Twilights widen the range of photic entrainment in hamsters.

The range of entrainment of the circadian rhythm of locomotor activity was compared in four groups of Syrian hamsters (eight animals per group) initially exposed to daily light-dark (LD) cycles with either abrupt transitions between light and darkness (LD-rectangular) or simulated twilights (LD-twilight). Lighting was provided by arrays of white light-emitting diodes; daytime illuminance (10 lux) and the total amount of light emitted per day were the same in the two conditions. The period (T) of the LD cycles was then gradually increased to 26.5 h or gradually decreased to 21.5 h, at the rate of 5 min/day. Under LD-rectangular, the upper and lower limits of entrainment were 25.0 to 25.5 h and 22.0 to 22.5 h, respectively, whereas under LD-twilight, 50% of the animals exposed to the lengthening cycles were still entrained at T = 26.5 h and 50% of those exposed to the shortening cycles were still entrained at T = 21.5 h. In a second experiment, two groups of hamsters were exposed to fixed T = 25 h LD-rectangular (n = 15) or LD-twilight cycles (n = 7). Only 33% of the animals entrained in LD-rectangular, whereas 86% of the animals entrained in LD-twilight. Free-running periods in constant darkness were longer following successful entrainment to T = 25 h but did not differ between the animals that entrained to LD-rectangular and those that entrained to LD-twilight. The widening of the range of entrainment observed in LD-twilight indicates that twilight transitions increase the strength of the LD zeitgeber. In LD-twilight, successful entrainment to T = 26.5 h was accompanied by an expansion of activity time to 16.52+/-1.22 h, with activity onsets preceding mid-dusk by 12.56+/-2.15 h. Together with earlier data showing similar phase response curves for hour-long dawn, dusk, and rectangular light pulses, these results suggest that the effect of twilights on the range of entrainment may involve parametric rather than nonparametric mechanisms.