of the Vertebrate Skeleton. 29 



segments ; and so the process must go on, the vertebrge in- 

 creasing in nmnber and extending further towards the head, 

 till the basis of a vertebral column is elaborated. So far as I 

 am aware, this hypothesis is in accord with the sum of the 

 facts, and gives an explanation of their relation to each other. 

 And not only does it account for the original existence of a 

 vertebral column, but for its subsequent modifications, and for 

 the repetition of the successive similar soft parts (muscles and 

 nerves) which are correlated with the bones. 



But so far we only account for the centrum of a vertebra. 

 In our usual conception of it, especially as seen in the fish's 

 tail, it includes an arch on the dorsal part, called the neural 

 arch, which covers the neural column, and a similar arch on 

 the ventral side, called the hfemal arch, which covers a blood- 

 vessel. In dissecting a fish, the muscles in the tail of the 

 dorsal and ha3mal sides of the animal are seen J;o be as like 

 each other as are the neural and heemal arches ; so that it will 

 be in accord with the mechanical basis on which this investi- 

 gation started to conclude that in both cases a like force has 

 produced a like result. 



But how? If we grant the differentiation of an initial 

 caudal segment of the notochord by muscular power, then as 

 those lateral muscles of the tail, acting obliquely, enlarge, 

 they would, with increasing force, become competent to set up 

 a separate ossification upon the notochord at each of the mar- 

 gins of their overgrowth. And these points, it is to be re- 

 marked, coincide with the points of origin from the centrum 

 of the lateral parts of the two arches. When once these 

 kinetic epiphyses are brought into existence, the lateral mus- 

 cular attachment would ensure their growth, and the dorsal 

 and ventral muscles would as surely draw them towards each 

 other above and below. Thus the fundamental plan of the 

 tail of a fish in its soft parts supplies the machinery necessary 

 to elaborate the hard parts ; and from their less bulk and the 

 greater relative power brought to bear upon them, it would 

 seem not improbable that the neural and haemal arches should 

 be ossified at an earlier period in the history of the organiza- 

 tion than the centrum. And this muscular power would be 

 competent, if the arches long remained separate from the cen- 

 trum, to draw them towards each other, so that the dorsal part 

 of every neural arch would abut against the dorsal part of the 

 arch next behind it. Thus there will come to be formed in- 

 terlocking facets between the arches, of which the anterior 

 will look upward while the posterior will look downward : in 

 most animals the neural arches actually have such • facets, 

 which are known as anterior and posterior zygapophyses. 



