474 LOUIS AGASSIZ. 



ler, are really very curious combinations of 

 plates, and when viewed in a slant light have 

 a decidedly sculpturesque and not ungraceful 

 effect. I have seen on our rustic tombstones 

 worse representations of angels, winged and 

 robed, than that formed by the central plates 

 of the interior surface when the light is made 

 to fall along their higher protuberances, leav- 

 ing the hollows in the shade. You see how 

 truly your prediction regarding the flatness of 

 the creature's head is substantiated by these 

 casts ; it is really not easy to know how, 

 placed on so flat a surface, the eyes could have 

 been very available save for star-gazing; but 

 as nature makes no mistakes in such matters, 

 it is possible that the creature, like the flat- 

 fishes, may have lived much at the bottom, 

 and that most of the seeing it had use for 

 may have been seeing in an upward direction. 

 None of my other specimens of bucklers are 

 so entire and in so good a state of keeping as 

 the two from which I have taken the casts, 

 but they are greatly larger. One specimen, 

 nearly complete, exhibits an area about four 

 times as great as the largest of these two, and 

 I have fragTnents of others which must have 

 belonged to fish still more gigantic. The 

 two other casts are of specimens of gill covers, 



