Our investigations on laboratory animals and on human subjects of different ages have shown regular changes in amplitude of the circadian rhythm of many homeostatic systems during postnatal ontogeny [1]. The amplitude of these rhythms rises initially and reaches a maximum in organisms of reproductive age. During aging the amplitude diminishes, and at times disappears completely. On this basis we have formulated a "spinning top" hypothesis. We judged changes in amplitude both visually and on the basis of the results of Cosinor analysis [2]. To obtain further proof and confirmation of our hypothesis we undertook dispersion analysis, which does not require a model describing the shape of the oscillations. The "Cosinor" procedure assumes a sinusoid, and on that basis it postulates that the decrease in amplitude calculated by the "Cosinor" procedure is connected only or partially with a change in shape of the oscillatory process, as it differs increasingly from a sinusoid. The second factor which can determine an apparent reduction of amplitude is connected with desynchronization between animals, for circadian curves represent the sum of the data for many individuals. In grouped "Cosinor" analysis better coincidence can be expected between circadian rhythms and synchronization, whereas in early ontogeny the mechanisms of protraction are only just maturing, and in the late stages they are gradually lost. Because of differences of age, definite dispersion can be expected between individuals. This factor must have a greater effect during aging than in the period of growth. The aim of the present investigation was to assess the possible role of these factors in age changes in the amplitude of the circadian rhythm established previously.
