Impacts of cotton acreage planted and weather on cotton yields were estimated for several major production areas. An ordinary least square regression model was developed using data from the 1955-80 crop years. Production areas evaluated included California, Texas High Plains, Texas Rolling Plains, Arkansas, and the Delta area of Mississippi. The variables selected for explaining cotton yields included spring temperature, summer temperature, autumn temperature, preplant precipitation, summer rainfall, acreage planted, and skip-rowed acreage. These variables generally accounted for a high proportion of the variability in lint yields. The model was less satisfactory for California than for the other 4 areas.