5 1 2 PISCES- -FISHES. 



be recognised. The large muscles which work the jaws are 

 noteworthy. Professor Cossar Ewart has described a 

 rudimentary electric organ in the tail region of Raja batis 

 and R. clavata, apparently too incipient to be of any use. 



Electric organs are best developed in two Teleostean fishes a S. 

 American eel (Gymnotus) and an African Siluroid (Malapterurus)., and 

 in the P'lasmobranch Torpedo. In Gymnotus they lie ventrally along 

 the tail, in Malapterurus they extend as a sheath around the body, and 

 in Torpedo they lie on each side of the head, between the gills and the 

 anterior part of the pectoral fin. In other cases where they are slightly 

 developed (certain Elasmobranchs and Teleosteans), they lie in the tail. 

 Separated from one another by connective tissue partitions, are numerous 

 " electric plates," which consist of strangely modified muscle substance 

 and numerous nerve endings. The electric discharge is very distinct in 

 the three forms noted above, and is controlled in some measure at least 

 by the animal. 



The skeleton.- -The skeleton is for the most part cartila- 

 ginous, but here and there ossification has begun, as a 

 crust over many parts, more deeply in the vertebrae, teeth, 

 and scales. 



The vertebral column consists of an anterior plate not 

 divided into vertebrae, and of a posterior series of distinct 

 vertebrae. Each of these has a biconcave or amphicoelous 

 centrum. From each side of the centrum a transverse 

 process projects outwards, and bears a minute hint of a 

 rib. From the dorsal surface of each centrum two neural 

 processes arise, and arch upwards for a short distance on 

 each side of the spinal cord. Between each two vertebrae 

 there is at each side a broad interneural plate, which not 

 only fills what would be a gap between the neural processes 

 and the slightly developed neural spine, but also links the 

 vertebrae together, so that on surface view the segmentation 

 of the vertebral column is far from obvious. In the caudal 

 vertebrae, what seem to be the transverse processes are 

 directed downwards, to form a haemal arch enclosing the 

 caudal artery and vein. In the lozenge-shaped spaces 

 between the vertebrae lie gelatinous remains of the noto- 

 chord. The vertebral column develops, as usual, from 

 the mesodermic sheath of the endodermic notochord. 



The skull is a cartilaginous case, with a spacious cavity 

 for the brain, a large posterior aperture or foramen magnum 

 through which the spinal cord passes, two condyles working 



