AZOTES ON CERTAIN CROSSOPTEEYGIANS. 1241 



55. Notes oil certain Crossopterygiaiis. 

 By Erik A : son Stensio_, Upsala.* 



[Received September 18, 1922 : Read November 21, 1922. J 

 (Plate I.t ; Text-figures 1-6.) 



Introduction. 



In the summer of 1921, just as my paper on the Devonian 

 Coelacanthids from Wildungen (Stensio, 1922 a) had been sent 

 to press, there arrived at the XJpsaJa Library an interesting 

 description of Ettsthenopteron, published by Bryant already in 

 1919, which accordingly could not be dealt Avith in my paper just 

 referred to. A few naonths later there also appeared a veiy 

 important paper by Watson on the organisation of the Cosla- 

 canthids (Watson, 1921). As many of the facts put forward by 

 Bryant and Watson in these papers throw new light on many 

 questions, I have found it of interest to give here some of the 

 views at which I have arrived, partly on the basis of these facts, 

 partly by examining my own material of Crossopterygians. I shall 

 begin with a detailed description of what is known so far of the 

 neurocranium in Dictyonosteus, then turning to Eustheno])teron, 

 Porolepis, and the Coalacanthids. 



Dictyonosteus arcticus Stensio. 



(PI. I. figs. 1-4.) 



The material of this fish to be described below has already 

 been dealt with by me in a preliminary description in 1918 

 (Stensio, 1918 6), and has further been referred to in my memoirs 

 on the Triassic fishes from Spitzbergen and in my paper on the 

 Ccelacanthids from Wildungen (Stensio, 1921; 1922 5, 1922 «, 

 pp. 196-198, 204-206). It consists only of an anterior pai't of 

 the primordial neurocranium with the dermal bones partly 

 preserved. The part is large, measuring 21 cm. in length and 

 indicating that the head of the fish to which it belonged must 

 have attained a very considerable size. 



Prirjiorclial Neicrocranium. 



The occipital region is totally lacking, as is also the labyrinth 

 region with the exception of a very short anterior part. The 

 remaining two regions, the orbitotemporal and ethmoidal regions 

 respectively, are as a whole rather completely preserved, though 

 partly crushed and broken. 



The whole orbitotemporal region and the anterior part of the 



* Communicated by Prof. D. M. S. Watsoij, F.R.S., P.Z.S. 

 t For explanation of the Plate, see p. 1271. 



