GIDEON HULATA - Agricultural Research Organization, Fish and Aquaculture Research Station, Dor, Mobile Post Hof Hacarmel 30820, Israel
ANA MII.STEIN -  Agricultural Research Organization, Fish and Aquaculture Research Station, Dor, Mobile Post Hof Hacarmel 30820, Israel
AMOS GOLDMAN -  Agricultural Research Organization, Department of Natural Resources, The Volcani Center, P.O. Box 6, Bet Dagan 50250, Israel

The multiple regression analysis presented in this paper was applied to data from experiments carried out during 1974- 1985 at the Fish and Aquaculture Research Station, Dor, Israel. The objective of these experiments was to test the possibility of complete or partial substitution of supplemented feed by manure. Due to the large number of variables affecting fish yield, or profit, multivariate statistical methods were used. Nonlinear relationships existed between fish yield (or profit) and each of the management variables (amounts of feed and manure, density and weight at stocking of common carp and tilapia). Multiple regression analysis identified stocking density of common carp and an interactive combination of amounts of feed and manure as the major factors affecting yield and profit. Manure had a stronger effect on profit than did the feeds. The highest profit was obtained in treatments in which a low ration of supplemental feed was added to a high amount of manure and not in the treatment in which only supplemented feed was applied. The analysis presented hem indicates the potential of multiple regression for the purpose of extracting information from fish culture data and identifying predictors of increased production and profit.