TRIASSIC FISHES FROM SPITZBERGEN 
225 
The ectopterygoid ( Ecpt , text fig. 72, PI. 3 i, figs. 1, 3 ; PI. 32 , fig. 5) extends along 
the lateral border of the palatoquadrate right to the quadrate, as in Boreosomus arcticus , 
but it is considerably higher than in this latter species. 
The entopterygoid ( Enpt , text fig. 72; PI. 3 i, figs. 1, 3 ; PI. 32 , fig. 5) is in its anterior 
part of about the same height as the ectopterygoid, becoming rapidly higher posteriorly. 
Its posterior part, which covers the medial surface of the whole metapterygoid and that 
of almost the whole quadrate, shows, however, very curions conditions, at it is subdi¬ 
vided into a number of oval or rounded tooth-bearing plates. The transition between 
this posterior part and the anterior part takes place gradually. 
It seems at first sight rather natural that in this condition of the entopterygoid 
we should see a good proof of the theory of the phylogenetic origin of certain membrane 
bones through the concrescence of teeth 
(cf. Gaupp, 1905, pp. 61 3 —615), i. e. that 
we . should have the posterior part of the 
entopterygoid at a primitive phylogenetic 
stage of development here. Close reflection 
will show us, however, that an opposite 
interpretation of the facts in question is 
possible and even more probable than 
the first-mentioned one. 
As we now known(cf.pp. 126,134,146) 
a larg-e, high entopterygoid, reaching in a 
caudal direction right to the posterior 
. . Text fig. 72. Acrorhabdus bertili n. sp. 
margin of the palatoquadrate, is developed Parts of the viscera i ske i e ton. After p. , 74 . % 
prop, . 
Rhipidistids. It has also been possible to byal; Ecpt, ectopterygoid; Enpt, entopterygoid; Hm, hyo- 
mandibular; Mco f mixicoronoid; Qu y quadrate; Sthy? possibly 
show in this work that the entopterygoid stylohyal; Sympl, sympletic; ohy, indication of ossification in the 
posterior end of the ceratohyal; pr.op, processus opercularis. 
component of the pterygoid of the Coela- 
canthids is large and high and has about the same extension as in Rhidipistids and 
Stegocephalians and we have found that a rather Rhipidistid-like entopterygoid was 
developed in certain Paleoniscids (Boreosomus) too. 
The facts mentioned support, in my opinion, the view that a Rhipidistid-like entopterygoid 
is a very old formation, which was not acquired independently by primitive Tetrapods, Crosso- 
pterygians and Actinopterygiansbut was inherited from an ancestral form common to them all. 
The strong - development of the entopterygoid in Boreosomus would thus be primitive, 
if this argument is correct, while the conditions found in Acrorhabdus bertili would indicate 
the beginning of a reduction process, whose continuation would lead to total disappear¬ 
ance of the caudal, larger part of the bone. The remaining anterior part would then 
correspond to the bone that is called the entopterygoid in recent higher ganoids and 
Teleosts, and Acrorhabdus bertili would thus, with regard to the entopteryg oid, have 
entered upon a direction of development leading to the Teleostean type . x ) 
! ) It is. noteworthy in this connection that both in Ami a (van Wijhe, 1882, p. 280) und Polypterus (van Wijhe, 
1882, p. 253; Pollard, 1892, p. 412) we find teeth developed on the medial side of the metapterygoid and quadrate 
Amia ). These teeth indicate that certain remains of the posterior part of the entopterygoid exist, though they were 
fused to the metapterygoid. 
Stensio, Triassic Fishes from Spitzbergen. 29 
