TUK DERMAL FlN-RAYS OF IRISHES. 469 



ill diameter towards the distal extremity. The horny fin- 

 rays — which I shall call the ceratotrichia — never become 

 ossified, and are never jointed. Moreover, as a rule, they ai'e 

 very numerous, and form more than one layer, are set close 

 together, are evenly distributed over the fin, and occasionally 

 may branch. In the adult fish, except near the distal g'row- 

 ing edge of the fin, they are deeply embedded in the con- 

 nective tissue some Avay from the epidermis (figs. 1 and 7 ct.), 

 and overlap to a considerable extent the cartilaginous fin- 

 supports. They are developed on both sides of the fin, and 

 proximally embrace the median cartilaginous rays. The 

 proximal ends of the ceratotrichia are enclosed in the sheet 

 of connective tissue to which the special fin-muscles arc 

 attached (fig. 1, cut.). These muscles, developed from the 

 muscle-buds which sprout from the myotomes, are the 

 appendicular muscles which occur in the fins of all fish, and 

 are parallel to and correspond with the endo-skeletal rays. 

 Outside the ceratotrichia and the surrounding layer of con- 

 nective tissue lie the denticles, and lastly the epidermis. 

 The denticles bear no relation either in position or in number 

 to the underlying dermal rays. 



Klaatsch (21) has given some details concerning the early 

 development of the ceratotrichia. According to him cells 

 derived from the epidermis migrate inwards into the under- 

 lying mesoblastic tissues, where they secrete the rays. The 

 horny substance appears first in the centre of these '' sclero- 

 blasts," and subsequently enlarges and becomes surrounded 

 by a multicellular sheath. These statements are supported 

 by very unconvincing figures. 



My own observations are directly opposed to Klaatsch's 

 interpretation. Sections through embryos of Scyllium or 

 Acanthias properly preserved, and treated with appropriate 

 stains (such as borax-carmine and picro-nigrosin), show 

 conclusively that a continuous and definite basement mem- 

 brane separates the epidermis from the mesoblastic tissues 

 over the whole surface of the animal (figs. 7 — 12 h.m.). At 

 what stage this membrane first makes its appearance I am 



