Starter fertilizers have been utilized in several crops for many years and recently in the Mississippi Delta to possibly increase cotton production. Studies were established at Stoneville on a Dundee silt loam from 1987 through 1990 to evaluate cotton response to narrow-band applications of starter fertilizer. Urea and ammonium polyphosphate (11-37-0) were evaluated at rates from 0 to 15 lb N/A (in 3-lb increments) applied in a 4-in band over the planted row shortly after planting. Supplemental fertilizer was broadcast so that each plot received an equal quantity of urea or 11-37-0 to total 100 lb N/A (85 lb N/A applied preplant as urea-ammonium nitrate solution). Yield results from this 4-year study indicated that neither 11-37-0 nor urea, applied in narrow bands directly over the row, produced consistent yield increases in irrigated cotton. When means were determined over 4 years, significantly higher lint yields were obtained with 6 lb N/A (8.8 lb P/A) applied as 11-37-0 as compared to either 0 or 3 lb N/A (0 or 4.4 lb P/A). The highest lint yield was obtained with 12 lb N/A as urea but this yield was not significantly higher than those produced from other treatments. After 4 years there was no significant difference between the two sources. In other studies at Stoneville where 12 gal/A of 11-37-0 was used (15 lb N/A and 22 lb P/A) in combinations with other P and K fertilizer placement combinations, 11-37-0 starter fertilizer produced significant yield increases 2 out of 3 years. When averaged over the duration of the study (1988-1990), an additional 48 lb lint/A/year was obtained with 11-37-0. In another study (1989-1990) under non-irrigated conditions with 12 fertilizer K placement combinations, application of 11-37-0 resulted in significant yield reduction in 1990 and no response in 1989. The soil test P level was high in this study, indicating less chance for a yield response. Since consistent yield responses have not been obtained with starter fertilizer, either with urea or 11-37-0, consideration should be given to individual field situations in determining the need for starter fertilizer. The 11-37-0 does provide an additional means of supplying needed P. However, applying starter fertilizer at planting slows down the planting operation, and some consideration must be given to this fact.