Tannins are antibiotic against various cotton pests. Thus, breeding for increased concentrations or toxicity of tannins should increase resistance to multiplepests. To determine the optimum leaf location for comparing tannins, concentrations were measured in each leaf of the main stem of eight pest-resistant Texas race stocks (Nos. 254, 277, 695, 810, 953, 1055, 1122, and 1123) and two cultivars (TAMCOT CAND-E and Stoneville 213). In the first 10 leaves counting from the cotyledonary node, maximum differences in tannin concentrations between the race stocks and the cultivars occurred in leaves 4 through 6. The mean tannin concentrations in leaves 4-6 were 54, 55, 86, and 148% higher in stocks 810, 254, 1122, and 953, respectively, than in 'TAMCOT CAMD-E'; tannin concentrations in the other stocks were not significantly higher than those in the cultivars. Only three race stocks (953, 1055, and 1122) had significantly higher tannin concentrations (65, 57, and 43%, respectively) in leaves 9-13 from the base than the commercial cultivars. When leaves four and five back from the terminal were compared one month after initial flowering of the cultivars, all race stocks had two to four times more tannin than did the cultivars. These results are probably misleading, because tannin concentrations increased progressively with each leaf formed, especially beyond leaves 10-12, and the photoperiodic race stocks formed greater total numbers of internodes and leaves than the cultivars once flowering and fruit set began. Consequently the high tannin concentrations measured after the initiation of flowering probably reflect the photoperiodic nature of the stock and would not be expressed in determinant hybrid progeny. Measurements of tannins in composite samples from leaves 4-6 from the base are recommended for breeding programs; measurements from leaves 10-12 may give useful supplemental information.