838 DR RAMSAY H. TRAQUAIR ON FOSSIL FISHES COLLECTED BY THE 



furnished with an aculeate spine, first one being directed forwards, the remaining 

 five backwards. . * 



Description. — The specimen represented in PL V. fig. 1, enlarged by one-half, 

 is three and a half inches in length, and is the largest complete one in the collection, 

 though fragments occur indicating a somewhat larger size. This example does not 

 show the contour of the head properly, but that deficiency is supplied in fig. 2, in 

 which the rounded blunt form of that part is seen in its entirety. Fig. 3 represents 

 another small specimen, the contour of which is shortened up, a phenomenon which, 

 as well as the converse condition of lengthening out, is frequently observed in 

 palaeozoic fishes which had no ossified vertebral column to keep them in shape 

 during fossilisatiOn. 



The configuration of the fish and the arrangement of its dermal scutes may best be 

 understood by a reference to the accompanying restored outline, the contour of which 

 is based on that of the specimen represented in PI. V. fig. 1, the details being 

 however, completed by an examination of numerous other examples. 



o» 



Fig. 3. — Restored outline of Birkenia elegans, Traq., one-half larger than natural size; d, dorsal fin. 



From the elegantly fusiform shape of the body and the completely heterocercal 

 deeply cleft inequilobate caudal fin, we might at first sight fancy that we had before 

 us a member of the Pakeoniscid family, with the rows of scales running the wrong 

 way ! The resemblance is, however, entirely superficial. 



The head is bluntly rounded, and covered with small scutes, spindle-shaped in 

 outline, and very peculiarly arranged. On the top of the head they are disposed in 

 four areas separated by two cross lines, one longitudinal-median and the other 

 transverse — the scutes of each area having their long axes parallel with each other, 

 but directed at acute angles to those of its fellow of the opposite side, and also of the 

 area immediately in front. Of course, in the figure only the two areas, anterior and 

 posterior, of the right side can be seen. Then in front, just behind the rounded 

 snout, the little scutes swirl round a circular space nearly where we would expect the 

 orbit of a pakeoniscid fish to be, but the area of this space is occupied by scutes like 

 those of the rest of the head, but arranged with their long axes vertical. Below and 

 behind this space is another rounded marking, around which the adjacent scutes also 

 pass in a swirling manner, but what organ this can represent it is meanwhile impossible 

 to guess. It is on the wrong aspect of the head for an orbit, — it cannot be a mouth, 

 because it is paired, having a fellow on the opposite side. It is not even certain that 



