in other warm regions including Hawaii and southern Florida. The inflorescence has been used for sweet sap production and sugar syrup. Also many products of the tree including leaves, trunk, fruit and underground seedling are used as is a thin orange pulp coating the fibre of the mature fruit which is consumed fresh or dried as a paste.

The tree grows well in poor acidic sandy soils and produces juice during the dry season in regions where plant growth is negligible in the absence of irrigation. The palm tree receives no agronomic inputs other than what is applied to companion crops.

The following study was focused on the use of sugar palm juice as the sole energy source in the diet for growing-fattening pigs.

Each family had between 12 and 20 trees. Seven of the families also produced sugar and their data were used to derive an economic comparison between use of the juice for sugar or pig production. Each month, from January to April, juice yields were measured on three consecutive days for all the trees being tapped by each family. The samples comprised both male and female trees.

The juice was collected twice daily at 07:00 am and 15:00 pm. The tapping process consists of beating the inflorescence with a wooden mallet or tongs and then slicing the tip during 5 to 8 consecutive days until the juice began to flow. The beating and slicing process is continued until the complete inflorescence is exhausted. Bamboo recipients called Ampongs with a diameter of 10- 15 cm and a length of about 30-40 cm (hold 3-5 litres of juice) are used to collect the juice. Each tree requires an average of 4- 6 ampongs according to the amount of inflorescence at the same time. The unique feature of tapping the sugar palm is the manipulation of the inflorescence, that is, the partial crushing of young tissues. Clearly, the cells of the crushed tissues are not actually killed in the process.

The feeding trials were carried out at Or Pheasang Village, Kandoeung Commun, Baty District, Takeo Province, Cambodia between January and May, 1995. Seventy two males and females of castrated crossbred pigs (Yorkshire x Duroc x Landrace x Hainam) with an initial weight of 20 kg were distributed to twelve farmers who annually produce sugar palm syrup. Six pigs were assigned to each farmer usually in two subgroups of three each. Each farmers had access to at least 12 palm trees which it was calculated would provide enough juice to satisfy the needs of pigs when fed ad libitum. All the pigs were vaccinated after 15 days of adaptation period against Hog Cholera, Pasteurellosis, Salmonellosis and Erycipela diseases. They were also treated with anti-parasitic drugs (Ivomic injectable) once during the experiment. The pigs were identified with an ear notch system.

The pens were constructed from palm trunks, with thatched roofs of palm leaves and with a solid concrete floor providing an average of 1.66 m² per pig. The feed trough in each sub-group was divided in three sections and there was one water trough shared between both sub-groups with water always available. A continuous- flow, plastic-tube biodigester (Bui Xuan An et al 1994) was installed and connected to the manure outlet from each pen in order to produce fuel (biogas) for each household. During the experiment, the air temperature was between 20-30 ?C taken twice a day at 7:00 am and at 5:00 pm; the humidity was between 45 and 76% measured two times daily at the Meteorological recording unit situated at the National Cattle Breeding Station, 2 km from the experiment site. Rainfall was negligible as the period January to April is typically the driest time of the year.

Pigs were purchased by farmers individually at the end of December. For the adaptation phase the feeding system was divided into three periods. The first period of five days was two parts rice bran one part of boiled soya bean seed and palm juice, the second period was one half rice bran one half of boiled soya bean seed and palm juice and the last period was one third of rice bran two thirds of boiled soya bean seed and palm juice. After this adaptation period, the pigs were given 100% of the experimental diet consisting of 400 g/day of soyabean seed (weighed before boiling), sugar palm juice free choice, 5 g of salt and 5 g of lime daily mixed in the juice and 500 g/day of water spinach. At all times the pigs had free access to water.

The daily requirement of whole soya bean seed was weighed and soaked in water for 12 hours during the night and boiled in the morning for about two hours. The boiled soyabean was mixed with fresh palm juice, the lime and the salt and given three times a day at 07:00 am, 12:00 m and 5:00 pm. The water spinach was given at 500 g/day as a vitamin supplement. There was free access to water.

