702 DR RAMSAY H. TRAQUAIR S DESCRIPTION OF 



outer surface to the matrix ; often, indeed, the scales themselves are so split, that 

 neither outer nor inner surface, but a section, is visible. 



The specimen shown in fig. 1 equals 14^ inches in length, and 2| inches in 

 breadth opposite the ventral fins. The other measurements are as follows : — 



From tip of snout to opposite the origin of dorsal fin, . . 8 inches 8 lines. 



„ „ of ventral ,, . . 6 inches. 



,, ,, of anal „ . . 9 inches 1 line. 



„ „ of caudal „ . . 12 inches 4 lines. 



Scales. — These are rather small in proportion to the size of the fish, but are 

 much larger along the back and upper part of the sides than on the belly, where 

 they are very minute. The larger scales are rather narrow, being, as Agassiz 

 remarks, much higher than broad. The upper border of each scale projects in a 

 very strong pointed spine, which is received into a corresponding excavation in 

 the deep surface of the scale next above. Below the base of the spine the upper 

 margin of the exposed portion of the scale is seen to be deeply concave, while 

 the lower margin is rounded off into a blunt point, which fits on to the con- 

 cave notch-like upper margin of the exposed portion of the scale next in order 

 below. The whole of the exposed surface is ornamented by very delicate, nearly 

 parallel, wavy ridges, whose general direction is from above downwards towards 

 the lower angle of the scale, sometimes anastomosing with each other on the way. 



The small scales along the body become more nearly equilateral, and those on 

 the continuation of the vertebral column into the upper lobe of the tail-fin are 

 also very small, and of an elongated lozenge shape. The largest scales are, how- 

 ever, those along the upper border of the tail. They are triangular in shape, 

 acutely pointed behind, and notched in front, thus resembling very closely the 

 corresponding scales in Amhlypterus and Eurynotus, but being very much larger 

 in proportion to the other scales. They are placed over each other in an imbri- 

 cating manner, whereas very little imbrication is observable in the scales of the 

 rest of the body, 



Pectoral fins. — The specimen figured does not exhibit the pectoral fins, and I 

 have seen no example of this fin complete to its termination. A fragment in my 

 collection shows the pectoral to have been wide and powerful, the anterior rays 

 strong, and with the transverse articulations farther apart than in the other fins, 

 at least at their commencement. Another fragment shows that the anterior 

 border of the pectoral fin was ornamented by a series of rather small imbricating 

 fulcral scales. 



The ventral fin of the right side is shown in fig. 1, V, its extremity being cut 

 off, like those of the other fins of this specimen, by the edge of the nodule. Its 

 base is wide, measuring in the present specimen one inch across, and the number 

 of rays may be estimated at no less than 35. In the more posterior finer rays 

 bifurcation may be observed, but the stronger anterior ones are cut off before 



