TRIASSIC FISHES FROM SPITSBERGEN 
2 5 3 
Of the pectoral fins — presumably rather well developed — only a few lepidotrichia 
and endoskeletal cordials are present; as will be seen, the lepidotrichia are unjointed, at 
least in their proximal and middle thirds. It is -also notworthy in this connection that 
the pectoral fins possibly had a rather high position. 
The scales are partly poorly preserved, partly weathered, so that nothing certain 
can be discovered as to their shape. Along the anterior margin, however, their medial 
surface had a relatively robust ridge. The sculpture consists of dense striae running as 
usual downwards and somewhat forwards. The striae are, however, thicker, fewer, and 
the whole considerably straig'hter than in Platysomns nathorsti, to which it may be added 
that none of them has been found parallel to the lower or upper margin of the scales 
as in this species. 
Remarks. — The specimen P. 140 just described is clearly distinguished from 
Platysonins nathorsti by its scale sculpture. It presumably represents a new species. But 
until more material is available I have only placed it under the heading Platysomns sp. 
The species seems to differ from Platysomus in the high position of the pectoral 
fins, and by the development of the lepidotrichia of these fins, and because of this it 
may probably with better preserved material turn out to be a representative of a new 
genus. With regard to the high position of the pectorals the fish resembles to some 
extent Ecrinesomus. 
Geological occurrence and locality. —- Fish horizon, Mt Trident. 
Family Catopteridae. 
Historical remarks. 
The family Catopteridae was established by Woodward 1890 (1890 a, pp. 15 —16) 
for the Triassic genera Catopterus J. H. Redfield and Dictyopyge Egerton. While these 
genera were at that time generally placed together with Acentrophorus and Semionotns 
among the so-called «Lepidosteids» (cf. Traquair 1877 b, p. 565—567; Zittel 1887, p. 2o3; 
Newberry 1888, pp. 50—65; etc.), Woodward was now able to show convincingly that 
their characters, although incompletely known, indicated instead a close relationship to 
the Palaeoniscids. For these reasons Woodward maintained that the family Catopteridae 
ought to «be assigned to whatever great subdivision of the «Ganoids» is made to 
include the Palaeoniscids and their allies*. In 1895 (1895 b, pp. V—VIII and pp. 1 — 9) 
he grouped the families Palaeoniscidae, Platysomidae, Catopteridae, Acipenseridae and Poly- 
ontidae in the sub-order Chondrostei. 
In 1901 Schellwien (pp. 17—28) put forward the view that the species described 
as Semionotns letticns O. Fraas and Semionotns altolepis Deecke really showed a number 
of peculiarities that were incompatible not only with the genus Semionotns but also 
with the family Semionotidae and the Protospondyli in general. Schellwien also expressed 
the view that the two species most probably ought to be referred to the Catopterids. 
In accordance with this df. Allessandri (1910, pp. 49—51) in his monograph on Triassic 
