1022 REPORT OF COMMISSIONER OF FISH AND FISHERIES. [42] 



appears in the front part of the dorsal fin-fold, which is occupied by a 

 papilla ; by the sixth day this has been prolonged into an appendage 

 nearly as long as the tail of the larva. At its base it consists of a short, 

 rigid, vertical portion, to which is attaclied a long flexible filament, to 

 the sides of which are appended alternately three ovate, flattened, pig- 

 mented bodies, of which the terminal one is largest. In the vexillifer 

 stage, before the final metamorphosis into the adult, this flexible por- 

 tion of the dorsal appendage may support as many as ten of these now 

 nearly acuminate, flat, pigmented bodies. The base of this singular 

 appendage consists internally of a rigid, homogeneous axis of connect- 

 ive tissue, in which no ossification ever occurs. In the vexillifer stage 

 it finally has its base attached in front of the permanent dorsal, and 

 disappears when the adult condition is reached. 



It is clear that the development of this appendage is similar to that 

 of a fin, being a verj- elongate diverticulum of the epiblast, into which 

 pigment and connective tissue are proliferated from the mesoblast. 



A somewhat similar development of long dorsal and ventral filamen- 

 tous ai)pendages or rays, bearing opposite bract-like processes placed 

 at intervals, occurs in the case of the young of TraehypteruSj two to four 

 inches long, which seem to be aborted and lost: during later life. In Fie- 

 rajifer the vexilliferous appendage is not the rudiment of a true ray, but 

 in Trachypterus the l)asal portions of these filaments seem to be retained 

 as rays. 



The development of the tail of Tracliypterus is, however, not a little 

 singular. In the young of the size mentioned above, it is homocercal, 

 and, if the existing figures are reliable, it is structurally heterocercal. 

 Subsequently, however, the longest rays of the caudal assume an up- 

 right direction at right angles to the axis of the body, while a few short 

 hypaxial rays arise from the hinder and ventral border of the knob-like 

 swelling which terminates the tail. The development of the tail of this 

 group is evidently widely diflerent from that of other forms, and it is 

 not a little singular that the change of direction of the principal rays 

 of the caudal should apparently occur during the post-larval period of 

 growth. 



The Tem^iivkixhX^ Stylephoriis cliordatus must also be considered in this 

 connection, since it presents the peculiarity of having a body onlj' 11 

 inches long, with a caudal band-like appendage 22 inches in length. 

 " The caudal is directed upward, and has its rays connected by a rather 

 firm membrane; the tail terminates in a narrow band-like appendage, 

 about twice as long as the body." (Giinther, Cat. Fishes Brit. Mus., 

 Ill, 307.) This " baud-like appendage" is undoubtedly homologous, as 

 far as one is enabled to judge from Shaw's figure,* with the caudal fila- 

 ment of Chimccra monstrosa^ or, in other words, the opisthure of Style- 

 pJiorus is 22 inches long. This appendage of Stylephorus is below the 

 upright five-rayed caudal, which it seems to me is not such, but prop- 



* Naturalist's Miscellany, VIII, pi. 274. 



