Counting circadian cycles to determine the period of a circasemilunar rhythm in a marine insect.

Semilunar and lunar rhythms are often controlled endogenously, but the mechanisms of their respective free-run periods, when external factors are absent, are mostly unclear. In this investigation, the authors studied the mechanism controlling the period of the circasemilunar emergence rhythm of a marine midge, Pontomyia oceana, in southern Taiwan. Experimental approaches were adopted with various artificial light-dark (LD) periods, or T, from 22 to 28 h per cycle in the first experiment, and 18 to 30 h per cycle in the second experiment, as treatments on the same cohorts of midge larvae. The responses in emergence days were directly proportional to the magnitude of the treatments, just as that predicted by the frequency demultiplication hypothesis. A counting mechanism is thus the only hypothesis supported by this finding. To further test whether it is endogenous oscillations that are counted, submultiples as well as multiples of 24 h, i.e., 6, 12, 24, and 48 h per cycle, were used as T. The midges under all these treatments emerged at similar days. This result supports the hypothesis that endogenous circadian oscillations, not external LD cycles, are counted in this circasemilunar emergence rhythm of the marine midge. This paper reports a first case supporting the frequency demultiplication hypothesis in a circasemilunar rhythm that is based on counting the cycles of endogenous circadian rhythms.