TRIASSIC FISHES FROM SPITZBERGEN 
273 
species (cf. Allis, 1905, Pis. XVIII, XX; Berg, 1911, Pis. IV—VIII); a further support 
for this consists in the fact that neither the Palaeoniscids, Platysomids, Catopterids nor 
Saurichthyids (cf. Part II) show any indication at all of anything- similar. 
For similar reasons I also consider it obvious that the cheek and opercular skeleton 
is also reduced in the sturgeons. We have as a matter oft fact found a fine example 
of a similar phenomenon in Axelia among the Coelacanthids, as we have seen above 
(pp. 98—99), 
The connection between the palatoquadrates anteriorly under the rostrum is, as we 
know, a very remakable character in the sturgeons, while by Bridge (1878, pp. 723, 724), 
and others has been interpreted as a primitive selachoid one. The Palaeoniscids and the 
closely allied forms show, on the contrary, the ordinary conditions, the palatoquadrates 
in them being separated anteriorly and each of them having articulated with the ethmoidal 
region by means of an autopalatine or perhaps sometimes by cartilage; this is also the 
case with the Saurichthyids, which, as we shall see, have in several respects sturgeon¬ 
like characters. It thus seems very probable that we are concerned here with a secondary 
character, a selachoid adaptation, in the sturgeons. 
The absence of premaxillaries in the sturgeons must of course also be looked 
upon as secondary, as these bones' are found in Saurichthyids, Palaeoniscids, Catopterids 
and Platysomids. That the premaxillaries could be reduced is seen in other Teleostomes 
as well such as Dipnoi and possibly the Coelacanthids too. 
In the mandible of Acipenser we generally find, as is known, a couple of investigating 
bones. In old specimens, on the other hand, Parker (1873, p.256) says he has found an 
angular, dental, mixicoronoid («splenial») and mentomandibular. In the Liassic Chondrosteus, 
which can already be considered to be a sturgeon, Traquair (1887, p. 256) states that he 
has observed a dental and two other bones, one of which he takes to be the angular, 
the other as the articular. According to what we can assume, however, the dental both 
of Acipenser and Chondrosteus is a dentalo-splenial and the articular is probably the same 
bone as the one we called the supraangular in the Palaeoniscids above, and other bones 
too, as well as the development of the mandible as a whole, seem to have been Pala- 
eoniscid-like. The evidence seems thus to show that the small number of bone elements 
in the mandible of the sturgeons is secondary and is probably due chiefly to reduction. 
In the Saurichthyids too we find, as we shall see, the same number of bones in the 
mandible as in the Palaeoniscids. 
With regard to the development of the sensory canals the sturgeons certainly 
show a rather great resemblance to Palaeoniscids, Platysomids and Catopterids, but at 
the same time great differences can be observed. Thus in the sturgeons the supraorbital 
sensory canal anastomoses at its posterior end with infraorbital canal, then continuing 
in the normal way forward on to the ethmoidal region, where it runs forwards and 
downwards through the septum between the two nasal apertures of either side. In Pala¬ 
eoniscids and the forms closely related to them the supraorbital canal has no connection 
posteriorly with the infraorbital canal and always extends into the parietal, to which is 
to be added the specially important fact that it never continues along in the ethmoidal 
region medially of the nasal aperture. It extends forward to the upper corner of the 
antorbital instead, where it is connected uninterruptedly with a portion of the infraorbital 
Stensio, Triassic Fishes from Spitzbergen. 35 
