50 



DE. W. &. EIDEWOOD ON THE 



vertebra of the series differs from the second only in that its 

 crural plate is still broader, and that the intercrural plate behind, 

 which should lie over the hinder part of the centrum, has been 

 pushed entirely off, and the third neural spine of this vertebra 

 («s') has suffered a similar backward displacement. The centra 

 4 and 5 are the first to show the doubling. They are shorter 

 than the preceding three, but are longer than the half of each of 

 these. The second of the two centra (5) has over it a crural 

 plate and the anterior half of the following intercrural, and 



Fig. 1. — Seylliuin catulus ; caudal vertebrEe, left side, natural size. 

 Fig. 2, — Scyllium catulus; transitional vertebrae, in the region of tlie cloaca, 

 immediately below the first dorsal fin ; left side, natural size. 

 c, centrum ; cr, crural plate ; ic, intercrural plate ; ns, ns', neural 

 spines ; 7^, haemal arch ; hs, htemal spine ; r, last rib ; d, foramen for 

 dorsal root of spinal nerve ; v, foramen for ventral root. 



The dotted lines indicate the hypothetical limits of the perfect vertebrse. 



above these are a half and a whole neural spine. The anterior 

 centrum (4) has belonging to it a crural plate and the posterior 

 half of the displaced intercrural in front, and a whole and a half 

 neural spine. That only one half of the intercrural belongs to 

 this vertebra is clear from the fact, that if in figure 2 the line be 

 erased which separates the centra 4 and 5 and their corresponding 

 crural plates, the now single vertebra will be a facsimile of those 

 marked 2 and 3, except for the fact that there are two hgemal 

 processes instead of one, 



