Transgenic cotton (Gossypium hirsutum L.) cultivars tolerant to topical applications of glyphosate herbicide are now commercially available to producers. However, little information exists with regard to the tolerance of various cotton cultivars as affected by the timing of glyphosate applications. Therefore, the response of several glyphosate-tolerant transgenic cotton cultivars to various application timings of glyphosate was investigated. Two separate studies were conducted in 1995 and one in 1996 at the Delta Research and Extension Center located near Stoneville, MS. Treatments for both studies in 1995 consisted of seven glyphosate-tolerant cultivars (Coker 312-RR, Hartz 1215RR, Hartz 1220RR, Hartz 1244RR, Hartz 1330RR, Hartz 1380RR, and Hartz 1560RR) that were either unsprayed (that is, untreated control) or sprayed topically with 1.0 kg a.e. ha-1 glyphosate plus a non-ionic surfactant at 0.5% v v-1 at one of three different growth stages (four-, five-, and six-leaf stage). A different strain of each glyphosate-tolerant cultivar was evaluated in the two studies in 1995. Treatments in 1996 consisted of four glyphosate-tolerant cultivars (Deltapine 9683RR, Deltapine 9685RR, Deltapine 9687RR, and Coker 312RR) that were either unsprayed (i.e., untreated control) or sprayed topically with glyphosate at 1.0 kg a.e. ha-1 plus a non-ionic surfactant at 0.5% v v-1) at one of three different growth stages (three-, four-, and five-leaf stage) and two unsprayed conventional cultivars (Deltapine 5415 and 90). All transgenic cultivars responded similarly to glyphosate, and no differences in vegetative growth parameters or in total lint yield were found among any of the application timings in either year. However, mapping data revealed a sequential decrease in first position sympodia boll retention of the first three fruiting branches as topical applications of glyphosate were delayed. Retention values were 61, 44, 34, and 19% for the untreated, four-leaf stage, five-leaf stage, and six-leaf stage in one study in 1995 and 62, 55, 41, and 21%, respectively, for the same treatments in the second study. In 1996 retention values were 58, 45, 38, and 25% for the untreated, three-leaf stage, four-leaf stage, and five-leaf stage applications, respectively. The loss of early-season fruit caused by topical glyphosate applications resulted in a slight delay in maturity as measured by nodes above white flower counts. Although vegetative growth and total lint yield were unaffected in this study, it appeared that early-season fruit retention was negatively correlated and maturity was delayed as the timing of topical applications of glyphosate herbicide was delayed from the third true-leaf stage to the sixth true-leaf stage.