
Some circannuals persist and phase-drift under constant environmental conditions (Figure 11/IV). When the deviation of the natural period from the calendar year is small, a full-cycle scan by the circannual phase would take longer than the individual's life span. In natural lighting, in continuous light, in continuous darkness or on a fixed 12-hourly alternation of light and darkness, a circannual of large amplitude with little or no damping from year to year characterizes the ovarian weight of the Jamuna River catfish (Figure 11/IV). Under the particular natural lighting and other conditions studied, the best-fitting circannual period was very close to the exact 365-day calendar year, namely 371 days; under constant environmental conditions the best-fitting circannual period was longer than exactly one year (393 days in continuous light and 400 days in continuous darkness) (Figure 11/IV). These findings suggest endogenicity.
Physiologic monitoring combined with chronobiologic analysis can recognize the presence of a heightened risk. The circannual amplitude of diastolic blood pressure, assessed longitudinally by several 24-hour profiles of clinically healthy women in North America and Japan, correlates negatively with the familial and personal risk of developing high blood pressure or related diseases later in life, as does the amplitude of the hormone aldosterone (Figure 11/III). The circannual amplitude of aldosterone correlates negatively with the circadian MESOR of diastolic blood pressure as well as with a questionnaire-derived cardiovascular disease risk index. Discrimination and classification techniques applied to these data have singled out aldosterone as a classifier of cardiovascular disease risk while corroborating the circannual stage-dependence of the discriminating power of aldosterone as a classifier of this condition.
For breast cancer risk assessment, circadian profiles are less discriminating than circannuals, Group means of the 24-hour MESORs of circulating prolactin of clinically healthy Minnesotan women, at high or low risk of developing breast cancer, and of Japanese women, at a low such risk, are plotted for each season (Figure 11/IIA); a circannual variation becomes apparent (Figure 11/IIB); its amplitude is larger for the low-risk than for the high-risk Minnesotan population (not shown) and is largest in a Japanese population, at the lowest breast cancer risk when the study was carried out. The individual circannual prolactin amplitude correlates negatively with the total risk load assessed by questionnaire (P=0.025), Figure 11/IIC. The circannual amplitude of the concomitantly assessed TSH correlates positively rather than negatively with the total risk load for developing breast cancer (P=0.019), Figure 11/IIC. The circadian amplitudes of these same two hormones also correlate with the risk of developing prostate cancer.