64 GALAXIAS FROM MOUNT KOSCIUSKO, 



which group truttaceus may be taken as typical, the other by a 

 long, slender body, to which atlenuatus and its allies are to be 

 referred; yet in this one small species I am confronted with 

 individuals varying from one-fifth to one-eighth in the propor- 

 tionate measurement of depth to length, and with a corresponding 

 difference in colour from a dull dark brown without or with but 

 very slight indications of markings to bright golden beautifully 

 blotched, spotted, or barred with black. These differences, how- 

 ever, great as they appear to a casual glance, are entirely 

 attributable to the nature of the locality and the water which 

 the individual fish inhabits, the stout, sombre-coloured form being 

 found in the deep still pools and small subalpine tarns, the slender 

 brilliant one in the more rapid gravelly or sandy shallows where it 

 is exposed to the sunlight; but between these twolimital forms every 

 conceivable variation, both of contour and colour, may be found. 

 The distribution of GalaxUts, comprising as it does the southern 

 extremities of the three great continental areas which converge 

 upon the Antarctic Circle, is unique among fishes, though the 

 Marsipobranchians. of the genera Geotria and Caragola and 

 the recent members of the clupeoid genus Diplomijsius* somewhat 



* The genus Diplomt/itus was instituted by Prof. Cope (Bull. U.S. Gaol. 

 Survey Terr. 1877, p. 808) for the accommodation of certain fossil forms of 

 Tertiary Clupeids from the Green River portion of the Wasatch Beds, which 

 are situated in the central region of the United States, where it is numerous 

 both in species and individuals. Three recent species are known, two of 

 which — no'-u' hollandue and apraffeUldex—^ttilong to the fauna of south- 

 eastern Australia, and the third (Clupea nofarcDifhiis, Gi'inth. )jto that of Chile. 

 Not being aware of its earlier severance by Cope, I proposed (Records Austr. 

 Mus. ii. p. 24, 1892) to separate, under tjie name Hi/])erlophm, all those 

 Herrings in which a predorsal serrature was present, but, my attention being 

 kinilly drawn thereto by Dr. Smith Woodward, I used Cope's name for 

 Valenciennes' Meletta novce-hollandiai in a subsequent work (Edible Fish, 

 and Crust. N.S Wales, p. 184, 1893). At present, however, lam uncertain 

 ■whether Diplomy>itus can properly be retained for these forms, as Dr. 

 Eigenmann in 1891 diagnosed the family Diplomystidce—oi which presum- 

 aljly the central genus is Diplomyslm — for certain South American Nema- 

 tognaths, and I have not as yet been able to learn the date of this genus; 

 if, however, it is prior to Cope's the clupeoid fishes must take the name 

 Hyperlophus. 



