1896.] NEW YORK ACADEMY OF SCIENCES. G7 



Dr. Newberry never intended to establish definite homologies 

 between these two spines, far less suggest for a moment that 

 there was genetic relationship between them, for the phylogeny 

 of the widely divergent types of rays is too perfectly known to 

 admit of the possibility of their tail spines being a primitive char- 

 acter — anymore in fact than the lateral tail spine of Acai^thurus. 

 His suggestion was merely that the spine of Edestus might be 

 located " in the position of the second dorsal fin on the back or 

 tail of a plagiostome fish (op. cit. p. 9.)." 



But the series of forms above described gives us now a some- 

 what more definite groundwork for a study of the relationships 

 of Edestus. As a flattened spine (E. heinrichsii) it bears evi- 

 dence of having been so imbedded in the integument that a 

 straight row of enlarged shagreen denticles alone protruded 

 above the surface. Each denticle with its underlying basal seg- 

 ment would thus be interpreted, as the present writer has already 

 suggested,* as a metameral element, each denticle with its sec- 

 ondary cusps representing the sum of the shagreen papilhie oc- 

 curring in a particular region (median dorsal ?) of a metamere, 

 its gouge-like underlying portiovi the fusion of the bases of these 

 denticles. From these conditions the spine has advanced in its 

 path of specialization, becoming more decurved until finally it 

 assumes a hook-shaped outline. Its hinder end tends at the 

 same time to become differentiated, and in the narrow end of the 

 shaft in E. lecontei no trace of segmentation can be determined. 



From its distinctly metameral character, as shown in the sim- 

 pler t^'pes, Edestus is certainly to be given a very primitive 

 rank among ichthyodorulites wherever its position ma}- have 

 been. That a large portion of its basal elements was inserted 

 in the integument seems exceedingly probable for all forms, and 

 in this event the curved spines must have been placed either 

 upon the summit of a fleshy, rounded ridge, or within a trench 

 as an erectile spine — the latter doubtless less probably. It is at 

 this point worthy of note that the narrow end of the shaft of 

 E. leco7itei heurs no trace of segmentation. Can this be interpre- 

 ted as evidence that the hinder end projected above the body 

 surface ? In any event does the present evidence gainsay the 

 thesis of Newberry that the spine is bilaterally symmetrical and 

 must, therefore, have found its place in the median line of the 

 body? As far as the writer is aware, all evidence is in favor of 

 this view; no specimen can be pointed to as showing definitely 

 traces of asymmetry. But, on the other hand, it cannot be 

 doubted that this bilateral condition might have attained in 

 spines which were not strictly in the median line. In view of 



*Fislies Living and Fossil, pp. 28-30. 



