Nervous Systems 847 



response occurred to the CS (sound). Much evidence indicates that con- 

 ditioning and extinction occur not in sensory or motoneurones but rather in 

 interneurones. 



CONCLUSIONS 



Nervous conduction is a cellular process which varies little among different 

 groups of animals. High speed of reaction is associated with long fiber 

 tracts, large fiber diameter, and nerve sheaths. Thick myelin sheaths and 

 nodes are primarilv \ertebrate developments. Syncytial giant fibers evolved 

 at least twice— with cell bodies scattered along the nervous system in annelids 

 and crustaceans, and with cell bodies grouped in single ganglia in cephalopod 

 molluscs. Unicellular giant fibers are large neurones, and they grade in a 

 series into small fibers; enlargement of neurones has occurred many times. 



Synapses may be of two types, relay and integrative. Relay synapses are 

 regions of redistribution, are usually polarized, but do not require facilitation. 

 Polarization appears to depend on convergence of axons; in non-polarized 

 junctions equal surfaces are presented. Integrative synapses make incoming 

 impulses count for more or less than one in eliciting outgoing messages; re- 

 inforcement occurs bv facilitation and diminution by inhibition. 



The correlation between chemical and electrical events of synaptic 

 transmission is not clear, but limited sampling shows synaptic potentials 

 followed by som.a potentials of slower time course than axon potentials. 

 Most of the distributional properties of synapses evolved very early. In the 

 coelenterate nerve net, for example, there is already facilitation in such 

 high degree that it determines a sort of polarity at junctions. Central antag- 

 onism with reciprocal innervation of muscles is well dex^eloped in annelids 

 and may occur in lower phyla. FaciHtation may be early or delayed; early 

 facilitation represents addition of converging subliminal impulses and gives 

 the time course of decay of the excitatory process, whereas late facilitation 

 may be associated with convergence via delay-paths or with after-negativity of 

 motoneurones. Inhibition also may be direct or delayed: direct inhibition 

 may result from anelectrotonic conduction in short interneurones: delayed 

 inhibition may result from blocking by impulses in suitably timed delay-paths, 

 or from after-positivity. 



Peripheral conducting svstems or subepidermal networks ha\'e been largely 

 replaced by local reflex centers for activating fast muscles. It is quite possible 

 that the subepidermal plexus in the earthworm and in molluscan feet (but 

 not in echinoderms) is even now vestigial. In most phyla, diffuse networks 

 for the control of visceral musculature persist. 



Among invertebrates central nervous and neuromuscular mechanisms are 

 closely interconnected in evolution. In the coelenterate nerve net system 

 there is little distinction as to whether facilitation is neuroneural or neuro- 

 muscular. Some of the so-called peripheral conduction in worms and molluscs 

 may be muscular. In arthropods the neuromuscular junctions with multiple 

 innervation, facilitation, and inhibition are as important as the central ner- 

 vous system in gradation of movement. (Ch. 16) 



As one passes toward the most specialized groups— cephalopod molluscs, 

 insects and higher crustaceans, birds, and mamrjials— one sees increasing ceph- 

 alization in the nervous system. The sensory basis for cephalization is clear, 



