394 S. V. Wood—Origin of the Loess. 
loam with, for the most part land shells only, thus resulted, and 
was laid over stratified river beds with freshwater mollusca, it would 
not be in one continuous bed of unstratified loam 60 feet thick, as 
Sir Charles Lyell says he found it, but either films of this inundation 
mud with alternations of stratified beds with river shells would 
occur, or (and more likely) this mud would become mixed with the 
beds with river shells; for it is only beyond the area occupied by 
the permanent stream that the successive films of Nile inundation 
mud accumulate to form a continuous mass; and this, according to 
Lyell, proceeds at the rate of only 34 inches in a century. If the 
objection as to the height to which the Loess not thus underlain 
rises, be met by the answer that there was a cutting down of the 
valleys by the rivers during its accumulation, then the unstratified 
Loess with land shells ought to be found overlying stratified beds 
with river shells at all levels, instead of adjoining the rivers, as it 
is said to do, only. Lyell (Principles of Geology, vol. i. p. 461) 
mentions that the remains of three Mississippi Buffalo fish occurred 
at a depth of four feet from the surface of the Loess plateau above 
. Vicksburg, and 200 feet above high-water mark. If the presence 
in it of fish-remains were one of the general features of the Loess, 
this would outweigh all objections to its fluviatile or lacustrine 
origin, but I do uot find such alluded to as the case in any of the 
descriptions of the Loess by American geologists that I have met 
with; so that, considering the Mississippi Valley has been for ages, 
and probably during both glaciations, inhabited by races of men 
who have lived on fish obtained from the rivers, this isolated 
occurrence in unstratified material, only four feet from the surface, 
requires some collateral support to make it good evidence on the 
subject. 
In his description of the Alaska Clay, Dall mentions a six-inch 
bed of marl of freshwater shells (Pisidium, valvata, ete.), of bog 
origin, as present at one spot in the clay ; and it is obvious that if 
the Loess has originated in the sliding way that I suppose, such 
occasional patches of freshwater shells must have become incor- 
porated with it. If from shrinkage in the river volume, or from a 
cutting down of its bed by the river, the ordinary deposit of the 
river became land along its edges, an atmospheric formation with 
only land shells, originating and sliding in the way I suppose, would 
be carried over it, and would thus give rise to that overlay of beds 
containing only river shells, by Loess with only land shells, which 
Lyell says occurs near the Mississippi.’ 
The descriptions of the Loess in different States of the Union by 
American geologists are too numerous for me to attempt to give an 
analysis of them; but though not altogether harmonious as to all 
1 In the section of the Trowbridge railway cutting at page 719 of the second part 
of my Newer Pliocene memoir (Quart. Journ. Geol. Soc. vol. xxxviii.) reproduced 
from a paper by Mr. Mantell, the loam (Drift) with Elephant teeth (y of my memoir) 
which ] refer to this sliding origin during the minor glaciation is shown as not only 
enveloping the Jurassic formations of the hill through which the cutting was made, 
but as spreading over the gravel of the river Biss. 
