126 A NEW FAMILY OF AUSTRALIAN FISHES, 



The caudal peduncle in the male is a little deeper than long, in 

 the female a little longer than deep. 



The development of the dorsal, anal, and ventral fins shows 

 similar sexual distinctions; thus, the flexible spines of the first 

 dorsal, the posterior rays of the second dorsal and of the anal, 

 and the outer rays of the ventral fins are prolonged into 

 filaments in the males, while in females and immature males this 

 character is inconspicuous or absent. 



Though not the oldest, this genus is by far the most important 

 of the group, whether as regards its degree of specialization, 

 area of distribution, or number of species. 



Up to the year 1878, when Castelnau first described this genus 

 under the name Ariste-iis, all but one of the authors (Richardson, 

 Giinther, Kner, and Steindachner), who had written on the fishes 

 which are here collected together in one family, had recognised 

 their affinity to the Atherinids, the exception being Dr. Peters; 

 and though Castelnau himself, first in proposing to separate in a 

 distinct family his closely allied genus Zantecla ( = Metanotcenia), 

 -which, as he says, " comes near the AtheriniJce" definitely gives 

 in his adhesion to this view, and two years subsequently endorsed 

 this recognition by proposing to separate from that family his 

 two new genera, Atherinosoma and Neoatlierinn, which he coupled, 

 notwithstanding their manifest differences, as Xeoafherinu/a', he 

 nevertheless, in spite of his acquaintance with two of the genera 

 — Mdanotoenia and Neoatherina — and his acknowledgment of 

 their connection with the true Atherinids, commits the extra, 

 ordinary error of referring Aristeus to the Gobiidce, a family with 

 which it has not the slightest affinity, either in its external or 

 its internal structure; this error is perpetuated by Maclea}^ and 

 others. 



In 1886, in a paper on the fishes obtained by the collectors of 

 the New South Wales Geographical Society's Expedition to New 

 Guinea, I descriljed two ver}^ distinct species from the Strickland 

 River, substituting for Aristeus Peters' name Nemalocentris, this 

 being, so far as I knew at that time, the earliest attempt to 



