72 GALAX/AS KR(J.M MOUNT KOSCIUSKO, 



'jcdaxiaa nebidosa, Maoleay, is probabh^ the same as G. scribe, 

 Val. The variation in the number of the dorsal and anal rays 

 cannot Vje considered of any value in this genus if the small unar- 

 ticulated anterior rays be included, the number of these being 

 extremely inconstant; there is no other character of sufficient 

 consequence to warrant tlieir separation except the size of the 

 eye, which is stated by A^alenciennes to measure " two-fifths of 

 the length of the head,'' a proportion which is quite unknown 

 among the members of the genus, and is very unlikely to be 

 correct. G. rostratus, Klunz., should also be compared with 

 scriba. 



Galaxias auratus, Johnston. Through the courtesy of Mr. 

 Alexander Morton of the Tasmanian Museum, I have had an 

 opportunity of examining two fine examples — 225 and 185 milli- 

 meters in length — of the form inhabiting the Great Lakes, Tas- 

 mania, which lie at an altitude of 4000 feet above the sea level. 

 These specimens I believe to be mere varieties of G. triUtacens, 

 modified by their surroundings. 



Galaxias iiulicus, Day. From the first I looked with distrust 

 on the possibility of the occurrence of a species of this genus in 

 Indian waters, and I am, therefore, pleased to find that Dr. Gill 

 not only shares that distrust, but has had the courage to publish 

 his disbelief (Nature, liii. p. 366). Should the species on further 

 examination prove to be a true Galaxias, its pi-esence in the 

 Indian littoral would seriously interfere with the theory of an 

 antarctic origin for these fishes. 



It will be observed that no less than seven species {truttaceas, 

 altennatus, ocellatus, cersiculor, cyUiidricus, ddicatukis, and 

 amoe.nus) are said to be resident in or in the immediate neigh- 

 bourhood of the Yarra, and since the two first are well known to 

 be of wide distribution and variable appearance I must be per- 

 mitted to doubt the specific value of all or most of the residual 

 forms, for none of which have their authors pointed out such 

 distinctive characters as would enable one, from a study of the 

 descriptions alone, to determine their specific value. Too much 



