Experiments were carried out to determine whether a semidian (12 h) rhythm in flowering response operates in Pharbitis nil as the basis for photoperiodic time measurement, The effect of 5 min far-red light followed by 85 min dark (FRD) given 4, 8, 14 and 22 h before the end of a 48 h photoperiod on night-break timing and critical night length was determined, When given 4 h before the end of a 48 h photoperiod, an interruption with FRD advanced the phase of the circadian rhythm in the night-break inhibition of flowering, In contrast, earlier interruptions of the photoperiod had no effect on the phase of the rhythm, The critical night length was modified by FRD given 4 h (shortened) or 8 h (lengthened) before the end of the photoperiod; when given at other times FRD did not alter the critical night length, The results are discussed in relation to the basis for photoperiodic timekeeping, with particular reference to suggestions for the involvement of a semidian rhythm, A circadian model based on the concept of limit cycles is described.
