804 PERIODIC FUNCTIONS IN MAMMALS 



functions, a comparison of the environmental "lock-in" of rhythm in 

 human beings and mice will reveal important differences in phase. This 

 interesting species difference notwithstanding, certain basic mecha- 

 nisms such as the adrenal cycle are shared by several mammals. Some 

 type of a periodic cellular sequence of events in turn most likely 

 characterizes many forms of life, including those devoid of conven- 

 tional hormones and/or nerves. 



The term "periodicity analysis" will be used with reference to 

 various methodological and other considerations arising from the 

 study of physiologic 24-hr periodicity (Halberg, in press). The choice 

 of this term, not qualified as to length of period, is made for the sake 

 of brevity. The term circadian [(L) circa, about + (L) dies, day] will 

 be used to denote daily periods which may differ from 24 hr by not 

 more than a few hours. Circadian periods thus may be slightly shorter 

 or longer than 24 hr. Other definitions have been given elsewhere 

 (Halberg, 1953, 1955a, and in press). 



METHOD 



In mammals, the lighting regimen of the environment affects, of 

 course, many body functions, the approach to its effect on a given 

 function, in turn, being largely determined by the available techniques. 

 But over and above the choice of a given biochemical or biophysical 

 procedure, the analysis of lighting effects upon body rhythms, or the 

 study of the rhythms themselves, poses additional methodological 

 problems. The choice of experimental animal, the uncertainties of 

 measurement and sampling procedure (Halberg, 1946; Koehler et 

 al., 1956), the conditions of study, the variabihty of time series, and 

 the analytical statistics that become thus indispensable — all of these 

 are general problems of periodicity analysis. Our particular approach 

 to them has repeatedly been discussed (Halberg, 1953, 1955a, and in 

 press). Only one point may deserve emphasis herein, in view of its 

 broader potential usefulness in biologic research. It is the desirability 

 of obtaining numerical estimates of the period and the amplitude in 

 our work on a given biologic periodicity, just as we do in dealing 

 with a physical periodicity. But to get these two estimates reliably is 

 much more difficult in the case of physiologic rhythms than in that of 



