Br, A. Gunther — Tertiary Fishes of Sumatra, 435 



the best preserved of all. The snout lies on the side ; behind the 

 eye the skull is twisted, so that its base is visible. Also the 

 abdominal column of vertebrae shows its ventral aspect, the neural 

 spines being so bent as to lie in the same plane as the parapophyses. 

 The right and left series of parapophyses have become detached 

 from the vertebral column, both lie along the right side of the 

 specimen ; and still more outwards, towards the left, a series of 21 

 dorsal spines has left a somewhat faint impression. The caudal 

 portion, with the fins, lies on its side. 



We see from this specimen that the snout was not much longer 

 than the remainder of the head (IJ inches). Some of the foremost 

 vertebrge were confluent into one piece, equal in length to the three 

 following vertebrae. The abdominal portion consists of 27 vertebrae, 

 with compressed centres, not grooved along their ventral surface. 

 The caudal portion consists of 31 vertebras, much shorter than those 

 of the abdominal portion ; 12 of them are in front, 5 below, and 14 

 behind the dorsal. The soft dorsal consists of 9 or 10, the anal 

 of 8 or 9 rays. The caudal is short, with feeble rays and rounded 

 posterior outlines. The dorsal and anal are highest in front. 



On the left-hand side of the posterior part of the head there is 

 visible a flat, sub-crescentic bone, anteriorly with a projecting arti- 

 cular process, from which a gently curved ridge runs along the 

 whole length of the bone. The surface is finely granulated, the 

 granules being partly arranged in radiating striae. I consider this 

 to be the operculum. 



c. The impression of the anterior half of the body of a specimen 

 smaller than the preceding. 



d. The anterior portion of the specimen, of about the same size as 

 specimen a. 



e. The impression of a fish six inches long. 



This fish is probably the one determined by Prof Geiaitz as 

 Fistularia hcenigii, Agass. (see Geol. Mag. 1875, p. 481) ; but our 

 fish has the central caudal ray not prolonged, and therefore does not 

 belong to Fistularia. 



This is evidently the same fish which has been described and 

 figured by Dr. W. v. d. Marck as Protosyngnatlius sumatrensis. The 

 specimens in the British Museum appear to be in a better state of 

 preservation than those examined by that palaeontologist, and, there- 

 fore, better adapted for the determination of the affinities of this 

 fish. 



Pseudeiitropius verheehii. (PI. XV. Fig. 2.) 



The following description is chiefly taken from two specimens in 

 the collection, of which one, if complete, would have been about 

 nine inches long ; only a very indistinct impression of a part of the 

 head is preserved. From the dorsal fin backwards this specimen, 

 which lies on its side, is in tolerably good condition. The second 

 specimen lies on its back, and is only 6^ inches long. Other 

 larger examples are more fragmentary, and do not assist much iii 

 the examination of this species. 



Nothing can be said of the form of the head, except that it appears 



