Livestock Research for Rural Development 17 (4) 2005 Guidelines to authors LRRD News

Citation of this paper

Promotion of cassava leaves silage utilization for smallholder dairy production in Eastern coast of Tanzania

P Y Kavana, Kiddo Mtunda*, Adebayo Abass** and Vianney Rweyendera**

Livestock Research Centre, P.O. Box 5016, Tanga, Tanzania,
*Sugar Cane Research Institute, P.O. Box 30031, Kibaha, Tanzania.
**IITA -Tanzania, C/O ARI-Mikocheni, P.O. Box 6226, Dar-es-Salaam, Tanzania.


A preliminary study was conducted from March to July 2004 to investigate the effectiveness of ensiling cassava leaves in reduction of free cyanogens and the feeding value of cassava leaf silage. A feeding trial was conducted in Muheza district involving 10 crossbred dairy cows (Friesian x Boran) owned by smallholder dairy farmers.

Preliminary results indicated that ensiling of cassava leaves was effective in reduction of free cyanogens. Ensiling for three months reduced free cyanogens from 289 to 20 mg/kg of silage. Animals that were provided with cassava leaf silage produced more milk than the control group (9.9 vs 7.6 litres/cow/day). Milk fat produced by cows during cassava leaf silage feeding period was higher than before the silage feeding period (4.0 vs 3.3 %). No significant difference (P>0.05) was observed in terms of Solid Not Fat (SNF) content of milk between the two periods. It was envisaged that cassava leaf silage had a high proportion of by pass protein that contributed to the increase in milk production of the experimental animals.

This study indicates that cassava leaf silage has a potential for improvement of milk production from crossbred dairy cows during the dry season. However, it has been recommended that more farmers should be involved in different locations so as to derive a general conclusion.

Key words: by pass protein, cassava leaf silage, free cyanogens, HCN


Generally, dairy animal production in Tanzania is challenged by seasonal availability of forage. There is a shortage of feeds during the dry season while surplus forage material exists during the rainy season. In order to alleviate the dry season feeding problems for dairy animals, it has been considered plausible to conserve the surplus forage material that exists during the rainy season. Hay and silage making have been described as appropriate methods of forage conservation. However, hay making in eastern coast of Tanzania is rather difficult because in most cases the optimal hay making period coincides with the wet season which limits sun drying of forage material. Silage making therefore seemed to be appropriate because the technique is less weather dependent (Wilson and Bringstock 1981). Nevertheless, conservation of forage in form of silage is not common to smallholder dairy farmers in Tanzania. One reason for that is lack of simple methods of silage making.

The most limiting nutrient for dairy cattle during the dry season is protein due to inherent low crude protein content of tropical grasses that are used as the basal diet. To minimize the problem of low protein content in dairy rations, leguminous forages such as Leucaena leucocephala, Gliricidia sepium, and other herbaceous legumes are provided separately or ensiled with grasses for dry season feeding. However, little has been done to explore usefulness of cassava leaves as dairy cattle feed in cassava growing areas of Tanzania. According to Montaldo (1977), cassava plants can withstand defoliation for several years if they receive adequate fertilization and irrigation. Under such management there is a potential of harvesting up to 4 tonnes of protein per hectare annually. Nevertheless, cassava leaves are wasted in many parts of Tanzania after harvesting of cassava roots. Utilization of cassava leaves as dairy cattle feed could open an avenue to proper utilization of cassava plants to contribute in poverty alleviation.

Many authors reported that ensiling of cassava leaves is an appropriate method to conserve them for dry season feeding (Limon 1992; Bui Van Chinh et al 1992; Du Thanh Hang 1998; Ly and Rodríguez 2001). This study was carried out to initiate utilization of cassava leaves as feed resource in cassava growing areas in Eastern coast of Tanzania. Many researchers in Asia showed that cassava leaf has a high protein content ranging from 16.7 to 39.9% of dry matter with almost 85% of crude protein fraction as true protein (Ravindran 1991). Ensiling has been reported as an effective way of reducing cyanide (HCN) content in cassava (Tewe 1991; Nguyen Thi Loc et al 1996). Stable minimum HCN content (20 mg/kg of fresh cassava foliage) occurred after 56 days of ensiling (Chhay Ty al 2001). It was suggested by Gomez (1991) that maximum concentration of HCN in animal feeds is 100 mg HCN/kg feed on dry matter basis. This study therefore tried to establish the trend of HCN reduction during fermentation as well as the feeding value of Cassava leaf silage for crossbred dairy cattle in Eastern coast of Tanzania.


