VERTEBRAL COLUMN OF FISHES. 



63 



CL, J» 



Kndo- iiml exo-ske- 

 let.-il elements of a 

 caudal vertebra of 

 a Plaice (I'lcuro- 

 ncctcs). * 



these in the abdominal region ai'e lateral, usually 

 stand out and support ribs ; but in the caudal region 

 they bend down and coalesce at their extremities. 

 The caudal vertebrje of some flat-fishes {Plenronectickc, 

 fig. 20.), the Polypterus and the Mur^nse, would seem 

 to disprove this homology of the haamal arches, since 

 transverse processes from the sides of the body co- 

 exist with them, as they do in the Cetacea. But, if 

 we trace the vertebral modifications throughout the 

 entire column in any of these fishes, we shall find 

 that the ha;mal arches are actually parts of the trans- 

 verse processes ; not independent elements, as in the 

 Cetacea ; but due to a progressive bifurcation : this, 

 in Murcena Helena, for example, begins at the end 

 of the transverse processes of about the twenty-fifth 

 vertebra, the forks diverging as the fissure deepens, 

 until, at about the seventy-third, the lower fork de- 

 scends at a right angle to the upper one (which re- 

 mains to represent the transverse process), and, meeting 

 its fellow, forms the hfemal arch, and supports the 

 antero-posteriorly expanded haemal spine. In the Plaice 

 a small process is given off" from the expanded base 

 of the descending parapophysis of tlie first caudal ver- 

 tebra, which increases in length in the second, rises 

 upon the side of the body in the third, becomes dis- 

 tinct from the parapophysis in the fourth, and gra- 

 dually diminishes to the ninth or tenth caudal vertebra, 

 when it disappears. These false transverse processes 

 never support ribs. 



The atlas may usually be distinguished by some 

 slight modification of the anterior articular end of the 

 body, by the persistent suture of the neural arch, or 

 by the absence or detachment of its pleurapophyses : 

 but none of these characters ai*e constant. Peculiar 

 pi'ocesses ai'c sometimes sent off" from the under part 

 of the centrum : two very long and strong processes 

 from this part are articulated with the basi-occipital 

 in the great Sudis (Arapaima gigas). The second 

 vertebra is never characterised by an odontoid pro- 

 cess ; but the absence of this is not to be accounted 



• Compare this figure with nature, and with the figures of a corruspondinj; ver- 

 tebra in n. pi. 5., and in xxvui. p. 58. The names assigned by GeolFioy St. IJilaire 



