800 HE (Os (Cla ANNES EISILI ONY. 
not necessary to suppose that the periodic extensions of the floods 
were destructive of the vegetation over all the flat region. In 
some portions not only could vegetation persist, but the land 
mollusks and other animals dependent upon the vegetation could 
find a temporary retreat from the flood on the taller vegetation 
that may have prevailed. 
After each of the periodical retreats of the water there 
would be left extensive silt-covered tracts facily exposed to the 
sweepings of the wind and from these, when dried, dust could be 
derived in great quantities to be borne away over the adjoining 
lands and lodged in their vegetation. The material thus 
derived would be essentially identical with the glacio-fluvial 
deposition, and thus the hypothesis seeks to account for the 
glacial element in the constitution of the eolian portion of the 
loess. The presence of land mollusks in the upland eolian loess 
finds in this way a ready explanation, while their presence in the 
lowland loess mingled with aqueous mollusks finds an almost 
equally obvious elucidation ; for not only would the upland shells 
be washed into the lowlands, as we observe they are at the 
present time, but they would periodically invade the lowlands in 
the intervals between submergence and would be caught and 
buried there. Occasionally the shells of the lowland and aqueous 
mollusks would be borne to the uplands by organic agencies, 
and possibly in rare instances by the severest type of winds, 
and hence their occasional presence there is not remarkable. 
To make this a good working hypothesis it would appear 
that there must be an accommodation between the breadth and 
fluctuations of the fluvial deposits and the extent and massiveness 
of the eolian deposits, for if we suppose the glacial floods to be 
confined within narrow channels, the sweeping ground of the 
winds would have been too scant to give origin to the great 
mantle of silt then attributable to them, for we must remember 
that in proportion as the river work is narrowed the wind work 
is expanded. It is obvious that the eolian factor will cut away 
its own ground if pushed too far. 
There is little question that loess-like accumulations are now 
