Livestock Research for Rural Development 12 (1) 2000

Citation of this paper

Fresh water fish silage as protein source for growing-fattening pigs fed sugar palm juice

Khieu Borin, Sim Chou and T R Preston*

Department of Animal Production and Health, 
Ministry of Agriculture Forestry and Fisheries, 
Phnom Penh, Cambodia.
* University of Tropical Agriculture Foundation
PO Box 2423, Phnom Penh, Cambodia



Eighteen families who owned at least 10 sugar palm trees and were interested to rear pigs participated in this study which was carried out in Kandal village, Ang Snoul district in Kandal province, 40 km west of Phnom Penh city. The juice from 157 sugar palm trees of both sexes (119 males and 38 females) was used as the principal energy source for fattening pigs.   

The average daily juice yield varied among families (P=0.001) in the range of 3.00 to 4.92 kg/day with a median value of 3.89 kg/day. The corresponding values for the Brix (total sugars) content were a range of 11.6 to 12.8% (P=0.001) and a median value of 12.3%. 

Seventy two pigs (23 males and 49 females) of different crossbreeds were reared during 5 months on a diet of ad libitum sugar palm juice, 500 g/day of rice bran and 400 g/day of fresh water fish silage which was made by mixing the partly eviscerated fish with sugar palm syrup and rice bran (Fresh weight ratios were: Fish, rice bran, sugar palm syrup; 50:10:40) and storing anaerobically in sealed plastic bags. The average daily intake of sugar palm juice was 8 kg.  The protein intake was estimated to be 130 g/pig/day. The growth rates of the pigs varied among families in the range of 325 to 476 g/day with an average of 405±47 g/day. 

Keywords: Silage, Fresh water fish, sugar palm juice, pigs, protein


The cereal grains traditionally grown and fed to livestock in industrialized countries are not appropriate crops for tropical latitudes where opportunities for year-round plant growth give comparative advantages to perennial crops capable of much higher yields of digestible energy per unit of area and of time (Preston 1998). Feeding systems using the products and/or by-products of sugar cane, the African oil palm, cassava and the sugar palm have been developed for all classes of livestock and are slowly finding acceptance in many tropical countries (Sarria et al 1990; Preston 1995; Ocampo 1994; Khieu Borin and Preston 1995; Perez 1997). In the early stages of development of these feeding systems, soya bean meal or ensiled soya seeds, alone or with fish meal, were used to supply the protein. However, soya beans do not adapt readily to tropical latitudes and it is therefore desirable to develop alternative protein sources to lessen dependence on imports.

Normally, the farmers in Cambodia start to fatten pigs after harvesting the rice when the by-products - rice bran and broken rice - become available. The usual number of pigs in each household is from 2 to 5 raised predominantly for fattening. Under the traditional system the pigs are raised as scavengers and almost never are given any protein-rich supplement. 

The Tonle Sap inland lake plays an important role in fish production in Cambodia.  It is thought that it provides about 70% of the protein consumed by the 10 million or so people of Cambodia. The Tonle Sap lake was reported in 1990 to occupy an area of 460,000 ha.  It is one of the richest inland fishing lakes in the world and is considered to be nearly 10 times as productive as the best fishing grounds in the North Atlantic, with fish yields of at least 65 kg/ha/yr if calculated on the basis of the dry season area of the lake. The fishing activities start when the water flows out from the lake at the beginning of the dry season (December). There are two times in January and February that the mass of fish is harvested along the Tonle Sap river.

The research reported in this paper examines the possibility of using, as a protein supplement for pigs, the small fish harvested from the Tonle Sap river which are rejected as unsuitable for human consumption.

Materials and methods

Ensiling the fish

Small fish taken at the peak harvest periods from Tonle Sap were partly eviscerated before mixing with rice bran and sugar palm syrup in proportions of (% fresh weight): fish 50,  rice bran 10, sugar palm syrup 40.  The sugar palm syrup had a Brix (soluble sugars) value of  75%. The mixed ingredients were put into thick plastic bags, packed tightly and the bags sealed to keep the silage under anaerobic conditions.

Location and experimental plan

The experiment was carried out between December, 1996 and April, 1997 (dry season) in Kandal village, Ang Snoul district, Kandal province. Rainfall in this area is about 1,200 mm per year, mostly concentrated in the months of August-October. Eighteen farmers who traditionally produced sugar palm syrup were chosen for the experiment. Each farmer had access to 10-25 palm trees, which guaranteed that the juice yield would be sufficient to feed 4 pigs at libitum.


A total of 72 crossbred pigs (23 males and 49 females) were used in the present experiment. They were purchased by the individual farmers with the assistance of the researchers. The initial weight ranged from 14-35 kg; they were of unknown age. All the pigs were vaccinated against hog cholera, pasteurellosis and salmonellosis. They were also treated with an anti-parasitic drug (Ivomic injectable) once after the vaccination. Individual pigs were identified by the spots or print mark on the skin.

Housing and environment

The four pigs on each farm were housed in a  pen (total area 6.0 m²) made from local materials, most of which came from the sugar palm tree. The roof was thatched with palm leaves and the floor was covered with cement.  

Diets and feeding

The diet was comprised of 500g of rice bran, 600g of fish silage (FWFS), sugar palm juice ad libitum and 5 g of salt per pig per day. In addition the pigs were offered water hyacinth (Eichhornia crassipes) at the rate of about 1 kg per pig per day. At all stages the pigs had free access to fresh water. 