All pigs were weighed individually after 15 days of the adaptation period and during the experiment at intervals of 15 days. The yields of the palm juice and the brix (% dissolved solids) were recorded every month on three consecutive days. The sap from different collectors of the same tree was poured into a big container, well mixed and a sample taken for measuring brix using a hand refractometer.

The results were analysed by analysis of variance (ANOVA) using the General Linear Model program of Minitab (Minitab Statistical Software version 9.0 (Minitab Inc., 7081 Enterprise Drive, State College, PA 1680-708, USA).

All the sugar trees were located around the farmer households. The yields of juice are in Table 1. There was considerable variation among the families with yields per tree ranging from 2.76 to 6.37 kg/tree/day, presumably reflecting the skill of the tapper in extracting the juice from the inflorescence, although soil fertility may also have been a factor. Yield declined progressively as the harvest season progressed from a maximum in January of 5.76 kg/day to a minimum in April of 2.77 kg. There was also a significant difference between male and female trees with the latter producing 50% more than the former.

Table 1: Daily yield of sugar palm juice according to family, month and sex of the tree
FAMILY	kg/day	MONTH	kg/day
Pring Houy	2.76	Jan	5.76
Chhan Mak	3.08	Feb	4.81
Hay Yang	3.54	Mar	4.20
Thorn Punn	3.93	Apr	2.77
Sim Henn	3.98	SE	±0.15
Houy Kiel	4.53	Prob	0.001
Map Chreb	4.63
Pauv Pauv	4.69
Thol Onn	4.85
Thorn Chreb	5.08	SEX	kg/day
Tha Khorn	5.15	Female	5.24
Yem Khnol	6.37	Male	3.53
SE	±0.24	SE	±0.11
Prob	0.001

The pigs in each of the 12 households consumed palm juice avidly. As soon as they heard the sound of the bamboo collectors knocking each other the pigs became restless and crowded to the feed trough in anticipation of being fed. The daily liveweight gains were reasonably good ranging from 450 to 350 g/day. More important was profitability which was positive in all households and closely linked with the rate of growth (Figure 1). The comparison with sugar production is particularly interesting (Figure 2). Four of the seven families made no profit from making sugar when the opportunity costs of fuel were taken into account. By contrast, the highest profit from sugar was still less than the lowest profit from pigs.

 Figure 1.  Profit and live weight gains of pigs fed sugar palm juice and boiled soya beans
  on farms in villages in Cambodia

Figure 2.  Comparative profits from using sugar palm juice to feed pigs
 or to process into palm sugar

Feeding palm juice to pigs is not only profitable for people in the rural areas but also benefits biodiversity and the environment as firewood is not required. The farmers also save time to do other things when the palm juice is fed to pigs. It takes 2-3 hours to climb the trees and 1-2 hours to feed the pigs and clean the shed; but at the same time they are charging the biodigester with manure and can use the effluent as fertilizer for fish ponds, water plants or rice and fruit trees, with no harmful effect on the environment. For sugar production, they spend about 16 hours per day in collecting firewood and they must stay close to the stove where they boil the juice. It was stated by Dolbert (1991), in his survey of sugar palm syrup production in Cambodia, that the whole family is occupied almost all the time with sugar production.

It was shown that sugar palm juice can be used as the sole energy source for growing and finishing pigs. The pigs were healthy and had reasonably good rates of daily weight gain considering the low protein intake of only 150 g per day.

The average yield of juice per tree was 4.5 kg daily with a sugar content of 16%. In a 5 month harvest season this is equivalent to a production of 108 kg of sugar per tree. Assuming that the area occupied by each tree is 50m² this is equivalent to 21.6 tonnes of sugar per hectare per year which under rain-fed conditions indicates that the palm tree is both productive and efficient in use of solar energy.

Thanks are due to the team of Lutheran World Service Integrated Rural Development Program in Takeo/Kandal, especially to Mr Lor Bunnath and Oscar Paez Castro for their assistance and guidance. The research is part of the requirements for the MSc degree in Integrated Livestock-based Farming Systems for Sustainable Use of Natural Renewable Resources financed by the Swedish Agency for Research Cooperation with Developing Countries (SAREC).

(Received 1 October 1995)