Materials and methods

Study Area

Tanga region occupies the north-eastern corner of Tanzania between 4 - 6 °S and 37 - 39 °E. The region covers 3% of Tanzania and provides land for 7% of the country's population. The region is divided into 6 administrative districts namely, Korogwe, Lushoto, Muheza, Pangani and Tanga. This study was conducted in Tanga and Muheza districts. Rainfall pattern of the area is bimodal with rains falling in the months of March through June and between October and December. Average maximum temperature varies between 26 and 33 °C while minimum temperatures are between 20 and 24 °C. Relative humidity is around 70%. The coastal belt is predominated by low fertile light sandy soils. The interior soils are medium fertile red loam soils but valleys are characterized by blocky cracking clay. Tanga municipal is an urban area where no cultivable land is available except in peri-urban areas. The peri-urban in this case refers to the peripheral areas of agglomerations often characterised by scattered and isolated residential developments along main highways (Bryant and Johnson 1992; Mlozi 1997). The area is characterised by availability of public amenities such as water, electricity and transport facilities. Milk marketing is easier than in other districts and farmers get better prices of milk and inputs because of reduced transport costs (Rutamu and Uden 1999). There is less infestation of tick and tsetse transmitted diseases. All feeds are brought in from outside the municipal area in the form of green fodder or hay. Manure remains in the area with little usage. About 20% of all dairy cattle in Tanga region are kept in Tanga urban district. Muheza district includes a downtown with similar features as that of Tanga urban. In addition, the district covers high altitude areas that are cool and suitable for pure exotic breeds of dairy cattle. In general, Muheza is more fertile and has relatively higher rainfall than other districts (1600 mm/year). Coconuts, oranges, cassava, bananas, sugar cane and spices are the main crops in the district.

Forage Material

Cassava leaves, twigs and roots were purchased from farmers in Tanga peri-urban and Kicheba village in Muheza district. Sweet varieties were used for this study.


Cassava root chips were used as water-soluble carbohydrate additive (Photo 1).

Photo 1. Chipping of cassava roots Photo 2. Chopping of cassava leaves

Plastic bags with capacity of up to 15 kg of forage material were used as silos. This method was chosen for easy transportation of packed silage to the clients.

Ensiling Process

Harvested cassava leaves were allowed to wilt for a minimum of 12 hours in a well-ventilated shed. Chopped forage material (Photo 2) was thoroughly mixed with cassava chips in 1:4 ratio (cassava root chips: leaves). The mixed forage material was put in two inserted polyethylene bags and compressed by foot to expel air. The polyethylene bags were tightened by using rubber bands and stored to allow fermentation to occur.

Experimental Animals

Five crossbred (Friesian x Boran) dairy cows owned by 5 smallholder dairy farmers in Muheza District were used for the feeding trial. Data from another 5 crossbred cows in the same farms that were not provided with experimental diet was used for comparison. The weights of the animals ranged from 350 to 400 kg.

Preparation of Silage Samples for Analysis

The silage from 5 bags selected at random was thoroughly mixed and then a sample of about 1000 g was taken. The sample was sufficient to provide enough sub-samples required for chemical analyses, determination of pH and dry matter (DM) content. The samples were kept in polyethylene bags and stored in a deep freezer at -10 °C until when they were transported to Sokoine University of Agriculture (SUA) and Tanzania Food and Nutrition Centre (TFNC) laboratories.

Determination of pH

Sample weighing 40 g was soaked in 200 ml of cool distilled water for 12 hours. The mixture was then filtered and supernatant divided into 4 aliquots for determination of pH using a pH meter.

Determination of Dry Matter and Chemical Composition

Dry matter of samples was determined by freeze drying while samples for determination of chemical composition were dried in an oven at 60 °C for 48 hours. Dried samples were ground to pass through 1 mm screen in a hammer mill. Samples were chemically analysed for determination of crude protein (CP) according to standard procedures (AOAC 1990). Neutral Detergent Fibre (NDF) and Acid Detergent Fibre (ADF) were determined as described by Göering and Van Soest (1970). Crude protein fractionation was performed according to Licitra et al (1996).

Feeding Trial

The feeding experiment was conducted between May and June 2004 to determine the feeding value of cassava leaf silage in terms of milk yield and quality. Five crossbred dairy cows (Friesian x Boran) owned by 5 smallholder dairy farmers were used. Another 5 cows on the same farms served as controls. The experimental animals were allowed to feed according to normal farmers practice and then provided with 10 to12 kg of cassava leaf silage in the evening after milking. The animals had access to silage for the entire night. The animals were provided with silage for 9 days prior to the data collection period of 21 days. The amount of silage given and the amount left at 7.00 am in the morning was recorded. Silage intake was calculated by subtracting amount of silage eaten from the silage provided to the animal.  The main ration (basal diet) that was available during the day comprised a mixture of grasses namely Digitaria mombasana, Urochloa pullulans, Bothriochloa radicans, and Panicum maximum species. All the animals were provided with 2 kg of maize bran during milking.

