150 NOTES ON THE REPrvODUCTION OF TELEOSTEAN FISHES 



of embryonic fin is present in front of the true anal, and, in fact, 

 the embryonic marginal liu is still continuous. The tail is in the 

 heterocercal condition, the urochord projecting freely. At 18 mm. the 

 pelvics, with well-developed rays, have united on the ventral surface, 

 the anus having migrated in a posterior direction. The caudal fin 

 is honiocercal. An isolated fragment of the embryonic marginal fin 

 persists between the anus and the anal fin. 



At 22 mm. the fragment of embryonic fin is still present. The 

 anus has nearly, but not quite, reached the limit of its posterior 

 mifrration. Even in the adult condition there is between the anus 

 and the anal fin a greater interval than in most Teleosteans, and I 

 imagine that this may be due to the rather recent suppression of 

 an anterior part of the anal, now represented only by the vestige 

 of the embryonic fin fold. The second dorsal and the anal fins have 

 the adult formula, viz., 1/14, 1/16. The first dorsal is still but little 

 developed. No scales are as yet visible. 



Compared with similar stages of A. Boycri the larva of A. 2')Tesbyter 

 can be distinguished by the smaller eye, and by the greater length 

 in relation to the degree of development. A young A. Boyeri of 

 32*5 mm. exhibits a stumpy fin-ray midway between the first and 

 second dorsal fins, and in front of and behind this ray are a series 

 of tubercles which are evidently of a similar nature.* These structures 

 represent, I imagine, the vestiges of a continuous dorsal fin, and 

 afford support to the supposition that the restriction of the dorsal and 

 anal fins is of comparatively recent date. 



The larval A. presbyter of 22 mm., though presenting the broad 

 features of adult Atherine conformation, is still far from exhibiting 

 the adult pigmentation. The lateral " stole " in particular is very 

 imperfectly represented. Young A. Boyeri of the same size are much 

 more advanced in this respect as in others. 



I suppose that sand smelts, on account of the robust character of 

 the larva, and its capability of assimilating comparatively large 

 organisms, could be artificially reared with much less trouble than 

 most other marine food-fishes, but their economical value is hardly 

 sufficient to encourage the attempt. The larval stages appear to me 

 to be chiefly interesting from the taxonomic point of view. It is 

 generally conceded that the Atherines and the Grey Mullets are closely 

 allied, yet in their ontogeny they differ most widely. The eggs of 

 the former are, as we have seen, large, demersal, and furnished with 

 long attachment processes. Some, at least, of the Grey Mullets have 



* Vestigial fin-rays have been observed in the larvoe of another fish ; but I cannot 

 recall either the species or the name of the observer. 



