A double circadian oscillator model for quantitative photoperiodic time measurement in insects and mites

From studies on photoperiodic time measurement in insects it is known that different night lengths at the same side of the critical night length can have different inductive strengths. This means that nights of different length, and either longer or shorter than the critical night length, can have qualitatively different values. Nevertheless, few photoperiodic-clock models have been developed that are based on quantitative night-length measurement. In this paper a model is proposed that consists of two independent, circadian mechanisms. Both mechanisms determine the length of a night and give it a quantitative value, which is either zero or positive. One of the mechanisms (LN system) generally gives a scotophase a positive value when it is "long" (i.e. lights-on occurs when the LN oscillator is in its descending phase), whereas the other (SN system) gives a scotophase a positive value when it is "short" (i.e. lights-on occurs when the SN oscillator is in its ascending phase). In this particular context, therefore, "long" and "short" do not necessarily mean longer or shorter than the critical night length. The reasons for two time-measuring systems instead of one are: first, in some insects only long nights are accumulated, not short nights, or vice versa. Second, long nights are less sensitive to temperature than short nights. Third, in some cases it seems that long and short nights are determined in a different manner. These observations indicate that long and short nights could be determined by separate mechanisms. Responses generated by the proposed model parallel those observed experimentally with the spider mite, Tetranchus urticae and the aphid, Megoura viciae. General properties of the model are discussed and compared with Zaslavski's quantitative clock model which shares some features. Copyright 1998 Academic Press Limited