The circadian basis of ovarian diapause regulation in Drosophila melanogaster: is the period gene causally involved in photoperiodic time measurement?

Females of a wild-type strain of Drosophila melanogaster (Canton-S), and of several clock mutants (period), were able to discriminate between diapause-inducing short days and diapause-averting long days with a well-defined critical daylength. The critical daylengths of a short-period mutant (pers) and a long-period mutant (perL2) were almost identical, both to each other and to that of Canton-S. The critical daylength of an arrhythmic mutant (perol), however, was about 3 hr shorter than that of Canton-S, and that of per- was about 5 hr shorter. Exposure of Canton-S females to Nanda-Hamner experiments, consisting of a 10-hr photophase coupled to a dark phase varying between 4 and 74 hr, showed (1) that the photoperiodic clock in D. melanogaster measures nightlength rather than daylength, and (2) that photoperiodic time measurement is somehow based on (or affected by) constituent oscillators in the circadian system. Nanda-Hamner results for the period mutants all showed similar profiles regardless of genotype, or the presence or absence of per locus DNA. These results suggest that photoperiodic induction and locomotor activity do not share a common pacemaker in D. melanogaster, and that the per gene is not causally involved in nightlength measurement by the photoperiodic clock, although flies in which the per locus is missing (per-) or defective (perol) show an altered critical value.