318 Mr. A. S. Woodward 07i some 



these are clearly caudal, with robust, gently arched neural 

 and hgemal spines ; as noted by Davis, there is a much- 

 expanded hypural bone. The remains of tlie paired tins 

 prove til em to have been very small, and the pelvic pair 

 must have been opposed to the anterior part of the dorsal 

 fin. The latter is situated completely within the anterior 

 half of the trunk, and seems to be borne by twelve supports. 

 Its two foremost rays are comparatively small and short ; the 

 third is excessively elongated, as shown in Davis's figure, 

 articulated in its distal two thirds, but not subdivided ; the 

 following rays, which are shorter and both divided and articu- 

 lated distally, rapidly decrease in length. Behind the fin 

 Davis recognizes " a number of fin-rays extending some 

 distance towards the tail" — a deceptive appearance due to 

 the crushing of the imperfectly preserved scales. The anal 

 fin is very small, probably with not more than seven or eight 

 short rays, and separated from the caudal fin by a space 

 about equal to the length of its own base-line. The inequality 

 in length of the lobes of the caudal fin noted by Davis is 

 evidently due to accidental distortion in the fossil. The 

 scales are rather large and quite smooth, and appear to me 

 to be cycloid, without any trace of serrations. 



If this fossil be carefully compared with the specimens from 

 Hakel in the British Museum , rightly labelled " Glupea 

 Bottce, Pictet & Humbert," by the late William Davies, it 

 will be found to agree in every essential particular. Al- 

 though Pictet and Humbert failed to discover the extreme 

 elongation of the third dorsal fin-ray, this character is 

 distinctly shown in some of the specimens just mentioned. 

 The fish, however, does not belong to the genus Glupea, as 

 indicated by the absence of ventral ridge-scutes, while it is 

 excluded even from the family Clupeidje by the structure of 

 the upper jaw. The specimens in the British Museum prove 

 that the rod-like premaxilla forms the complete upper margin 

 of the mouth, excluding the equally slender maxilla. Clupea 

 Bottw, with which Pseudoheryw longispina is included, may 

 thus be regarded as the type of a new genus of Scopelidge, 

 Kematonotus, defined as follows: — Trunk short and robust, 

 and maximum depth at origin of dorsal fin. Mandibular 

 suspeiisoriuni nearly vertical; jaws delicate and teeth minute. 

 Vertebrce about 30 in number, half being caudal ; ribs 

 moderately robust. Paired fins very small, the pelvic pair 

 opposed to the dorsal, which is situated completely within the 

 anterior half of the back, short-based and much elevated, 

 with at least one ray excessively elongated ; anal fin rela- 

 tively small and remote; caudal fin stout but deeply forked. 



