FISHES. — MCCULLOCH. 115 



The spines and rays of the dorsal, anal, and ventral fins 

 are all more or less rough with minute spinules, while each 

 of the spines is also cut into numerous grooves. The dorsal 

 spines increase uniformly in length backwards ; the first is 

 placed a little behind the verticle of the hinder orbital margin. 

 The anterior rays are the highest, the others decreasing rapidly 

 backwards ; the base of the soft dorsal is little more than half as 

 long as that of the spinous portion. Anal originating below the 

 last two dorsal spines, and terminating behind the verticle of 

 the last ray ; its form is similar to that of the dorsal. Pectorals 

 short, the rays slightly roughened at their bases ; the upper 

 ones the longest. Ventrals greatly produced, reaching to the 

 base of the fifth anal ray ; their margins are rounded, and the 

 last rays are wholly comiected with the body by membrane. 

 Caudal rays smooth, the margin of the fin slightly rounded. 



Colour. — Brassy-coloured in formalin, clouded with some 

 grey blotches on the upper half of the body. Membrane of 

 spinous dorsal and anal dark, becoming black between the 

 ends of the anterior rays. Ventrals black, the rays lighter. 

 Margin of caudal dark grey. 



Described and figured from a specimen 197 mm. long ; a 

 second example, 173 mm. long, does not differ from the type. 



P. velahundus differs from all other species of the genus in 

 being not much less than half as high as long. In this and 

 other characters it resembles Pseudopriacanthus, Bleeker, but 

 differs from that genus in having the posterior dorsal spines 

 longer than the median ones, and a well-developed pre- 

 opercular spine. 



Locs. — -Off Burrewarra Point, near Bateman's Bay, New 

 South Wales, 60 fathoms ; 29th May, 1914. 



Between Cape Naturaliste and Geraldton, Western Aus- 

 tralia, 20-100 fathoms. 



Family APOGONICHTHYID.^. 

 Genus Amia, Gronovius. 



Amia fasciata, Shaw. 



The several colour forms of Amia fasciata named novem- 

 fasciata, compressa, aroubiensis, and robusta. can only be 

 separated from each other with difficulty, and they are per- 

 haps best regarded as subspecies ; they cannot be regarded 

 as varieties of A. fasciata, as has been done by Giinther, be- 

 cause the typical form of that species varies in itself, and its 

 variations do not overlap the characters of the supposed 

 subspecies. I have examined a representative series of some 