Experimental Design

The experimental design was a completely randomised design (CRD) with farmers deemed as replicates. The farmers involved had at least two lactating crossbred dairy cows. The other lactating cows that were not provided with cassava leaf silage were considered as a control group. The animals were kept in different pens for the whole night. Individual cow records of milk yields prior to the experimental period were regarded as covariate.

Statistical analysis

The data were analysed by using General Linear Model (GLM) of Statistical Analysis System (SAS Version 612) computer package. The following statistical model was used to compare the effect of diets on milk yield and quality parameters.

Υijk = µ + αi + b(χij - x) + Ωk + еijk


            Υijk = The performance of the j-th cow assigned to i-th feeding regime in k-th Farm
            µ = Overall mean
            αi = The effect of the i-th feeding regime
            b = Regression of Υijkon χij
χij= The record of the covariate in the individual cow indicated by the subscripts
            x = Overall mean of the covariate
            Ωk = Effect due to k-th farmer's management
            еijk = A random error specific to each individual cow


Cassava leaf silage had a higher crude protein and proportion of acid detergent soluble crude protein (ADSCP) than maize bran and the basal diet (Table 1).

Table 1. Chemical composition of feedstuffs (as % of DM for CP, NDF and ADF)







Cassava leaf silage






Maize bran






Basal diet






CP = crude protein; ADSP* = Acid detergent soluble crude protein expressed as percentage of total crude protein; NDF = Neutral Detergent Fibre; ADF = Acid Detergent Fibre.

Ensiling the mixture of cassava leaves and chips reduced levels of HCN from 289 to 20 mg/kg after 3 months  (Table 2). The dry matter content of the silage was 42.9% and the pH 4.85. Average daily intake of silage was 3.73 kg DM.

Table 2. Free cyanogen content of cassava roots and forage material before and after ensiling


Free cyanogens, mg/kg

Fresh cassava leaves


Fresh cassava leaves mixed with root chips


Fresh cassava chips


Cassava leaf silage


Cassava root chips silage


Average milk yield of cows that were provided with cassava leaf silage was higher than control animals (Table 3: Figure 1). There were no differences among farms during the experiment, when the data were corrected by covariance for pre-experiment yields   Milk fat content was also increased (Table 4).

Table 3. Mean values for milk yields before and during the experiment according to treatment and farm





Treatment effect


















Farm effect
































# Adjusted by covariance for pre-experiment milk yield
## Un-adjusted yield

Figure 1: Average milk yield of cows supplemented with cassava leaf silage and control animals

Table 4. Milk quality before and after cassava leaves silage feeding


Crude protein, %

Milk fat, %

SNF*, %

Before silage feeding




After silage feeding




*SNF = Solid Not Fat


Acid detergent-soluble proteins (ADSCP) comprise proteins that are more slowly degraded in the rumen than buffer-soluble fractions of crude protein. A larger proportion of ADSCP is thus believed to escape the rumen and be finally degraded in the lower gut (Elizalde et al 1999). It is therefore indicated that cassava leaf silage has the potential of contributing a larger proportion of amino acids for milk synthesis than the maize bran and basal diet fed to lactating crossbred dairy cows during the study. The potential of dried cassava foliage (cassava hay) to replace concentrates for lactating cows has been observed in several experiments (see Wanapat 2001). Evidence for 'bypass"  protein characteristics in cassava leaves was reported by Ffoulkes and Preston (1978) who showed that fresh cassava leaves could completely replace soya bean meal in a diet of molasses and urea fed to fattening cattle.  Promkot  and Wanapat (2003) reported that cassava hay had a content of bypass protein comparable with that in cottonseed cake. Wanapat (2001) suggested that the condensed tannins contained in cassava leaves could have a potential role in forming tannin-protein complexes that escape rumen degradation to reach duodenal digestion.
The drastic decrease in the level of cyanide after ensiling implies that ensiling is an effective way of reducing HCN in cassava leaves and roots before feeding them to dairy animals. This is line with many reports in the literature (see Ly and Rodríguez 2001). The observed variation in performance of lactating crossbred dairy cows among farmers indicates the need to involve more farmers in different locations in order to derive a general conclusion from the study.
The increase in milk fat percentage after feeding cassava leaf silage is similar to the effect of feeding cassava hay to lactating cows (Wanapat et al (2000). 



The authors wish to acknowledge financial support obtained from International Institute of Tropical Agriculture (IITA) that enabled execution of this study. Special thanks to Officers in Charge of LRC Tanga, SRI Kibaha and IITA Mikocheni,  for permission granted to utilize all facilities available during execution of this study. We would like to convey our sincere gratitude to all farmers who volunteered their animals to be involved in this study. Extension staffs in Muheza district are highly appreciated for their cooperation and commitment. Lastly but not least, we thank all colleagues who participated in one way or another to support this study. With this attitude, we can make our clients move a step towards better life.


