1256 BR. ERIK A. SXENSIO : NOTES 



conclude with the guidance of Bryant's figures where it must 

 have been situated. 



The bone which is held by Bryant to be the postfrontal is 

 clearly seen to consist of two independent bones in both my 

 specimens (text-fig. 4), and I also find there that the posterior 

 one of these two bones, Avhich is undoubtedly a compound 

 supraorbital {so), extends so far backwards as to meet the 

 anterior end of the dermosphenotic. The anterior bone (Ant), 

 on the othei" hand, is situated mainly anterior of the orbit, 

 forming the anterior boundary of the entrance to this, and is 

 certainly to be interpreted as an antorbital bone. In the main, 

 the supraorbital and the antorbital are correctly drawn in my 

 restoration published in 1921 (Stensio, 1921, text-fig. 57; the 

 suture line between the nasal and antorbital has fallen away on 

 the left side of this figure). 



The postrostral elements of the fish have fused to a rather large 

 unpaired plate (Fir, text-fig. 4), as described by Bryant (called 

 by him interfrontal), and my restoration in 1921 (text-fig. 57), 

 is thus incorrect in this respect. 



The paired nasal bone is well seen in one of my specimens. It 

 is situated along the lateral edge of the postrostral plate, 

 extending forwards to the anterior end of this plate, backwards 

 on the other hand somev/ha.t farther (N'a, text-fig, 4). Its length 

 thus exceeds that of the postrostral plate. The breadth is rather 

 small, and constitutes only about one-half of the length. Through- 

 out its length the bone is pierced by the supraorbital sensory 

 canal. My restoration in 1921 thus showed correctly its 

 approximate shape and relations. It is also notewoi'thy in this 

 connection that tlie left nasal hone of the specimen actually seems 

 to consist of two or three independent elements, the conditions 

 thus much resembling those in Dictyonosteus. 



That the shape and extension of the nasal bone in Bryant's, 

 restoration (his text-fig. 2a) must be incorrect is thus evident. 

 That this is the case is also seen from pi. iii. fig. 2 in Bryant's 

 paper in question, where the bone is rather well displayed in 

 ventral aspect and shows a longitudinal rounded ridge caused by 

 the sensory canal in it. 



iinterior to the postorbital plate and the nasals, I find in the 

 smallest of my specimens (the one that has the nasals well 

 displayed) fragments of a transverse series of bone-plates, Avhich 

 are the rostrals and probably also the interrostrals, and finally 

 aiiterior to these the rather narrow premaxillaiies. The number 

 of the rostrals and interrostrals cannot be ascertained, and it is 

 also impossible to make out anything as to the exact shape either 

 of any of these bones or of the premaxillaries. A rostral plate 

 is also observed by Bryant (p. 9). That he has not seen more 

 of these plates is easy to explain, as they are very difficult to 

 observe and in larger specimens have perhaps fused more or 

 less intimately with the premaxillaries or other neighbouring- 

 bones. 



