30 



Harker, Kansas, six specimens. I have also found specimens of the same 

 species at these two localities ; on one hill near Decatur, I found no other kind 

 of leaves with them. Near Fort Harker, the leaves of this species are mixed 

 with those of Laurus. All the specimens of Platanus obtusiloba, twenty-two 

 in number, come from near Beatrice, Southern Nebraska. Representatives of 

 this species have not as yet been found elsewhere. The specification of the 

 various forms of leaves of Sassafras, as may be seen in the descriptive part, is 

 confirmed by local distribution. Sassafras mirabile, S. cretaceum, S. hark- 

 erianum are from Fort Harker ; Sassafras mudgii and S. obtusum from Salina. 

 Most of the Pterospe.rmites, too, have been found at Fort Harker. The Salina 

 locality is along the bluff's of the river formed by a succession of low hills, 

 whose faces are more or less abruptly cut and exposed by erosion. Two of 

 these hills are separated by a narrow depression formed by a spring, and the 

 top of both is on the same level, as can be seen by the exposure of the upper 

 strata of red shale. One of them is strewn with fragments bearing specimens 

 of Menispermites obtusilobus, a species which I have as yet not seen from 

 any other locality, though it is there in abundance ; the other has only speci- 

 mens of Sassafras. A distribution of this kind can result only from the prox- 

 imity of the trees from which the leaves have been derived, and confirms the 

 opinion that the formation of the Dakota group is the result of muddy flats 

 whose surface, raised perhaps in hillocks above water-limits, and already solid 

 ground, was cut like an immense swamp, here and there interspersed by rare 

 groups of trees and bushes. Of course the main portion of this surface was 

 subject to continuous changes in the successive modifications resulting from 

 the heaping and displacement of matter by water, and thus the leaves were 

 distributed either at the same places but at different levels, or at the same 

 level but at different localities. 



The only objection to this supposition is the absence of roots or of 

 fossil trunks, which as yet have been rarely observed in the shale of the 

 Dakota group. 1 If we consider that the low islands bearing trees were 

 rare, and especially if we think how great is the disproportion in quantity 

 between trunks of trees with their roots, and the leaves which they bear 

 and disseminate annually for a series of years, we may easily account for 

 this apparent anomaly of leaves found petrified without accompanying trunks 



1 Fossil silicified wood has been found, however, in Nebraska, in the lowest strata of the Dakota 

 group, and also at tho upper part of the same formation. Near Sioux City, the shales are blackened by 

 the heaping of rootlets, as they are generally in the clay-beds preparing for tho formation of peat. 



