go KANSAS UNIVERSITY QUARTERLY. 



species as O. ellsworthensis, but a collection of a large number of forms may show 

 proper gradations. The forms so far collected by the writer do not point to such 

 a conclusion. 



Ostrea canonensis, n, sp. Plate XXI, Figs. 3, 4. 



Shell (left valve) less than medium size, partly crescent shaped, 

 moderately convex; margin crenulate; test thin, nearly uniformly 

 so; beak small, pointed; ventral border forming concave curve, 

 smooth, no crenulations; dorsal border forming convex curve, 

 crenulate; posterior border short, rounded; exterior surface orna- 

 mented with concentric lines of growth; interior of the shell moder- 

 ately smooth. 



Measurements: 



Length 25 mm 



Width 20 " 



Height- 10 " 



Locality: Coal Canon, Eilswortli county, Kansas. 



Geological horizon: Lower Cretaceous. 



Remarks. — Had I found this species associated with O ellsworthensis I would 

 have called them the same species and the former the young of the latter. F"or 

 here as in other forms of ostreidce there exists as much difference between the 

 forms of young and adult as there is between these two forms But as they are 

 not found associated, and several forms of each were obtained, they have been 

 designated as different species. 



Ostrea cong-estB. Conrad. Plate XX, Figs, i, 2, 6. 



I am not positive that tlie form here figured and designated as 

 O. congesta does not differ specifically from that species. Writers 

 upon the Upper Cretaceous of Kansas have applied the name, O. 

 congesta, to two forms which are separated by at least two distinct 

 geological horizons. The first of these two forms occurs in the 

 Ostrea beds of the Ft. Benton group and the second in the lower- 

 most Niobrara beds. 



Although these forms differ in many particulars they have certain 

 common characteristics. Both are attached by the whole under 

 surface of the lower valve; both are attached to luoccranii: both 

 are members of crowded groups or colonies; both are decidedly ir- 

 regular in form, consequent upon their method of growth. 



They differ in thickness of test, the Benton form being much 

 thinner; they differ in size, the Niobrara form being much larger; 

 they differ in that the muscular impression in the Benton form is 

 indistinct, in the other it is not usually so. 



As these two forms do not grade into each other it is fair to 

 assume that they are, at least, varieties if not distinct species. 

 Figures heretofore published best represent the Benton form while 



