TH15 DliiRMAL l<lN-ItAYS OF FISHES. 493 



scales similai- to those on the body cover the caudal fiu ; 

 transverse sections might possibly reveal underlying cera- 

 totrichia. 



Amongst the Ichthyotomi, Xenacauthus is the only genus 

 in which dermal rays have been described. They appear to 

 be typical ceratotrichia, differing only from those of Selachians 

 in that they were perhaps grouped together near the endo- 

 skeletal rays in the paired hns (Fritsch [10]). 



No very definite dermal rays have been described in the 

 Ostracodermi. Most authors seem to agree that these 

 aberrant fish had median fins provided with rows of small 

 scales resembling those on the body in histological structure. 

 Writing of the Asterolepidaa, Traquair says, "No distinct 

 'rays^ can be seen '^ (37). In his recent description of 

 Bi rkenia elegans, the same author describes the caudal fiu 

 thus:— "Along the dorsal margin we have a special band of 

 tiny, naiTow, oblique scales, which would remind us of the 

 ridge-scales or ' fulcra ' of the upper lobe of a palceoniscid 

 tail, were it not that the band here apparently consists of two 

 rows of scales one above the other. The fin-membrane is 

 also covered with narrow scales, which tend to be arranged 

 linearly so as to give the fin a very decidedly rayed appear- 

 ance. All the scales of the tail and of the caudal fin show 

 the same minutely tuberculated ornamentation which occurs 

 on the scutes of the body" (38). A. Smith Woodward, 

 describing Cephalaspis Murchisoni, Eg., says, ''There 

 are no well-defined fin-rays, the supporting structures being 

 small, oblong, calcified plates, closely fitted together, and 

 placed end to end in vertical parallel series" (42). My own 

 observations on the Cephalaspidte, especially on a specimen 

 in the Oxford Museum, lead me to the conclusion that the 

 fin-rays of Cephalaspis, at all events, were well-defined and 

 typical lepidotrichia as regards their shape and disposition. 

 The limits of the individual segments in this specimen are 

 indistinct; but they are set end to end to form definite parallel 

 rays of considerable width, and set very close together, — very 

 much as in Acipenser, for instance. The presence of a very 



