444 E. T. NEWTON ON A NEW FISH FROM 



being no indication of a crest such as exists in this region in some 

 species of Portheus. The hinder part of the skull has the sides much 

 depressed, so that the points of the pterotics are on a. much lower 

 level than the epiotics or supraoccipital (fig. 1). The supraoccipital 

 bears a crest which projects backwards, but does not reach so far in 

 this direction as do the points of the epiotics. Anteriorly the bone 

 is broken, so that one cannot be sure whether it joined the frontals 

 or not ; and the small triangular space seen in front of the crest 

 (fig. 2) is due to this breakage. The frontals (fr) are two large 

 bones, separated by a distinct suture, and forming a considerable 

 proportion of the upper surface of the skull ; the right one joins the 

 pterotic (pt.ot) and possibly also the supraoccipital (s.o) ; but the 

 boundaries of the bones cannot be traced : the left one has the pos- 

 terior margin broken away in such a manner as to leave a space 

 between it and the pterotic, and thus displays what appears to be 

 another bone lying beneath, which perhaps may be a parietal. I 

 cannot trace any definite division between the last-named bone and 

 the sphenotic (sp.ot); but this is not surprising, considering the 

 manner in which other sutures of the skull are obliterated. If I 

 understand Prof. Cope's description of the skull of his Saurodontidce 

 (loc. tit. p. 188), he regards the bone I have marked ep.ot as the 

 parietal; but this explanation, for reasons given below, I cannot 

 accept in the present instance. Upon the left side the frontal 

 appears to send down a preorbital process (fig. 1, p. orb) ; but the 

 relations of this process cannot be clearly made out, and it is possible 

 that it may be a process of the prefrontal bone. 



Anterior to the frontals, upon the upper surface of the skull, there 

 are two bones (fig. 2) separated by a median longitudinal suture ; 

 towards the front of these an osseous band passes across at right 

 angles, obliterating the suture. The whole of these bones may form 

 part of the ethmoid only ; or, as seems most probable, the parts behind 

 the transverse band may be the prefrontals; this, unfortunately, 

 cannot be decided definitely, because the relations of the parts 

 beneath cannot be known, being hidden by the matrix. In a side 

 view of this region (fig. 1) one downwardly directed process is seen 

 to be given off at the transverse band, and another a little further 

 back ; the latter is most likely the palatine process of the prefrontal, 

 and the former the process of the ethmoid, which, according to Cope, 

 overlaps the anterior condyle of the maxilla in this genus. The 

 space between these two processes no doubt lodged the olfactory 

 organ. 



Bones of the Auditory Region. — The back of this skull presents the 

 five points commonly found in teleostean skulls. The central point 

 is doubtless formed by the supraoccipital, the inner lateral ones by 

 the epiotics, and the outer ones by the pterotics. Prof. Cope, as already 

 stated (I. c. p. 188), seems to think these inner points in the Saurodon- 

 tidse are the parietals ; but we should naturally expect, if this were 

 so, that the lines indicative of the direction of growth would radiate 

 from some anterior portion of the bone, whereas in this specimen 



