ARRANGEMENT OF STRUCTURAL ELEMENTS. 225 



motor nerve to the muscle which it controls, although 

 the connection here is secondary. In the same way the 

 sensory epithelial cells arouse impulses in the fibres with 

 which they are associated, although again the connection 

 is not genetic. Here and there in the central system 

 nerve fibres are found, apparently terminating in small 

 disc-like expansions on the surfaces of cells. In some 

 cases the nerve fibre terminates in an elaborate brush, 

 which forms an enclosing basket about the body of a 

 second cell. In other cases the neuron forms its final 

 brush in the neighbourhood of the neuron, or the 

 dendrons of another cell. 



On the other hand, it is not easy to make out a close 

 relation between the branches of the different cells, 

 as they interweave with one another. In many cases 

 where it is assumed that the impulse passes from a 

 terminal brush of a neuron into the branches of 

 another cell, the specimens do not show continuity 

 between the two. The difficulty, therefore, is to conceive 

 the anatomical relations, so that the impulses may not 

 be stopped by interruptions in their course. 



It must be assumed either that these various branches 

 are structurally continuous with one another, although 

 the continuity is hard to demonstrate ; or that con- 

 tiguity of these structures is sufficient, the impulse in 

 some way passing across the interval ; but whichever view 

 is taken, the actual relations, as shown in sections, 

 require to be largely completed by hypotheses. As the 

 matter now stands, however, the greater weight of 

 evidence is in favour of the notion of contiguity. 



Many problems are bound up in this general question. 

 For example, an incoming impulse reaches the nervous 

 system by one of the dorsal root fibres. These branch 

 immediately after entering the spinal cord, one branch 

 passing cephalad and the other caudad. 



