1896.] NEW YORK ACADEMY OF SCIENCES. 65 



the most specialized. But tliere can certainly- be but little 

 doubt that Edesfus is an elasmobranchian fossil. The character 

 of the primary denticles, their shape, denticulated edges together 

 with their bank-like arrangement are almost conclusive evidence, 

 even if unsupported by similar histological conditions in enamel, 

 dentive and nutritive canals. 



If, in any event, the five species in the adjoining figure are 

 compared, as, for example, in the above table, a number of very 

 suggestive conditions are readily determined. These might be 

 arranged in somewhat the following order : (I) The primary 

 denticles, small in size in E. heinrichsii, tend to become larger, 

 and are relativelj^ largest in the terminal member of the series, 

 E. lecontei. They seem also to become more numerous. And 

 (II) in the same order the secondary (marginal) denticles tend to 

 become smaller and more numerous. (Ill) The shaft of the 

 spine becomes reduced, and at the same time narrows at one 

 end. (IV) In a general wa}^ the segmented elements of the 

 shaft tend to shorten their dorsal {%. e.^ on the side nearest the 

 row of denticles) margin and increase their ventral margin. In 

 this regard, E. davisii appears more highly differentiated than 

 E. lecontei. A satisfactory' comparison of the inclined direction 

 of the primary denticles cannot well be made : it would seem at 

 first that the decnrving of the shaft was accompanied with the 

 inclining of the denticles in a single direction, but unfortunately 

 this comparison does not hold good ; for where the shaft is 

 greatly decurved, as in Figs. 5 and 6, the inclination of the den- 

 ticles is in two directions, those in the middle of the spine re- 

 main erect, while those at the ends incline in opposite directions 

 from an angle of 90° to the longitudinal axis of the shaft. 



In interpreting these conditions generall}', one must, I believe, 

 be lead first to the conclusion that the forms in this series are 

 genetically related to each other, more or less nearly, and that 

 one of the two forms E. heinrichsii or E. lecontei must be re- 

 garded as representing the ancestral type, or vice versa. That 

 the form represented hy E. heinrichsii is to be placed at the 

 base of the series, next becomes evident when we consider in 

 evidence the following facts : (1) that it presents in its elements 

 the most perfect serial homology, (2) that its denticles are 

 smaller, i. e., more nearly like those of the neighboring integu- 

 ment, from which they took their nltimate origin, (.3) that the 

 secondary denticles resemble closely the primary ones. 



The establishment of this series of forms in Edestus seems to 

 the writer of far greater interest than first appears. For by 

 this means an important aid is secured both in interpreting the 



Transactions N. Y. Acad. Sci., Vol. XVI., Sig. 5, February 22. 1897. 



