372 NATURAL HISTORY BULLETIN. 



and were probably also affected by inequalities in forest and 

 plant distribution during the deposition of the loess, to 

 nothing of the variation in the winds themselves. Strong 

 winds carry the dust over the entire valley, and its final set- 

 tling and accumulation may be affected by various local causes. 



2. He further states that "the exclusive dependence upon 

 wind for the distribution of the loess finds it difficult to account 

 for the extensive level-topped terraces which frequently occur. 1 ' 

 This statement conveys an exaggerated idea of the extent and 

 frequency of such terraces. They represent a very small part 

 of the loess-area even along the Missouri river, and where they 

 do occur their vertical contour is determined largely by the 

 underlying deposits on which the loess accumulated. In 

 any event they are too insignificant when compared with the 

 extent of loess deposits, to form a safe basis for generalizations. 



3. He criticises the writer's explanation, which is con- 

 sistent with the seolian hypothesis, of the fact that loess is 

 thickest near the streams, and that its particles are here some- 

 what coarser, but he evidently failed to grasp its significance. 

 It is not here a question of moving sand upon the lower 

 heights. The particles of which the loess is composed are 

 all sufficiently fine to have been sustained in the air as dust. 

 They are not uniformly fine, but the differences between them 

 are not to be compared with the differences between particles 

 carried by water-currents, and those sustained in sluggish wa- 

 ters. Naturally the coarser particles would usually be deposited 

 nearer the chief source of supply, — namely the bars of the 

 streams, — and their retention would be favored by the usually 

 greater, though unequal, extent of forest growth, or at least 

 plant growth, in such places. But Professor Wright signally 

 fails to explain why under his hypothesis of great floods land- 

 snails, favoring high, comparatively dry grounds, should be 

 most abundant in the portions of the deposit nearest the thread 

 of the streams, and where the greatest amount of mud was de- 

 posited. 



In this connection it may be well to notice also the ob- 



