152 NOTES ON THE REPRODUCTION OF TELEOSTEAN FISHES 



appears that the former passes through a long-tailed phase, which is 

 not at all represented in the latter. The question is, Has this phase 

 been suppressed in the ontogeny of the Grey Mullet, or has it been 

 evolved as a specialised feature in the phylogeny of the Atherine? since 

 the adult resemblances probably justify us in regarding both as derived 

 from a common stock. I do not think that the knowledge which we 

 at present possess of the systematic relations of individual groups of 

 Teleostean fishes furnishes us with any answer. I believe it is 

 generally held that the forms with elongated abdominal cavities are 

 the more primitive, or it may rather be said that an elongated abdomen 

 is most commonly met with in what appear to be the more primitive 

 members of the Teleostomi. So far as concerns the families with which 

 we are now dealing, the elongated abdomen appears less primitive, since 

 the arrest in this elongation in Atherina would seem, from the evidence 

 of the persistent ventral embryonic fin and of the vestigial dorsal rays, 

 to result from the more recent restriction of the permanent dorsal and 

 anal fins in that genus to the proportions which are now common to 

 the adults of both families. The force of this evidence depends of 

 course on the assumption that a continuously rayed fin-fold is a 

 primitive condition, and is never achieved by a reversion from an 

 intermediate detached-finned condition, 



I am not acquainted with any British larvre of Mugil except those 

 already alluded to ; but a Mediterranean specimen of 14 mm. (figured 

 in my paper in A7in. Mus. Mars., and referred from local considerations 

 to M. auratus), appears to afford some evidence of the relative antiquity 

 of the Atherines and the Grey Mullets. In essential features of 

 conformation it is a true Mugil, but it exhibits a most distinct black 

 "stole" or lateral pigment band. This is a feature of the adult 

 Atherine, but not of the adult of any Grey Mullet with which the 

 specimen can be associated ; nor, as far as I know, of any Mugil at all. 

 The appearance of this "stole" as a transitional larval pigment-phase 

 of Mugil and its retention in Atherina mast be regarded, if of any 

 value as evidence of phylogeny, as indicating that the latter is the 

 more primitive form. The resemblance, however, may be merely 

 superficial, since I cannot say that the pigment stripe of the young 

 Mugil is ever associated with the peculiar characters of the " stole " of 

 the adult Atherine. 



Coming to the characters of the ova, the large demersal type appears, 

 imtna facie, to be that most suitable to the requirements of the 

 presumably fluviatile ancestors of modern Teleostei, which is perhaps 

 the most conclusive argument forthcoming. The attachment process 

 of the zona of the Atherine egg certainly indicates a high degree of 

 specialisation ; but, as such would presumably be lost in the evolution 



