Circadian rhythm is entrained in the 24-hour time interval by periodic factors in the environment, known as zeitgeber. But most rotating work schedules are outside the range of the entrainment of the pacemaker timing the human circadian sleep - wake cycle. It has been postulated that physiological and emotional disturbances occur in most human functions when the circadian rhythm is disturbed. So application of circadian principles to the design of shift schedules can aid in maintaining the temporal integrity of the circadian system and thereby minimize for the shift worker any detrimental consequences of circadian disruption. This study was a quasi-experimental study to test the effect of shift intervals for the clinical nurse on the circadian rhythm. Twenty nurses newly employed in general units of two hospitals were selected as an experimental group and twelve college nursing students as a control group. Both groups were selected according to an established criteria using a purposive sampling technique. Ten subjects were assigned to a weekly shift group and another ten to a biweekly shift group engaged in a semi -continuous shift schedule(sunday off) with a backward direction, that is, morning -evening - night shift. The control group worked a morning shift for 42 days. Oral temperature rhythm, waking tim, sleep - wake cycle, fatigue, and mental performance were measured during the experimental period. The data collection period was from April 30, 1990 to June 10, 1990. MANOVA, paired t-test, ANOVA, and Student Newman Keuls method were used for statistical analysis. The results are summarized as follows. 1. Phase delay in the acrophase of temperature rhythm was shown according to the backward rotating shift. A complete adaptation to work on the night shift was achieved between the sixth and ninth day of the night shift. 2. There was no difference in either waking time or sleep- wake cycle according to the duration of the working day for every shift group. Significant difference was found in the waking time and the sleep -wake cycle for subjects on the morning, evening, and night shift in both of the shift groups(weekly shift group : λ＝0.121, p＜0.01, λ＝0.112, p＜0.01, biweekly shift group : λ＝0.116, p＜0.01, λ＝0.084, p＜0.01). 3. There was no difference in fatigue between the first working day and the last working day for the control group and for the biweekly shift group. In the weekly shift group, physical fatigue was significantly different for the first day and the sixth day of the night shift(t＝-2.28, p＜0.05). Physical fatigue and total fatigue on the first day of the night shift showed a significant difference among the control group, the weekly shift group, and the biweekly shift group(F＝5.79, p＜0.01, F＝4.56, p＜0.05). There was a significant difference between the shift groups and the control group(p＜0.05), Physical fatigue, neurosensory fatigue and total fatigue on the last day of the night shift showed a significant difference among the control group, the weekly shift group, and the biweekly shift group(F＝12.65, p＜0.01, F＝7.77, p＜0.01, F＝9.68, p＜0.01). There was a significant difference between the shift groups and the control group(p＜0.05). 4. No difference in mental performance was seen between the first day and the last day of work in each case. An arithmatic test on the first day of the night shift revealed a significant difference among the control group, the weekly shift group, and the biweekly shift group(F＝3.79, p＜0.05). There was a significant difference between the shift groups and the control group(p＜0.05) . The digital symbol substitution test and the arithmetic test on the last day of the night shift showed a significant difference among the control group, the weekly shift group, and the biweekly shift group(F＝3.68, p＜0.05, F＝5.55, p＜0.01), and both showed a significant difference between the shift groups and the control group(p＜0.05). Accordingly, this study showed that..