THE REFLEX CIRCUITS 65 



were related through the associational tract (A, A) passing be- 

 tween them. 



In the case of the dog's experience just described the neural 

 mechanism was undoubtedly much more complex than our dia- 

 gram, though similar in principle, and the associative memory 

 process involved was probably vividly conscious (cf. p. 295). 

 But the simpler types of "associative memory" which have been 

 experimentally demonstrated in many of the lower organisms 

 may involve no more complex mechanism than this diagram, and 

 it is by no means certain that any conscious process is there 

 present. 



It must be kept in mind that in higher vertebrates all parts of 

 the nervous system are bound together by connecting tracts 

 (internuncial pathways). Some of these tracts are long, well- 

 defined bundles of myelinated fibers whose connections are such 

 as to facilitate uniform and clear-cut responses to stimulation. 

 Others are very diffuse and poorly integrated. Permeating the 

 entire central nervous system is an entanglement of very deli- 

 cate short unmyelinated fibers. This nervous felt-work (neuro- 

 pil) is much more highly developed in some parts of the brain 

 than in others. It is not well adapted to conduct definite ner- 

 vous impulses for long distances, but it may serve to diffuse or 

 irradiate such impulses widely. Where tissue of this sort is 

 mingled with myelinated fibers it is termed the "reticular for- 

 mation" (see pp. 65, 127, 158, 304). 



These manifold connections are so elaborate that every part 

 of the nervous system is in nervous connection with every other 

 part, directly or indirectly. This is illustrated by the way in 

 which the digestive functions (which normally are quite auton- 

 omous, the nervous control not going beyond the sympathetic 

 system, see p. 241) may be disturbed by mental processes whose 

 primary seat may be in the* association centers of the cerebra) 

 cortex; and also by the way in which strychnin-poisoning 

 seems to lower the neural resistance everywhere, so that a 

 very slight stimulus may serve to throw the whole body into 

 convulsions. 



It follows that the localization of cerebral functions can be 

 only approximate. Every normal activity has what Sherring- 

 ton calls its reflex pattern, whose anatomical basis is a definite 

 5 



