1877.] 205 [Garman. 



weaker tail, and a small spine placed so close to the body that the 

 radius of its circle of motion is so insignificant in comparison with the 

 extent of dorsal surface that it is impossible that it should depend on 

 capturing its prey by the same means as the former. The ventrals are 

 attached backward, and have not the same freedom of motion for 

 digging. The small eyes, the position of the mouth so near the cen- 

 tre of the ventral surface, and the extraordinary distribution of 

 nerves under the forward half of the body, tend to confirm me in the 

 belief that this genus finds its food by means of touch, on the lower 

 side rather than the upper, and captures by pressing down with the 

 enormous pectorals, when the struggles of the captive will help to 

 bring it to the mouth. A shorter and weaker prepelvic cartilage is 

 found in this genus. 



Behind the spines the tails of older specimens of these genera 

 become calcified to such an extent that the vertebrae are consolidated. 

 Once rigid they are easily broken, and it is only young individuals 

 that are found with the tail entire. The caudal spine is replaced by 

 a new one which appears under and posterior to it. If not readily 

 detached, as is often the case, the succeeding protected by the first 

 attains its growth and is in many cases followed by another reaching . 

 a size still larger. Should the specimen be growing rapidly the 

 hinder ones are carried back so that each is at a distance from the 

 other. Where the growth is slow the spines are bunched together, 

 the base of the later resting beneath and very little behind the pre- 

 ceding. 



Anacanthus, of the Thalassotrygons, has a compressed pelvic shaft, 

 arching forward and without spine or angle in front. 



In Trygon the shaft is slightly depressed, and more or less regularly 

 arched in front. 



Tceniura and Urolophus have the lateral halves more straight and 

 meeting in front in a very blunt angle. 



Pteroplatea has still more of a furculum type. 



Myliobatis, Aetobatis and Rhinoptera possess a greater forward 

 arch and the median angle is produced. 



C ephaloptera presents the perfection of the furculum type, and the 

 production of the angle in front bears a certain resemblance to the 

 subabdominal cartilage of the Potamotrygon. 



All of the species included in the Thalassotrygones the Myliobatidce 

 and the Cephalopterce have the ventrals directed backward. Their 



