234 



a. The serial number of the girdle-piercing nerves; 



b. The number of the post-girdle nerves; 



c. The number of nerves forming the collector ; 



d. The number and position of the nerve canals; 



e. The number of the fin rays; 



f. The number of the whole vertebrae. 



2) Asymmetry occurred in an appreciable number of cases. 



3) Differences occurred in the two sexes on the following points: 

 The position of the girdle is more rostral in the male than in the fe- 

 male. The post-girdle fin innervation area is greater in the male than 

 in the female, owing to the development of the mixipterygium. 



4) The female is, on the whole, more variable than the male. 



5) A well-marked correlation exists between 



a. The position of "the girdle and the number of collector 

 nerves ; 



h. The position of the girdle and the number of post-girdle 

 nerves ; 



c. The position of the girdle and the number of whole 

 vertebrae. 



6) No correlation was found between the number of the fin rays 

 and the number of fin nerves. 



7) At certain stages in ontogeny the number of collector nerves 

 is greater than in the adult. 



8) At certain stages in ontogeny the number of post-girdle nerves 

 is greater than in the adult. The most caudal two or three of these 

 form a posterior collector — a structure which is never found in the 

 adult. 



The facts recorded have been used as criteria between the two 

 rival theories of limb origin with the following results: 



1) To explain the variations on the side-fold excalation theory, it 

 must be assumed that excalation of segments is going on in the collec- 

 tor and pre-collector areas whilst, at the same time, intercalation is 

 taking place in the post-girdle area; or, in other words, that the 

 portion of the vertebral column in front of the girdle is tending to 

 split up into fewer segments, whilst simultaneously that portion behind 

 the girdle is tending to become divided into more segments. Leaving 

 on one side the improbability of two contiguous portions of the verte- 

 bral column undergoing at the same time two opposite processes, an 

 examination of the number of whole vertebrae associated with different 

 positions of the girdle lends practically no support to the view that 

 intercalation is going on in this area. 