Palm juice

A total of 157 palm trees  (119 males and 38 females) of unknown age were used. Most of them were located around the farmer households. The juice was collected twice daily (07:00 h and 15:00 h). A piece of bark from the tree Shorea cochinchinensis was placed in each juice collector as a means of inhibiting fermentation.


The yield of palm juice from each tree and the Brix value were measured every month on three consecutive days (the first three days of each month) during the period December-April. The juice yield was recorded with a hand balance. The sugar content (Brix value) was measured using a hand refractometer (Atago N1, Japan). 

The pigs were weighed individually after a 15 day adaptation to the diets. This was used as the initial live weight. Subsequent measurements of liveweight were done every 15 days throughout the experimental period using a "beam" balance. To facilitate the weighing procedure, the pig was pushed into a bamboo basket which was then attached to the balance. At least three persons were needed to operate this system. The liveweights were taken in the early morning prior to feeding.

Statistical analyses

The yield of palm juice and the liveweights of the pigs were analysed by the GLM (General Linear Model) ANOVA programme of Minitab Version 10 (1993). Sources of variation were families and error. The Basic Statistics option using the One sample "T" test was used to calculate the means and standard errors for the Brix values of the juice.

Results and discussion

Sugar palm juice

The average daily juice yield varied among families (P=0.001) in the range of 3.00 to 4.92 kg/day with a median value of 3.89 kg/day (Figure 1). The corresponding values for the Brix (total sugars) content were a range of 11.6 to 12.8% and a median value of 12.3%. These values are slightly lower than were observed in a previous study conducted in 1995 in Bati district, Takeo province, where average daily juice yields were 5.0 kg with a Brix of 13.3%. There was an overall tendency for the Brix value to increase with yield but the relationship was not strong (r = 0.23) (Figure 1).

wpe3.jpg (22036 bytes)

Both yield and Brix increased as the harvest season advanced from December through April (Figure 2).  In this case the the Brix value was closely related with yield (r = 0.97). This is an opposite trend to that observed in the earlier study (Khieu Borin and Preston 1995) where yield declined from the peak of 5.76 kg/day in January to the lowest value of 2.77 kg/day in April. There were also no significant differences in yield and Brix between sexes, which is in contrast with the previous study (Khieu Borin and Preston 1995) where female trees yielded more than males.

wpe4.jpg (14508 bytes)

The lower values for yield and Brix value and the different tendencies due to harvest period and sex of tree in the present study, compared with the earlier one, could be due to many factors as the sugar palm trees are very sensitive to climatic changes and manipulation of the inflorescences as was observed by Khieu Borin et al (1995).   


The growth rates of the pigs (Figure 3) varied among families in the range of 325 to 476 g/day with an average of 405±47 g/day. The average daily intake of sugar palm juice was 8 kg per pig, in addition to the 400 g of fish silage and 500 g of rice bran. This provided approximately 130 g daily of crude protein.  Despite this relatively low protein intake, the daily weight gain of the pigs was superior to that recorded in the earlier trial (365 g/day) when the protein supplement was 400 g daily of boiled whole soya beans providing about 156 g daily of protein (Khieu Borin et al 1995). At low protein intakes the balance of essential amino acids in the protein supplement is the major determinant of performance and in this respect ensiled fish is obviously superior to soya beans..

wpe3.jpg (18097 bytes)


The results of the present trial confirm the potential of the sugar palm tree to produce highly digestible carbohydrates, which can be used as the main alternative energy source for growing and fatterning pigs in small scale farms in the rural areas of Cambodia. In the months of January and February when fish is in abundance, the ensiling of the small fish is a practical and cheap method of preserving this resource. It is more hygienic and preserves better the nutritive value as compared to sun-drying on the sides of the roads.

Despite the encouraging results reported in this paper, the adoption of the technology is slow.  Evaporation of the juice to make sugar is still the preferred procedure, even though the overall economic returns are higher from feeding juice to pigs.  Some reasons for preferring sugar production are: 


The authors wish to express their sincere thanks to the families who actively participated in the experiment and shared the experience with us. The study was made possible through the support from the International Foundation for Sciences (Grant No B/2353).


Khieu Borin and Preston T R 1995 Conserving biodiversity and the environment and improving the wellbeing of poor farmers in Cambodia by promoting pig feeding systems using the juice of the sugar palm tree (Borassus flabellifer). Livestock Research for Rural Development  (7) 2:

Ocampo A 1994 Utilizacion del fruto de palma africana como fuente de energia con niveles restringidos de proteina en la alimentacion de cerdos de engorde. Livestock Research for Rural Development  (6) 1:1-7

Perez Rena 1997 Feeding pigs in the tropics.  APHP Series No 132, FAO: Rome

Sarria P, Solano A and Preston T R 1990 Utilización de jugo de caña y cachaza panelera en la alimentación de cerdos. Livestock Research for Rural Development  (2) 2:92-100

Preston T R 1995 Recent developments in use of sugar cane as livestock feed. World Animal Review No 82 pp 84-89.  FAO: Rome

Preston T R 1998 Livestock production from local feed resources. In: Proceedings VIII Conference of World Association for Animal Production. Seoul: Korea (in press)

Received 27 September 1999

Go to top