FISIIKH — WAl'I'K. 115 



to be identical with //. fypuH, Sclilegel, from Japan. Originally, 

 the authors wrote : — " But for the difference in the number of 

 the dorsal spines and some other minor variations, I should have 

 been inclined to consider this to l)e the species descriljed in the 

 'Fauna Japonica' as //. acutirostris." 



A reference to the descriptions of tSchlegel's species shows that 

 II. elPAiati(,s is quite distinct, and further suggests that //. actUi- 

 rostris is merely a young form of II. typus. 



From an examination of the good series of //. elevatufi obtained 

 by the expedition, I am able to supplement and amend the 

 original description as below : — 



B. vi. D. V. i. 26-28. A. iii. 14. V. I. 5. P. 14. C. 17. 

 L. lat. 66-68. L. tr. 14-43. 



Length of head 2-54, height of body 1*6 in the length (caudal 

 excluded). Diameter of the eye 3, length of the snout 2, and 

 the interorbital space 4, in the length of the head. 



The snout above is markedly concave, the occiput less so, and 

 bounded above by a very prominent bony boss. The interorbital 

 area is convex, marked by a deep median groove. Above the 

 bony boss the profile rises by a gentle convex curve to the third 

 or fourth dorsal ray, which marks the highest point of the body, 

 thence it suddenly descends to the caudal pedicel. The lower 

 profile from the snout to the first anal ray is almost straight, 

 thence, corresponding to the dorsal, it abruptly gains the caudal 

 pedicel. 



Lower jaw the longer ; cleft of mouth slightly oblique ; the 

 maxilla extends nearly to below the anterior nostril, the width 

 of its distal extremity less than five in the diameter of the eye. The 

 serrations of the preopercle and post temporal are weak, less so at 

 the angle of the former. 



Teeth. — Both jaws are furnished with small conical teeth, set in 

 a very broad band in front, much narrower at the sides. Vomer, 

 palatines and tongue edentulous. 



In the original description the dorsal fin formula is rendered 

 as vii. 24, but a recount of the spines of the type, and a com- 

 parison with perfect examples, shows that this is an error, the 

 first ray having been mistaken for and counted as a spine, 

 although noticed as a slender one. 



The first dorsal spine is very small, and the rest regularl}' 

 increase to the sixth, which is the longest, slightly shorter than 

 the head, being 1-1 therein. The relative length of the spines 

 is 1, 4, 7, 11, 17. The first two rays are produced, nearly 

 equal, as long as the head and body, or two and a half times 



