TRIA.SSIC FISHES FROM SPITZBERGEN 267 
P „ rvoodwardi is well separated from P. altolepis by the position of the dorsal fin 
and the development of the cranial sculpture and by its larger size, 
Geological occurrence and localities. — Perleidus woodwardi occurs rather 
commonly in the fish horizon. It seems to be most common in Dickson Land, where no 
less than 10 of the present 14 specimens were collected. The finds are distributed over 
the following localities: Mt Congress, Mt Sture, Mt Tschermak, Mt Rotunda (Student 
Valley), the south shore of Sassen Bay, Sticky Keep and Mt Andersson. 
Some remarks on the genera Catopterus, Dictyopyge, Colobodus 
s. str., Perleidus, Dollopterus and Meridensia. 
Skeleton of the head. — In Catopterus and Dictyopyge the osteology of the head is 
still little known. Eastman (1911, p. 53) mentions, however, in Catopterus an unpaired 
«ethmoid», so situated in relation to the frontals, that in my opinion it corresponds at 
least to the fused nasals and postrostrals, i. e. the membrane bones of the ethmoidal 
region whould show fairly Palaeoniscid-like conditions (cf. p. 242 above). The maxillary 
is said (p. 54) also to have a Palaeoniscid-like shape. Further it may be pointed out 
that the outlines of the head seem to be very Palaeoniscid-like (cf. Woodward, i8goa, 
PI. IV, figs. 5, 6, 7, 8). 
That with regard to the skeleton of the head Colobodus s. str. belonged to the 
Palaeoniscid type was shown by me already in 1916 (Andersson 1916 a, pp. 22— 23 ), and 
in 1920 Stolley (pp. 3 o— 32 , 35— 36 , 38—45) gave further proof of this. As in 1919 
I had an. opportunity to study a part of Stolley’s original material 1 ) and as the Insti¬ 
tution at Upsala possesses a quantity of fine Colobodus material both from German 
Muschelkalk and Alpine Triassic, I am now in a position to supplement rather essentially 
our former knowledge of the membrane bones of the head in the genus Colobodus. 
In the first place it is obvious that a large unpaired nasalo-postrostral (Na. ptr, 
text figs. 85, 86) is present, which probably includes in itself the interrostrals as well. 
The antorbidal (Ant, text figs. 85, 86) is a triangular plate, situated, as usual, laterally 
of the nasalo-postrostral, together with which it forms the boundary of the nasal opening. 
In front of the antorbitals and the nasalo-postrostrals there are three rostrals situated 
in a transversal row ( R, text figs. 85, 86). Otherwise the bones show the conditions seen 
in text figs. 85, 86 and 87. 2 ) 
As Stolley (1920, pp. 45—65) has shown, the head in Dollopterus 3 ) shows similar 
conditions to that of Colobodus. 
*) The private collection of Redacteur Konig of Heidelberg and the collection of the Geological Institute 
of Heidelberg. 
2 ) In ought perhaps also to be pointed out that the parietals both in C. bassanii and C. maximus are 
small and square, and that in both these forms the head skeleton is very similar. It is therefore impossible, at 
least at present, to follow Stolley (1920, p. 44) in separating C. bassanii from the genus Colobodus s. str. (cf. 
p. 254 above). 
3 ) Stolley (1920, pp. 64, 65) points out the agreement between the so-called Semionotus letticus O. Fraas 
and Dollopterus. I have myself had an opportunity during a visit to Stuttgart to make sure that the type specimen 
of this species really seems to belong to Dollopterus. Among other specimens in the Stuttgart Museum labelled 
S. letticus there were also a number that turned out to belong to Gyrolepis. Stolley (1920. p. 65) considered, 
however, that the ougth provisionally to incorporate S. letticus with the genus Perleidus. 
3 4 * 
