Variations of Stereolepis gigas. 491 



not one of importance." Tliis latest description, accompanied 

 by a figure, is taken from a sj)ecinien 14 inciies long, although 

 the species is known to grow to five times that length. Had 

 Dr. Jordan not overlooked my description and figure in 1807, 

 as he courteously informs me he inadvertently ditl, he coukl 

 not have stated that the spinous dorsal fin is appreciably more 

 elevated in the adult Megaperca than in the adult Stereolepis 

 (young specimens of the latter are, I believe, still unknown). 

 It seems hardly credible that so experienced an ichthyologist 

 as Dr. Jordan should overlook the enormous changes in the 

 comparative depth of the spinous dorsal v/hich take place 

 with age in all Bass-like fishes, and in order to emphasize this 

 point I here give outline-figures (a) of a small (1 foot long) 

 and {b) a large (4 feet long) Nile Perch [Lates niloticus). 



I have specially selected the Nile Perch as an example, 

 because, having been able to study a large number of 

 specimens, I have no fear of having confounded two species. 



In the young Megaperca the longest spines measure about 

 half the depth of the body, in the adult (photograph of the 

 type) exactly one fourtli. I therefore cannot accept the 

 differences in the dorsal spines as being due to anything more 

 tlian the usual changes which take place with age, ami until 

 the young of the Californian fish is known we may safely 

 assume that its first dorsal is much more elevated than in the 

 adult. In a letter addressed to me a short time ago Dr. Jordan 

 adds that the Japanese fish has " much larger scales." I have, 

 I think, disposed of the supposed distinction in the dorsal fin; 

 I will now give some facts against the second distinctive 

 character, which, so far as I know, is the only one that would 

 stand after a comparison of the figures given by me in 1897. 

 In 1895 I gave the scale-formula, compiled from different 

 sources (Japanese and American specimens), as 80-100 gl. 

 In the Californian specimen examined by me in 1897 I 

 counted 115 ^-^. In their young Japanese specimen Jordan 

 and Snyder counted 87 ^. In two specimens from Japan 

 (Sagami Bay), now preserved in the British Museum, and 

 measuring 19 and 15 inches respectively, I find 90 ^-^ in the 

 first, 105 ^ in the second, and my counting has been verified 

 by my colleague Mr. Regan. These numbers seem to me 

 to dispose entirely of the alleged difference in the size of the 

 scales as a specific character. 



34* 



