CRETACEOUS AND TERTIARY FORMATIONS 525 



is dark gray, but weathered surfaces, especially when moist, fre- 

 quently have a greenish gray or olive color. Beds of brown, 

 carbonaceous clay shale are very common and conspicuous. The 

 strata also contain much dark brown, ferruginous material, occur- 

 ring both in thin seams and concretions, the latter being most 

 numerous at certain horizons, and fragments of these cover the 

 slopes in many places. Great numbers of sandstone concretions 

 are present, some small, and others eight or ten feet in diameter. 



Only thin beds of coal, not over eighteen inches thick or less, 

 occur in the lower 300 feet or more of the Lance formation. Thus, 

 in the 250 feet of strata exposed at the mouth of Bacon Creek 

 there is practically no coal, the thickest bed being fifteen inches, 

 and the same is true for all the Lance shales and sandstones exposed 

 along the Little Missouri from the Pretty Buttes, five miles below 

 Marmarth, to the South Dakota line. But in the upper portion 

 of the formation, thick beds of lignite are found in many places. 

 In the vicinity of Yule, five or six coal beds are present in the upper 

 part of the member, and the coal of Bacon and Coyote creeks 

 occurs at about the same horizon. 



The basal beds of the Lance formation, together with the 

 underlying Fox Hills sandstone, are well exposed on Little Beaver 

 Creek, several miles southwest of Marmarth, and the relation of 

 the two has already been described in connection with the Fox Hills. 



The massive sandstone forming the top of the latter is seen to 

 have undergone erosion before the deposition of the very carbona- 

 ceous and argillaceous, brown and black sandstone, which shows 

 cross-lamination. Some of the depressions have been eroded to 

 a depth of six feet below the adjoining elevations (Fig. 6). 



The uneven surface of the Fox Hills shown here is perhaps 

 due to contemporaneous erosion, and practically continuous 

 deposition may be represented in this locality, as on the Cannon 

 Ball River. The thickness of the Lance beds in southwestern 

 North Dakota is approximately 600 feet. It is not possible here 

 to divide them into three members, since the upper sandstone 

 and the middle shale member are absent, and the strata are com- 

 posed throughout of rapidly alternating shales and sandstones. 



Plant remains are by no means as abundant in the Lance for- 



