1010.] CAI-DAI. KIN OK Tin; TKLKOS'J'OMI. 597 



vei-tel)rMl segment which h;is l)efii found. The doi'sal contrilni- 

 tiori to the ciuidid fin is cninparatively small. Of the last four 

 vertebi-ie all but the last hear epural bones and the last appears 

 to have its neiu-al arch greatly reduced. 1'he most jiosterior of 

 the epaxial fin-iuys ai'e supported by two or three poorly developed 

 dorsal caudal radials ; the proximal ends of these radials are 

 almost in contact with a triangular cartilage applied to the doi'sal 

 side of the notochord near the termination of the lu'ostyle ; this 

 cartilage probably represents a vanishing neui"al arch. 



Clupea pilchardus. (Clnpeidse.) (Plate XLVII. fig. 5.) 



The caudal fin of the Pilchard very closely resembles that of the 

 Ancbovy. The urostyle {ur.) is long and slender, and the noto- 

 chord (/?.) is produced beyond it, but not be3-ond the distal ends of 

 the endoskeletal supports, An opisthural cartilage (o.c.) is 

 attached to the end of tlie notochord, and extends between the 

 dermoti'ichia. Nine hypural bones are present, seven of which 

 belong to the last vertebral segment, as in hjngrnidis ; tlie 

 hypurals vaiy considerably in size, Epaxially, the dei-motrichia 

 are supported by two epu.rals {ep.) and three dorsal caudal radials 

 ((Z.c.r.) ; the latter are short bones situated between the last 

 epural and the notochord. Proxiraally the dorsal radials abut a 

 triangular bone {n.a}) — cartilage in E^yraid'is — on the dorsal 

 side of the urostyle : this triangular bone probably represents one 

 or more neural arches of vertebrae now nbsorbed into the urostyle. 

 Tlie List centi'um {l.v.) is pi'ovided with a large spatulat^ neural 

 ai-ch {n.a.) peculiarly constricted at its proximal end. 



Compivring the caudal fins of the two Clupeoids here described 

 and also that of the 8prat as detailed by Newton, it will be seen 

 that there is a striking similarity throughout ; the figure given in 

 Newton's paper is that of a young Olapea sprattios, and the 

 urostyle is seen to be composed of several centra which, no doubt, 

 are also present in C pilchardus and Eiigrcadis eiicrasicholus in 

 the young, but which are obliterated in the adult All show a 

 lowly specialized homocercal form, approaching very closely the 

 Aitiia type. 



Nqtopteru^ BORNEExgis. (Notopteridte.) (Plate XLVll. 

 fig. 6.) 



This fish is provided with a highly specialized caudal fin, and 

 also presents one or two peculiarities. The caudal is continuous 

 with the ventral median fin, but the caudal proper may here be 

 limited to tliat part composed of dermotrichia which ai'e attached 

 to the endoskeleton by means of a proximal forking, as distin- 

 guished from the ball and socket nature of the attachment of the 

 anterior tin-rays to tlie radials. Thus the caudal fin is supported 

 Iiy three hypurals and one epural ; hence it is not strictly hypo- 

 • cereal as Gregoiy states. No urostyle can be seen in the dissection, 

 and even clearing fails to give any certain clue as to its exact 

 outline in the substance of the l;\st hy)tui-al lK)iie. and what 



