1910.] CAUDAL Kl.V OK THE TELKOSTOMI. fi09 



large number of radiiils in addition to the epural ami li\ pin-iil 

 bones. Except for tlie fact that eleven vertel)ra' form tlie l)asis 

 of support for the tin instead of thirteen as in Molva, all remarks 

 made in i-eference to the latter genus apply equally to Gadiculiis. 

 The whole interest in this tail-tin lies in the radials ; in addition 

 to the epurals no less than six are present epaxially, and ventrally 

 three are to be seen alternating with hypiu'als. Moreover, these 

 radials may be either quite free, like ordinary radials, or they 

 may be more or less fused with the neighbouring hypurals or 

 epurals. This aflbrds one of the strongest cases for the theory 

 that hypural bones are formed by the coalescence of radials with 

 hiemal arches. In the other median fins two I'adials are present 

 between successive vertebral arches ; thus probably the hypurals 

 and epurals of the caudal have already absoi'bed one I'adial and 

 the tendency is to efiect a fusion of the second also. There can 

 be no doubt about these I'adials being true I'adials and not a 

 second neural or haemal ai'ch, since they never reach the centrum 

 itself and never embrace the spinal cord or hfemal vessels at their 

 proximal ends, as they would do if they wei'e arches. The spinal 

 cord can be seen to extend to the distal end of the last hypural 

 bone. This caudal also is therefoi'e a homocercal form. 



Genercd reinarks on tlie (Jadklte. 



There seems to be a certain amount of misconception among 

 writers concerning the caudal fin of the Gadidse ; one or two 

 quotations will make this cleai-. J. T. Cunningham, in ' Science 

 Progress,' 1897, says: — " The structure and development of the 

 tail described above (heterocercy) occur in the flat fish, but are 

 entii'ely wanting in the Gadidfe. In the latter the tail is per- 

 manently diphycereal and is composed of dorsal and ventral rays 

 which are equal in number and size, and, in fact, closely I'esenibles 

 the tail of the extinct C(elacanthida3. There can be little doubt 

 that even if the Gadidaj cannot be directly derived from the latter 

 family, they are descended from Crossopterygian Ganoids with 

 diphycereal tails and have never passed through a heterocercal 

 condition." And Boulenger, in the Ann. ct Mag. Nat. Hist. x. 

 1902, writes : — "I have reason to believe that Gadoids must have 

 been derived from such a group as the Berycidte, through forms 

 of which the Macruridaj Avith thoracic ventral fins composed of 

 seven to twelve rays are the nearest known examples, and in 

 which the caudal fin has entirely vanished. I regard the isocercal 

 condition of the Gadida^ as the result of the formation of a new- 

 caudal fin, the homocercal extremity of the \erteljral column 

 having been lost by the direct ancestors of these fishes." The 

 same writer in compai'ing the caudal fin in the Pleuronectidse 

 with that of the Gadida^ says : (in the Pleuronectidie) "• the tail, 

 whenevei'a caudal fin is well developed, belongs to the homocercal 

 type (heterocercal in the embryo) with coraparati\"cly few Y!\y?, 

 (20 or less)," — thus differing from the Gadoids. 



