Metamorphosis of Rana pipiens tadpoles may be retarded when the light phase of the light/dark (LD) cycle is shortened or when thyroxine (T4) is given in the dark because melatonin peaks during the dark. Injection of premetamorphic tadpoles in spontaneous metamorphosis with melatonin (1 5 pg) retarded tail growth and hindlimb development on 18L:6D but had no significant effect on 6L: 18D. During induced metamorphosis (30 pg/liter T4), melatonin injections retarded tail resorption on 18L : 6D and accelerated it on 6L : 18D, but did not affect the hindlimb. When melatonin was injected during T4 immersion at different times in the photophase on 18L : 6D (L onset 0800 hr), tail regression was retarded by melatonin at 1430 or 2030 hr. At 0830 hr, shrinkage of tail length was accelerated whereas tail height was not affected. Tail tips in vifro induced to resorb by 0.2 pg/ml T4 in Niu-Twitty solution regressed more slowly in the presence of melatonin (10 or 15pg/ml) than with T4 alone on both 6L: 18D and 18L:6D. The findings implicate melatonin in LD cycle effects on tadpole metamorphic rate in vivo, show the importance of the time of melatonin injections, and indicate that melatonin antagonizes the metamorphic action of T4 at the tissue level.

The influence of circadian (6L:18D, 18L:6D) and noncircadian (12L:18D, 15L:15D) light/dark (LD) schedules on the cell cycle was studied in the hindlimb epidermis of late premetamorphic tadpoles of the frog, Rana pipiens. Using tritiated thymidine autoradiography, percent labeled mitoses (PLM) curves were obtained for 3 cell cycle determinations on each LD schedule, starting at 0, 8, and 16 hr for the 24 hr, and at 0, 10, and 20 hr for the 30 hr LD regimens. From the PLM curves, the durations of the whole cycle (T), the DNA synthetic phase (S), and the pre-(G 1 +1/2M) and post-(G 2 +1/2M) synthetic gaps, which included mitotic time, were determined. The data were compared with previous work on 12L:12D. The shape of the PLM curves and the resulting phase durations were characteristic for each LD regimen. T was longer where L was less than D and shorter where L was greater than D. The relative duration of S decreased, while G 1 +1/2M increased, as L lengthened on the 24 hr regimens, whereas on the 30 hr regimens the opposite occurred. On both 24 and 30 hr schedules, as L lengthened the percentage of cell cycle time in G 2 +1/2M increased. The findings show the specific influence of a change in the LD schedule, including noncircadian regimens, on particular cell cycle phases in a vertebrate. In a separate experiment, the labeling index was studied over a 24 hr period in control tadpoles and in those treated with hydrocortisone or melatonin. Since only melatonin shifted the phase of the labeling index rhythm when injected during the light, this hormone may account for the effects of the LD schedule on the cell cycle.