Smallholders with dual-purpose cattle production systems in most Central America experience a shortage of forages during the dry season (4-8 month. As a result, substantially lower milk production and weight gain occurs. Dual-purpose operations seeking to maximize milk and beef production in the dry season can produce and feed hay or silage to their livestock. The increase in milk and beef production due to feeding hay or silage during the dry season on commercial dual-purpose cattle productions systems, the production costs of making silage or hay and the benefits as a result of feeding silage or hay are estimated and compared in Honduras and Costa Rica. Due to feeding silage or hay, farmers in Honduras and Costa Rica have increased milk production during the dry season. The costs of feeding hay are lower in both countries, although farmers feed a higher amount (kg DM/cow/day) than silage. Feeding silage or hay to milking cows is profitable in both countries. The income-cost ratio and the net benefit ($/cow/day) due to feeding these feedstuffs are positive. Farmers with milking cows in Honduras realise a higher income-cost ratio and net benefit due to feeding silage and hay than farmers in Costa Rica. Likewise, beef cows or young livestock supplemented with these feeding alternatives don't loose weight during the dry season. In Honduras as well in Costa Rica, it is profitable to feed silage or hay to beef cows or young livestock. Currently, no silage and a small hay market exist in Honduras. Possibly, these will develop due to an increasing demand for these profitable feedstuffs. In Costa Rica thanks to the efforts of agricultural organisations, a hay market of different qualities is developing.

The growth response of Tilapia fish (Oreochromis niloticus) to supplementation with fresh or dried leaves of a sweet variety of cassava was studied in 1 m deep ponds (each 4 m * 2.5 m * 1 m). Rice bran or rice bran mixed with cassava root meal was provided as the energy source (20-24% of feed DM), the mixed feed being given at the rate of 5% of fish LW. There were 30 fish per pond with average initial weight of 6 g per head. The treatments were arranged as a 2*2 factorial in a Completely Randomized Design (CRD) with 4 replicates. The first factor was cassava leaf processing (sun-dried or fresh); the second factor was energy source (rice bran or rice bran mixed with cassava root meal). The cassava leaves and roots were from a local “sweet” variety traditionally planted by farmers for human consumption. The experiment was conducted for 100 days. Daily gain in weight and length, and the ratio weight: length after 100 days of growth, did not differ among treatments. All the fish survived and there was no evidence of HCN toxicity. Water quality parameters were not affected by the treatments. There were no interactions between sources of protein and of energy. It is concluded that the feeding of cassava leaves with rice bran can be readily adopted by small scale farmers as both these feed resources are readily available in rural areas.