242 Dr. R. II. Traquair on the Structure 



that in the Palseoniscidae ; hut the minute articles of the 

 rays are iiner and more scale-like, ami, as M'Coy has aptly 

 expressed it, present " a deceptive resemblance to the scales 

 of the body." This view of the structure of the fins of C/iei- 

 rolepisy however, is denied by Pander, who afiirms that the 

 apparent joints of the fin-rays are in reality nothing but 

 scales which covered internal rays a])parently of a flexible 

 nature ; and such internal non-jointed rays he has actually 

 represented in tab. ix. fig. 2 of liis work. Here I feel my- 

 self compelled to dissent from the opinion of so high an 

 authority as Pander, and to agree with Agassiz and ]\PCoy — 

 as, in spite of the most careful examination of a large number 

 of specimens from various localities, I have never seen any 

 thing like the unartieulated rays represented in his figure, ana, 

 moreover, a ti'ansverse section of a small portion of the lower 

 lobe of the caudal, from a Cromarty nodule (PI. XVII. fig. 6), 

 effectually (to my eyes at least) demonstrates the contrary. 

 Here the whole thickness of the fin is seen to consist of the 

 right and left sets of imbricating demi-rays, no other hard 

 parts being visible. And although it is of course not im- 

 possible that such internal soft rays may have been present, 

 yet the stinicture as here shown exhibits the most complete 

 analogy, or rather identity, Avith that of the anal fin in Pol)/- 

 pterus and CaJamoichtliys^ in which certainly no other rays 

 exist save those whose ganoid, closely jointed, and imbricating 

 surfaces are seen on the outside*. 



The shoulder-girdle must next claim oui- special attention, 

 seeing that one of its elements seems to have escaped the 

 observation of previous writers, save Powrie, and to have 

 been by him completely misinterpreted. Of this the first 

 element, by which the arch was attached to tlie skull, is the 

 jirst supraclavi'cidar, or " suprascapular " (PI. XVII. fig. 3, 

 1st s.cl), a small rounded-triangular plate placed immediately 

 behind the posterior margin of the cranial shield, and distinctly 

 seen only in very few specimens. It is correctly indicated 

 by Pander, in tab. ix. fig. 6 of his work, by the number 46. 

 Articulated with this is the second supraclavicular {2nds.cl), 

 or " scapular," a more elongated plate, broadish above, but 

 getting suddenly narrower about the middle, and whose long 

 axis points obliquely downwards and backwards to articulate 



* Agassiz was nevertheless inclined to believe that in so7ne species of 

 PalaonisctiA (e. g. P. BlainviUei and P. VoUzii) the tin-rays were really 

 covered with scales (Poiss. P'oss. t. ii. pt. 1, p. 43). I do not, however, 

 find this idea coiToborated by the specimens of Palaoniscus BlainviUei 

 in the British Museum, which I have carefully examined ; P. VoUzii I 

 have not seen. 



