G. R. Wieland — Upper Cretaceous Turtles. 193 



bridge (as indicated by fragmentary portions accompanying 

 Yale specimen No. 625, and by marginals 4 and 5). 



Limbs. — Little known ; humerus and femur thalassoid, and 

 of nearly equal development. 



Habitat. — Infralittoral. Habit. — Conchifragous. 



Range. — Upper Cretaceous and Lower Eocene of Europe 

 and America. 



Systematic Position of Lytoloma and of Osteopygis.* 

 Where shall we place Lytoloma, which in common with 

 Osteopygis and Propleura has a reduced and somewhat Chely- 

 dra- or Stater otypus-like plastron, and a distinctly Chelone- 

 like carapace ? It is my belief, based on certain somewhat 

 fragmentary fossils which it is proposed to illustrate later, that 

 in the New Jersey Upper Cretaceous there were already pres- 

 ent forms more nearly related to Chelone than is Lytoloma, 

 although as Dollo well suggests such are far rarer than has 

 been assumed. Second, the free tenth marginal of Lytoloma 

 indicates that the swinging back of the ninth rib (or eighth 

 and ninth ribs), thus leaving the ninth or tenth marginal, as 

 the case may be, without rib-support, took place early, and was 

 correlated with the shortening of the femur and the develop- 

 ment of heavy front nippers. Third, it appears that Osteo- 

 pygis and Propleura belong to a side line, with long and still 

 chelic femora, which never accomplished the rib change just 

 mentioned and did not survive ; and fourth, Lytoloma origi- 

 nally sprung from this side line. If so, the latter genus devel- 

 oped by parallelism a carapace which, with the skull (that of 

 Osteopygis being yet unknown) and the thalassoid humerus, 

 brings it so near to Chelone as to make necessary the inclusion 

 of both these genera in the same subfamily. 



Having settled this point, the question remains as to whether 

 the two genetic groups containing Osteopygis', Propleura, and 

 Lytoloma on the one hand, and the living members of the 

 Cheloninse and their more direct ancestors in the other, shall 

 be included in the same subfamily. Osteopygis, the most 

 primitive of all the forms in question, is removed from Chelone 

 by its less modified limb structure, with all or nearly all the 

 claws present ; by its less reduced marginals, consecutively rib- 



* In my first paper on the Upper Cretaceous Turtles of New Jersey (this 

 Journal, vol. xvii, Feb., 1904), the opinion was tentatively expressed that 

 Osteopygis and Propleura might best be separated in a distinct family, 

 namely Cope's Propleuridse, but the utmost degree of separation any one 

 might suggest, now that Lytoloma has been more closely considered, would 

 be as a subfamily, — the Propleurinse. The position now assigned to these 

 forms is virtually that given them in a provisional classification of marine 

 turtles, proposed earlier (this Journal, vol. xiv, p. 108, 1902), and to which 

 I shall as yet adhere, although recognizing with Dollo the great difficulty, if 

 not impracticability, of satisfactorily dividing the Cheloniidae into sub- 

 families in the present imperfect state of our knowledge of the group. 



