The effects of starter fertilizer and other placement methods on yield and fiber quality of cotton grown under reduced and conventional tillage systems were studied in a three-year field experiment. Conventional tillage common to the region was compared with a reduced tillage system on a calcareous Orelia sandy clay loam located in the Coastal Bend region of southern Texas. Both reduced and conventional tillage were compared at fertilizer rates of 0-0-0, 40-40-0 and 60-60-0. Fertilizer placement combinations included: all knifed preplant, all broadcast and disc incorporated, 2/3 knifed preplant plus 1/3 starter, and 2/3 knifed preplant plus 1/3 sidedressed. Additional placement treatments included: pop-up fertilizer (13-13-0 placed with seed and 27-27-0 knifed preplant) and repeated annual placement of 40-40-0 and 60-60-0 in the same band as separate treatments with reduced tillage. Plant populations were generally unaffected by fertilizer rate and tillage systems. However, lowest plant populations were measured with pop-up fertilizer in two years of the three-year study. Lint yields and treatment response varied with years. Fertilizer response was limited to the initial 40-40-0 rate and generally was greatest when banded preplant in both reduced and conventional tillage. Band placement was superior to broadcast incorporated with reduced tillage in two years of the three-year study. Splitting fertilizer into preplant (2/3) and sidedress (1/3) had no beneficial effects on yields in the conventional tillage. With reduced tillage, substantial decreases in lint production were measured from similar sidedressing treatments under both suboptimal and above average rainfall conditions. No yield response was recorded to pop-up and starter fertilizers. Fiber strength was the only quality parameter affected by fertilizer rate and tillage.