The effects of daylength and testosterone on the initiation and progress of moult in Starlings Sturnus vulgaris

The effects of daylength and of testosterone implants, before and after the beginning of moult, on the timing and rate of primary moult have been quantified. Female Starlings Sturnus vulgaris were moved from natural daylength in February to 13 h or 18 h of light per day (13L: 11D or 18L: 6D). Some of the birds on 18L: 6D were left on 18L: 6D throughout the experiment and others were transferred to 13L: 11D after 6 weeks, before moult had begun, or after 12 weeks, after moult had begun. Birds kept on 18L: 6D began to moult before birds kept on 13L: 11D, but the subsequent rate of moult was the same in both groups. A decrease in daylength before moult started slightly advanced the onset of moult. A decrease after moult had begun increased the speed of moult. Castrated male Starlings on 18L: 6D were given testosterone implants for different periods before or after the beginning of moult. Testosterone treatment which ended before moult would normally have started had little effect. Treatment extending beyond the normal start of moult considerably delayed or even prevented the onset of moult. Moult was arrested in birds which received testosterone after moult had begun. On removal of testosterone implants, moult began again from the point where it had stopped, but in some birds, all of the feathers which had been regrown recently were dropped again and regrown. These results are discussed in relation to the different patterns of moult seen amongst different species.