6. THE RELAXATION TIME OF THE EPIGENETIC SYSTEM 97 



and these oscillations can occur in only one direction, i.e. the constraints are 

 such that initially one variable must increase before the other, since messenger 

 RNA must always be present before protein can be produced in the system. 

 Thus irreverisble thermodynamics imposes upon the motion of the biochemical 

 control system a constraint which has an immediate and obvious biological 

 interpretation in terms of the necessary relations holding between messenger 

 RNA and protein. The other condition required by thermodynamics, that the 

 system be far from an equilibrium state (in the usual chemical sense of this 

 term) if it is to oscillate, is also very clearly satisfied in our system due to the 

 almost complete irreversibility of macromolecular synthesis. 



It is thus possible to demonstrate compatibility between thermodynamic 

 laws and the fundamental dynamic processes which underlie the statistical 

 mechanics developed in connection with biochemical control mechanisms. The 

 question of consistency between theories developed in connection with physical 

 processes on the one hand and biological processes on the other, therefore does 

 not present any real difficulties and no contradictions arise. One is simply deal- 

 ing with different classes of phenomena; and it is logically and conceptually 

 simpler to construct a biological theory on the basis of processes which are 

 characteristic of the biological system one is dealing with rather than on the 

 basis of physical or chemical procedures which were designed to fit quite differ- 

 ent situations. There does not seem to be any reason why biologists should not 

 start their model-building at a level which is convenient for the description of 

 biological systems, rather than starting with physical and chemical principles. 

 Although some care must be exercised in this procedure so that no incom- 

 patibilities are introduced, it nevertheless appears to be a more reasonable way 

 of analysing biological systems than the much more difficult and often com- 

 putationally impossible task of deducing biological behaviour from the 

 "lowest" analytical level, that of chemistry and physics. 



