Fig. 5

Changes in melanocyte numbers (A), frequencies of apoptotic melanocytes (B) and collection of melanocyte carcasses on nylon membranes (C). (A) In wild type, melanocyte numbers do not differ among days during the first 11 days of development (F₇,35=1.20; P=0.3). In spa^b5 mutants, fewer melanocytes than wild type are present at 2.4 days (60 hours; F_1,10=30.87; P<0.001) and these cells are lost over the next several days (F_7,33=45.8; P<0.0001). On each day, dorsal stripe melanocytes were counted in 5-7 larvae of each genotype. (B) TUNEL staining is evident in spa^b5 mutant melanocytes by 4 days and persists at least through 8 days (n=43 larvae, 3903 melanocytes), but is not observed in wild-type melanocytes (n=15 larvae, 2025 melanocytes). TUNEL staining was assessed in dorsal stripe melanocytes in whole-mount, Fast-redstained embryos. (C) The timing of melanocyte carcass recovery on nylon membranes corresponds to the time course of melanocyte loss from spa^b5 mutant larvae. Error bars in A and B are 95% confidence intervals. Error bars in C are 95% confidence intervals, with intervals summed across days. Counts in A-C were gathered on different days using different groups of larvae and rearing densities.