No. 436.] APPENDAGES OF TREAT A TASPIS. 



239 



and more nearly parallel with the long axis of the head, the tend- 

 ency being to draw the basal plate more and more diagonally 

 across the margin of the shield, instead of straight across it as 

 during the middle of the stroke. This factor and the sudden 

 change in the intensity of muscular contraction at the beginning 

 and end of each stroke furnish a very satisfactory explanation of 

 the increased depth and width of the ends of the eroded groove. 



To return to the basal plate shown in Fig. 4. A close exam- 

 ination of its convex surface shows a second belt, not shown by 

 the photograph, farther back than the first where the surface 

 enamel has been very slightly eroded. It extends about two 

 thirds of the way across the lower portion and is coextensive 

 with the line of greatest elevation of the surface. It is hard to 

 see, on any supposition as to the location of the plate how this 

 abrasion could have been produced by contact with any part of 

 the animal's body to which the plate was attached. I assume 

 that the abrasion, if made during the lifetime of the animal, 

 must have been produced by contact with the bottom over 

 which the animal was moving. 



