BY J. DOUGLAS OGILBY. 129 



lessness which characterises Castehiau's work, may be easily set 

 aside or explained away; the main differences are as follows : — 



(i.) Gill-covers. — Castelnau writes ^ " opercle and preopercle. 

 without teeth or spines, the first with a double edge." This is 

 probably mere carelessness; by substituting " last " for " first " 

 the description would be quite correct. 



ii.) Dentition. — By turning to the foot-note p. 124 my readers 

 will find that I there suggest that certain of the teeth in Rliomba- 

 tractiis ma}^ be deciduous with age, and it is merely necessary to 

 carry this deciduousness a little further to arrive at a dentition 

 somewhat similar to that described by Castelnau. 



(iii.) Fin rays. — " Anal with two spines." I do nut think it 

 necessary to attach much importance to this character, seeing that 

 Castelnau was possessed of but one specimen from which to draw 

 xip his description. It ma}'- be taken for granted that in all these 

 small fresh-water fishes the first soft ray is liable to take the form 

 of an additional spine, and it would, of course, be but natural to 

 describe this genus as having two anal spines if the diagnosis was 

 taken from an example having this individual peculiarity. 



As an instance of this tendency I may mention that when some 

 j'-ears ago a species of Ainhasaia was present in great abundance 

 in the Parramatta and George's Rivers, I noticed that in a 

 number of specimens taken at random almost as many would be 

 found having two rays in front of the second dorsal as those 

 having one, and this increase was always coordinated with a 

 corresponding decrease in the number of soft rays, thus plainly 

 showing that this was not a structural character, but a simple, 

 though common, variation caused by the calcification of the 

 anterior soft ray. 



That Castelnau on the one hand was either unaware of or paid 

 no attention to this tendency to acanthination in fresh-Avater 

 fishes, while on the other hand placing undue prominence on the 

 presence of one or more additional spines, we know from his own 

 writings and from his treatment of JIacquaria australasica, of 

 9 



