436 



known about it than is known of the Cephalaspidae, L i m u 1 u s ought, 

 logically, to stand intermediate between the two genera Cephal- 

 aspis and Pteraspis. I am very far from saying, however, 

 that it should occupy that position. As to its peculiar surface orna- 

 mentation, the shield of Pteraspis is exceptional among the 

 Cephalaspidae and need not at present be considered. 



The position of Pteraspis is doubtful and depends entirely on 

 its resemblance, in the minute structure of its shield, to Cephalaspis. 

 Huxley says: "No one can, I think, hesitate in placing Pteraspis 

 among fishes. So far from its structure having no parallel among 

 fishes, it has absolutely no parallel in any other division of the 

 animal kingdom". 



It is very probable that Huxley would not have made this as- 

 sertion had he known of the structure of the shield of Limulus. 

 However that may be, he and his followers place the Cephalaspidae 

 among fishes on the very uncertain evidence offered by the presence 

 of doubtful scales and fin rays, which may as well be those of an 

 arthropod as those of a fish. I do not deny, however, that 

 Cephalaspis is a fish, but I do maintain that what there is left of 

 it shows pronounced and unquestionable arthropod characters. To 

 state the matter concisely and in a preliminary way, it may be said 

 that Limulus resembles the osteostracous Cephalaspidae 1) in 

 the general outline of the cephalic buckler; 2) in the configuration 

 and structure of the anterior margin of the shield; 3) in the outline 

 of the shield in cross-section; 4) in the presence of lacunae in the 

 inner layers of the shield; 5) in the position and arrangement of the 

 three ocellar tubercles of the parietal eye ; and 6) in the bony capsules 

 surrounding the inner surface of the lateral eyes. Limulus 

 resembles the Pteraspidian section 1) in the arrangement of the 

 four layers, outer, reticulated, cancellated, and inner layer, as seen 

 in sections through the middle of the shield; 2) in the very charac- 

 teristic parallel lamination of the inner layer, and the concentric 

 lamination of the trabeculae; 3) in the faint vertical striation of 

 these layers; 4) in the absence of distinct layers and the great 

 development of the bony network along the posterior lateral margins 

 of the shield; 5) in the presence of a large opening or canal in this 

 thickened margin ; 6) in the number and arrangement of the radiating 

 grooves on the inner dorsal surface for the attachment of muscles; 

 7) in the presence of convex lateral eye tubercles, which could not 

 be present in these fossils if the eyes were not originally covered by 



