9/9 ULTRA STABILITY IN THE ORGANISM 



A molecular basis for memory? 



9/8. What is necessary that any material entity should serve as a 

 step-mechanism in ultrastability ? Only that it should be a 

 step-mechanism, that it should be able to be changed by the 

 essentia] variables, and that it should have an effective action on 

 the reacting part R. 



As a dynamic system, the brain is so extremely sensitive, and 

 is such a powerful amplifier, that we can hardly put a limit to the 

 smallness of a physical change that still has a major effect in 

 behaviour. Even a change on a single molecule cannot be dis- 

 missed as ineffective, for many events in the nervous system 

 depend critically on how some variable is related to a threshold; 

 and near the threshold a small change may have great con- 

 sequences. 



It is therefore not impossible that molecular events should 

 provide the step-functions. There are plenty of events that 

 might provide such forms: whether a molecule is in the dextro- 

 or the laevo- state, in the cis- or trans- state; whether a hydrogen 

 bond does or does not exist; whether a double bond between 

 carbon atoms lies in this plane or that; and so on. Such bases 

 would have the advantage in the living organism of providing the 

 necessary function at a minimal cost in weight and bulk, matters 

 of considerable importance to the free-living organism. 



Pauling has discussed these possibilities and has suggested 

 limits that would narrow the field of search. If the molecular 

 entity is too small, thermal agitation will prevent it from showing 

 the constancy which it must have if it is to act as basis for such 

 behaviour as that of Skinner's pigeons (S. 1/14). If too large, it 

 will be unsuitable for the miracle of ' miniaturisation ' that has 

 actually been achieved in the mammalian brain. 



9/9. With all these possible forms of step-mechanism in mind, 

 it is difficult for us to say much that is definite about the feedback 

 channels from essential variable through step-mechanisms to the 

 reacting part R of Figure 7/5/1. Clearly there is not the least 

 necessity for the channels to consist of anatomical or histological 

 tracts ; for if the step-mechanisms were molecular, the channel to 

 them might be biochemical or hormonic in nature; while the 

 channel from them to R might be extremely short, and of almost 



127 



