PHYOIS YARRELLII. 45 



novelty, to dedicate so elegant and interesting a species to the excellent 

 and able author of the " British Fishes." 



Compared with the common " Abrotea" of Madeira (P. Medlterraiiciis Lar.) 

 it is not only a much smaller fish, but is considerably shallower in proportion to 

 its length ; the greatest depth, which is from the beginning of the dorsal to that 

 of the anal fin, being only one fifth part, instead of one fourth, of the whole length. 

 The length of the head equals or is very little less than the greatest depth. The 

 greatest thickness, which is on the head and shoulders, or from the eye to the root 

 of the pectoral fins, equals only half the depth : towards the tail the body becomes 

 much more attenuated, slender, and compressed. 



Although the mouth and gape are very large, the head is small, considerably 

 depressed below the dorsal line, with a remarkably strong hollow or depression on 

 the nape ; rising again, however, into a protuberance over the extremely large eye, 

 which in diameter is one third the length of the head, and is surrounded by a 

 faintly-defined bony ring or ridge. Muzzle prominent : the upper jaw a little 

 longer than the lower, which is furnished with a single short beard or barbule at 

 the tip beneath. Teeth scobinate, or brush-like in both jaws, small and recurved : 

 the outer row in the upper longer and more pointed than the inner, which are 

 blunt and short ; none exist upon the ethmoid or vomer. The tongue is short and 

 broad ; white at the tip ; the rest of the mouth inside and gullet being dark. The 

 whole body is covered with small and inconspicuous scales, which easily rub off, 

 leaving it naked and smooth. 



The lateral line is straight and channeled, appearing like an inlaid narrow sort 

 of chain. Its scales are slightly notched or cloven at the tip, with the appearance 

 when in situ of a little tubercle or wart rising through the notch at about every 

 fourth scale. 



The first dorsal fin is placed very forward, nearly on the nape, just behind its 

 depression. Its height equals or rather exceeds that of the body, and is double 

 that of the second dorsal fin immediately behind it. It is composed of five soft 

 rays ; the first of which is the longest, and simple : the foiu" others are gradually 

 shorter and forked ; the last is connected to the body behind by a web. Its 

 breadth or rather length at the base equals the diameter of the eye. 



The second dorsal fin commences close behind the first, in a vertical line with 

 the upper or anterior axil of the pectoral fins, and extends continuously nearly to 

 the origin of the caudal fin. Its edge is even, but it is slightly depressed or nar- 

 rower at about two thirds its length from its origin than forwarder ; and its hinder 

 end again is broader, ending abruptly in a point or angle caused by the production 

 of the hinder rays, and by the sudden diminution in length of the five or six last 

 of all. The front or fore part is also very slightly higher than the rest. 



The anal fin begins opposite the twelfth or fifteenth ray of the second dorsal, 

 ending a little before the termination of the same, and resembling it precisely in 

 shape, but having its peculiarities exhibited in a much stronger manner. It is 

 much higher in front, more depressed in the middle, and with the hinder end more 

 distinctly angular and pointed. Its first ray is very short ; the following in- 

 creasing gradually to the sixth or seventh. 



The pectoral fins are narrow, lanceolate, acuminate, placed about one third up 

 the side. They are about the length of the ventral fins, and their tips reach to 

 the origin of the anal fin. 



The ventral fins are placed close together quite underneath the throat, con- 

 siderably in advance of the pectoral fins. They consist each of a single flexible 

 forked ray or filament, without web or membrane, rather shorter than the head, 

 and reaching only to the vent, or not quite to the origin of the anal fin. 



