272 
ERIK A : SON STENSIO 
support the view that in this respect the sturgeons are regressively specialized, as 
Woodward especially (loc. cit.) and after him Goodrich (1909, p. 3 1.6) have maintained in 
opposition to Bridge (1878, pp. 717—731) and others, who saw in this a primitive cha¬ 
racter instead. Thus, as we have seen above, the primordial skeleton is very much ossi¬ 
fied even in the Rhipidistids, Coelacanthids and Palaeoniscids, and below (Part II) we 
shall see how this is even more true in the case of the Saurichthyids, which are rather 
closely related to the sturgeons. In primitive fossil Dipnoi, according to Traquair 
(1878, p. 5), at least the primordial neurocranium is strongly ossified, while in the recent 
forms it is cartilaginous practically throughout. Even in certain Teleosts (Gierse, 1904, 
p. 605—615; Handrick, 1901, p. 3 —5) the ossifications in the primordial skeleton may be 
reduced, and we thus see that similar reductions may occur in the most different forms 
among the Teleostomes. An additional support for the view of Woodward and Goodrich 
seems to me to be present in the fact that several ossifications actually exist in the 
primordial neurocranium of the sturgeons (Bridge, 1878, p. 698; Parker, 1882 a, pp. 175, 
176) although they have lost their importance and therefore develop so lat that they 
are as a rule only found in old specimens. 
As far as can be judged, however, these ossifications seem, according to Bridge’s 
(1878) and Parker’s (1882 a) account, to show conditions that are on the whole rather 
Teleostean-like, wdiile according to what we have seen in this work, the Palaeoniscids 
appear in this respect to be developed more in agreement with Polypterids and Crosso- 
pterygians. It is impossible to decide with certainty the position of the Saurichthyids in 
this respect, as the primordial neurocranium in them — at least in fully-grown specimens 
— is ossified in a single piece. 
As we have seen from my account above, certain Palaeoniscids, e. g. Birgeria, are 
rather specialized in the direction of Amia and the Teleosts with regard to the orbito¬ 
temporal region. In Birgeria the ventral part of the interorbital wall have thus become thin 
and at one place fenestrated, and there is also a myodome, though in a very primitive 
stage of development. To this it may also be added that in this fish the distance be¬ 
tween the foramen opticum and the canalis transversus is much shortened and that the 
fossa hypophyseos is very small, only forming a narrow vertical pit in the basis cranii. 
I do not, however, consider it probable that all the Palaoniscids showed the same con¬ 
ditions, but that a number of them had a thick interorbital wall and had the orbito¬ 
temporal region developed on the whole as in the recent sturgeons. 
The Palaeoniscids and even certain of the Catopterids have, as we have seen, indi¬ 
cations of a rostrum, and we thus find conditions here that, as it were, predisposed to 
the development of the frequently very long rostrum of the sturgeons. It is noteworthy 
in this connection that in the Saurichthyids the lengthening process affected not only 
the rostrum but the mandibula as well. 
In the sturgeons the parasphenoid is developed almost as in the higher ganoids, 
while at least in certain Palaenoiscids, as we have found, it shows a very curious 
specialization. 
In the sturgeons the rostrum is covered, as we know, by a rather large number 
of bone-plates, and that we are not concerned here with primary conditions, but with 
secondary ones, is already evident from the differences present in this respect in different 
