3. CONTROL SYSTEMS AND RHYTHMIC PHENOMENA IN CELLS 21 



enzyme on a metabolite will all take some time. A return signal from the altered 

 metabolic state to the genetic locus, acting in the form of a repressor (or co- 

 repressor), will then reduce the rate of mRNA synthesis at this site some time 

 after the increase has occurred. An oscillation will certainly occur in such a 

 system during a change of state, but it could be a damped oscillation (decreasing 

 to a completely stationary condition with no continuing oscillation) if the time 

 interval between synthesis of mRNA and feed-back of metabolite is small 

 enough, and if there are "self-damping" effects in the system (e.g. if the rate of 

 degradation of mRNA is a function of its concentration). The cell might have 

 found that such oscillations were a disadvantage to adaptation and survival, 

 and so they would be selected against in the same way that engineers usually 

 try to select against parasitic oscillations. 



There is an increasing body of evidence, however, which suggests that there 

 is some fundamental periodicity occurring in the dynamic organization of 

 cellular processes. The evidence comes from studies on rhythmic activities of 

 cells and organisms. This is a field of rapidly-expanding proportions, especially 

 in recent years with the extremely interesting and fundamentally significant 

 studies which have been made on the nature and widespread occurrence of 

 biological clocks. The most forceful proponent of the idea that rhythmic 

 activity is an all-pervasive feature of temporal organization in biological 

 systems is Pittendrigh (1960), and he has recently suggested (Pittendrigh, 1961) 

 that the primary oscillations underlying this organization arise as a result of 

 the occurrence of feed-back devices for the control of physiological activity; 

 "The Darwinian Daemon has certainly had plenty of physiologic oscillations 

 to work with, because his commonest device in installing regulators — from 

 control of heartbeat to that of protein synthesis — is negative feed-back. And 

 one of the innate tendencies of such feed-back systems is to oscillate." An 

 earlier proposal of a similar nature was made by Hastings and Sweeney (1959); 

 while Lwoff and Lowff (1962) have also explored the possibihty of an intimate 

 connection betaeen feed-back control processes and rhythmic or periodic 

 activity in biological systems at many different levels of organization, but 

 with particular reference to muscular activity. There would seem to be here 

 an opening for the establishment of a biological dynamics of control processes 

 whose fundamental postulates could apply to a very wide range of biological 

 system, for biophysics to demography, not to mention closely-related fields 

 such as economics. 



If a biological statistical mechanics and "thermodynamics" can be con- 

 structed on the basis of periodicities in the kinetics of biological control 

 mechanisms, then an enormous experimental and theoretical advantage will 

 have been gained for biology. The analysis of dynamic systems in terms of 

 periodic phenomena is as old as science, which began with observations on the 

 cyclic motions of the heavenly bodies. Mathematics has naturally taken on the 

 structure required of it by observational science, and "harmonic analysis" has 

 come to occupy an absolutely central position in mathematical analysis. It 

 might almost be suggested that man's mind has been constructed in such a 

 manner that he tends to see all process in the world in terms of cycles, as did the 



