The authors analyzed circadian features of two period-lacZ (per-lacZ) fusion genes in transgenic strains of Drosophila. Both genes manifest circadian fluctuations of mRNA levels, but fluctuations of only the larger chimeric protein are apparent. Fusion protein cycling is indistinguishable from the behavior of wild-type per protein (PER), including apparent temporal regulation of phosphorylation state. Several arguments indicate that the difference in the two constructs is proper regulation at the level of protein turnover: the smaller protein has much higher levels; a beta-galactosidase degradation products is visible in both strains but fails to manifest cycling, presumably due to a long half-life; and only the noncycling proteins accumulate as a function of adult age. The large cycling fusion protein also undergoes modest cycling in an arrhythmic per01 background. This is light dependent, resembles the regulation of the timeless protein (TIM) by light, and reflects a documented fusion protein-TIM interaction. The results are discussed with respect to the posttranscriptional regulation that is necessary for proper cycling of both PER and TIM as well as for clock function.