AOAC 1990 Official methods of analysis (15th Edition). Association of Official Analytical Chemists, Arlington, VA.

Bryant C R and Johnson T R R 1992 Agriculture in the city's countryside. Belhaven Press, London.


Bui Van Chinh, Le Viet Ly, Nguyen Huu Tao, Do Viet Minh 1992 ''C'' Molasses and ensiled cassava leaves for feeding pigs. Results of research 1985 -1990. Agricultural Publishing House, Hanoi. pp 46.


Chhay Ty, Ly J and Rodríguez Lylian 2001 An approach to ensiling conditions for preservation of cassava foliage in Cambodia; Livestock Research for Rural Development. (13) 2:


Du Thanh Hang 1998 Digestibility and nitrogen retention in fattening pigs fed different levels of ensiled cassava leaves as a protein source and ensiled cassava root as energy source; Livestock Research for Rural Development. (10) 3:


Elizalde J C, Merchen N R and Faulkner D B 1999 Fractionation of fibre and crude protein in fresh forages during the spring growth. Journal of Animal Science, 77: 476 - 484.


Ffoulkes D and Preston T R 1978 Cassava or sweet potato forage as combined sources of protein and roughage in molasses based diets: effect of supplementation with soybean meal. Tropical Animal Production   (3) :186-192


Göering H K and Van Soest P J 1970 Forage fibre analyses (apparatus, reagents, procedures and some applications). Agriculture Handbook No. 379. ARS, USDA, Washington, DC.


Gómez G G 1991 Use of cassava products in pigs feeding. Pigs News and Information 12:387-390.


Licitra G, Hernandez T M and Van Soest P J 1996 Standardization of procedures for nitrogen fractionation of ruminant feeds. Animal Feed Science and Technology, 57: 347 - 358.


Loeza Limon R L 1992 Ensilage of cassava products and their use as animal feed. In; Roots, tubers, plantains and bananas in animal feeding (Editors: D Machin and A W Speedy). FAO Animal Production and Health. Paper No. 95: 99-110.


Ly J and Rodríguez L 2001 Studies on the nutritive value of ensiled cassava leaves for pigs in Cambodia. In: Cassava as livestock feed (Editors: T R Preston and M Wanapat). July 23-25 2001, Khon Kaen University, Thailand


Mlozi R S 1997 Urban and peri-urban livestock farming and the environment. In: Proceedings of the 24th Scientific Conference of the Tanzania Society of Animal Production. Arusha August 5th - 7th 1997. pp 43 - 62.


Montaldo A 1977 Whole plant utilization of cassava for animal feed. In; Cassava as animal feed (Editors: B. Nestel and M. Graham) IDRC 095e. Ottawa, pp 95-106.


Nguyen Thi Loc, Ogle R B and Preston T R 1996 On farm and on station evaluation of cassava root silage for fattening pigs in central Vietnam. MSc. Thesis. Swedish University of Agricultural Sciences.


Promkot C and Wanapat M 2003  Ruminal degradation and intestinal digestion of crude protein of tropical protein resources using nylon bag technique and three-step in vitro procedure in dairy cattle;  Livestock Research for Rural Development (15) 11 Retrieved , from


Ravindran V 1991 Preparation of cassava leaf products and their use as animal feed. In: Roots, tubers, plantains and bananas in animal feeding (Editors: D Machin and Solveig Nyvold). FAO Animal Production and Health Paper No. 95: 111-122.


RutamuI B and Uden P 1999 Effect of farm location and management practices on growth performance of crossbred heifers in the coastal areas of Tanzania. In: Proceedings of 26th Scientific Conference of the Tanzania Society of Animal Production August 3rd - 5th 1999. pp 38 - 55.


Tewe O O 1991 Detoxification of cassava products and effects of residue toxins on consuming animals. In; Roots, tubers, plantains and bananas in animal feeding (Editors: D. Machin and Solveig Nyvold). FAO Animal Production and Health Paper No. 95: 81-95

Wanapat  M.,  Petlum A and Pimpa  O  2000a   Supplementation of cassava hay to replace concentrate use in lactating Holstein Friesian crossbreds. Asian-Australasian J. Anim. Sci., 13:600-604.

Wanapat  M 2001 Role of cassava hay as animal feed in the tropics  In: Cassava as livestock feed (Editors: T R Preston and M Wanapat). July 23-25 2001, Khon Kaen University, Thailand  

Wilson P N and Bringstock T D 1981 Improved feeding of cattle and sheep. London, UK: Granada Publishing Company. pp 324.

Received 22 November 2004; Accepted 21 December 2004; Published 1 April 20055

Go to top