during their Growth and Development. 9 



lateral line, which in the anterior part of the body unites 

 with the inferior, at a point situated beneath the first dorsal 

 spine. The palatine bones in their posterior part are armed 

 each with three or four small teeth ; and the ventrals are 

 formed bj one spinous and four soft rays. In the earliest 

 phases of their development the Gempyli have so different a 

 physiognomy that it would be difficult or impossible to recog- 

 nize them for what they are if one did not know the interme- 

 diate stages. The body is short and thick-set ; the first 

 dorsal attains its greatest elevation in front, and then decreases 

 rapidly ; we find two free spines in front of the anal fin ; the 

 denticulated spines of the ventrals are comparatively very 

 long, as long as (and even longer than) the spines of the dorsal ; 

 there are neither scales nor lateral line ; nor are the finlets 

 distinguishable ; the operculum and prajoperculum are spi- 

 nous. For what relates to the course of the development 

 and transformation I refer the reader to the figures on pi. iii. 

 [of the Danish memoir]. I have also described and figured 

 some y^r^ analogous stages of development in a fish of the 

 Thyrsites tribe, perhaps Nealotus tripes. Dicrotus armatus^ 

 Giinther, is certainly an analogous transitory form, probably 

 of Prometheus atlanticus. 



6. Thynnus ; Oecynus ; Pelamys ; Cybium and 



ACANTHOCYBIUM. 



In the group Thynnidse I distinguish the following genera: — 

 Orcynus^ Thynnus, Auxis, OrcynojJsis, Pelamys, Cyhium, and 

 Acanthocyhium. I must leave on one side certain other genera 

 which have been established, because I do not know them — 

 for example the genera Grammatorcynus, Gill [Thynnus bi- 

 iineatus) , Lepidocyhium and Apodontis, Benn.; the separation 

 of these two last genera from Cybium seems to me, however, 

 to be not well founded. Gymnosarda Gill [Pelamys nuda) 

 will no doubt prove to be a synonym of Orcynojjsis, G. [Pela- 

 mys unicolor ■=^ Thynnus 'peregrinus, Coll.*); even the specific 

 differences which serve as the basis of these two genera are 

 rather uncertain, and will need to be confirmed, although one 

 of the two species is a native of the Mediterranean (it has been 

 once found in the North Sea) and the other of the Red Sea. 



I retain the name of Thynnus f for " the small tunnies," 



* M. Giglioli has recently proposed for this genus the name of Pela- 

 michthys, which, however, must yield priority to that proposed by Mr. 

 GiU. 



t M. Giglioli designates this generic division by the name of Thynnich- 

 thys, a very happy denomination, but one which has already been 

 employed for a genus of Cyprinoids, 



