THE ORGANIZATION OF THE PRIMITIVE CENTRAL 



NERVOUS SYSTEM AS SUGGESTED BY EXAMPLES 



OF INHIBITION AND THE STRUCTURE OF 



NEUROPILE 



G. Adrian Horridge 



Gatty Marine Laboratory and Dept. of Zoology, University of St. Andrews, Scotland 



In 1957 I laid out an explanation of the transmission of excitation in the 

 nerve nets of various types of corals, and, by impUcation, of all coelenterates, 

 in terms of a model which was a network of connected units, each unit repre- 

 senting a neuron or through-conducting group of neurons. The essential 

 feature of this model of this special type of nervous system was that the 

 effective connexions between units of the model were randomly distributed, 

 and at the time I suggested that random connexions may make up some part 

 of the organization of other nervous systems. The present considerations are 

 an exploration of this line of thought as appHed to some other invertebrate 

 central nervous systems. A number of examples will be presented to show 

 that diffuse and widely ramifying neurons having a general activating or in- 

 hibitory effect occur commonly in invertebrates and it is suggested that an 

 explanation of their action does not require a set of specific connexions. 

 Alongside them, and interacting with them, are neurons which are more 

 hmited regionally and could have more specific connexions. As part of this 

 problem, therefore, it is essential to consider the formation of specific con- 

 nexions. Finally, it is suggested that there has been a progressive evolution 

 away from random connexions towards a greater individuafity and speci- 

 ficity of the interrelationships between neurons. 



The complexity of a model of a nervous system depends on the number of 

 specifications which must be laid out in order to determine the formal pattern 

 of connexions of the model. A simpler model is therefore one with fewer 

 such specifications. Here Ues the reason for considering the connexions within 

 the model to be random until evidence is available from the animal to prove 

 that patterns exist in the structure. This is essentially the application of the 

 null-hypothesis of statistics to the connexions of a model of the nervous 

 system which is based on a combination of anatomical and physiological 

 data. Order is not assumed, in general, in the study of relationships in any 

 natural phenomenon; it is looked for and then, at a certain level of significance, 

 is demonstrated to be present. 



395 



