C. R. Keyes — Eolian Origin of Loess. 299 



Art. XXVIIL— Eolian Origin of Loess ; by Chas. E. Keyes. 



That the formation of loess deposits has ever been aided, to 

 any appreciable extent, by the wind, has never, in this country, 

 gained much credence. Almost without exception the aqueous 

 hypothesis has been accepted in explanation of the deposits in 

 the Mississippi Valley, in spite of the many grave difficulties 

 presented in its general application. 



Of late years, some American glacialogists have begun to 

 suspect that possibly more than one agency has been involved 

 in the production of the loess ; that water is the principal 

 agency in some cases, and in others the wind ; while in some 

 deposits both are concerned, or both kinds exist side by side. 



The accompanying notes are presented on account of their 

 direct bearing upon the eolian side of the question. They are 

 confined to the deposits of the Mississippi Valley. The loess 

 bordering the Iowan and other ice sheets is not considered — 

 only those deposits that cover the bluffs of the great rivers of 

 the region. Hence, for convenience, they are called by the 

 long-used name Bluff deposits. 



The prime reason for excluding from the present discussion 

 the loess of the former ice fronts, is that it doubtless had a 

 very intimate connection with the glacial agencies. Another 

 reason is that the observations herein recorded were made 

 largely upon the Bluff deposits alone. From this it might be 

 inferred that an attempt is made to differentiate two great 

 deposits of loess — one water-laid and the other wind-driven. 

 Such is not the case. No means of discriminating between 

 the two kinds of loess are yet known to be formulated. 

 Should, however, the suggestion offered for the formation of 

 the Bluffs loess be the correct one, the presence from bottom 

 to top in the one, and the absence or existence only near the 

 top, in the other, of limonite tubules, and an unusually prom- 

 inent jointed structure may possibly prove to be reliable criteria. 

 The nature of the fossils should also furnish a key to discrim- 

 ination. 



The areal distribution of the Bluff loess is peculiar. Prof. 

 Chamberlin has recently* stated it as follows : 



" The loess is distributed along the leading valleys. These 

 embrace not only the great valleys, the Missouri and the 

 Mississippi, but some of the subordinate valleys, as the Illinois, 

 the Wabash and others. The loess is found along the Missouri 

 River from southern Dakota to its mouth ; along the Missis- 



* Journal of Geology, vol. v, p. 795, 1897. 



