Tropical Grasslands (1995) Volume 29, 142149
Component forage yield and quality of grass–legume cropping systems in the Caribbean
Agricultural Experiment Station, University of the Virgin Islands, Kingshill, St Croix, USVI
Development of compatible, persistent, grasslegume mixtures could alleviate acute, seasonal livestock feed deficiency in the semi-arid tropics. A 2-year experiment was conducted on St Croix in the Caribbean to determine component forage yield and quality of either sole grass or binary mixtures of forage sorghum cultivars or a millet-elephantgrass (M-E) hybrid with 2 native legumes (Leucaena leucocephala and Desmanthus virgatus), using alley-cropping technology. The soil was a mildly alkaline (pH > 7.8) Fredensborg clay (fine carbonatic, isohyperthermic, Typic Rendolls Mollisols). Annual dry matter yields of grass or grass plus legume ranged from 17–20 t/ha under normal rainfall but 9–11 t/ha in a droughty, second year. The perennial M-E hybrid-based system produced the highest dry forage yields of 24 t/ha initially, due to multiple harvests, but was not so productive in the second year. Crude protein (CP) concentration and in vitro organic matter digestibility (IVOMD) of grasses averaged 8.4% and 58.8%, respectively, regardless of cropping system. Leucaena forage and prunings for mulch were superior in CP (20% vs. 12%) and IVOMD (61% vs. 51%) to those of desmanthus. Leucaena and more persistent selections of M-E hybrids grown in an alley-cropping system could become a sustainable system for forage production in the Caribbean.