496 EDWIN S. GOODRICH. 



general structure of the lepidotrichia lias already been well 

 described by many authors^ among whom one may mention 

 especially Agassiz and Traquair in their admirable mono- 

 graphs (1 and 36). 



The lepidotrichia of Palasoniscus do not differ mucli in 

 their general relations from those of Acipenser. On exa- 

 mining a tail of a Paleeoniscid we are at once struck with the 

 strong resemblance the flattened shiny segments of the rays 

 bear to the scales on the caudal region of the body. Yet 

 even here the transition from the one to the other is not 

 perfect; and may even be abrupt (figs. 52, 53). In the allied 

 genus Amblypterus, however, as was pointed out to me by 

 Dr. Traquair, the transition is so gradual that it is scarcely 

 possible to say where the scales end and the rays begin 

 (fig. 55 A and b). 



An examination of this fish must, I think, dispose of all 

 doubt as to the general homology of the two sets of 

 structures. It need hardly be added that in the PalaBouis- 

 cidte the agreement in the details of histological structure 

 between the lepidotrichia and the scales is as close as in 

 their outward appearance. 



Agassiz believed that, in some species at least, the fins 

 were covered with scales, overlying the true dermal rays — 

 " dans les Palaeoniscus Blainvillei et Voltzii, du moins, il est 

 evident que les divisions transverses que I'on voit sur les 

 nageoires provieunent des series d'ecailles qui en recouvrent 

 les rayons, et qui meme sent placees de maniere a reposer 

 sur les bords avoisinans de deux rayons, . . . car en enlevant 

 soigneusement ces petites plaques, on voit en dessous celles 

 du cote oppose alterner avec I'empreinte des rayons" (1). 

 But Traquair has shown that this view is not correct, aiid 

 that no other separate set of ra3's underlies the scale-like 

 lepidotrichia — a conclusion which is confirmed by my own 

 observations. 



It is to be noticed that in Palseoniscus the proximal seg- 

 ment of each ray is considerably longer than the others, and 

 passes beneath neighbouring body-scales for some distance. 



