S. V. Wood—Origin of the Loess. B95 
its features, they are so that, with the exception of freshwater 
shells, such as Limnea, occasionally, the mollusca in it are of terre- 
strial or amphibious habit. As regards the absence of stratification, 
and the general relation of the Loess to other formations, Hilgard 
says that, though unstratified, it often shows bedding planes, and that 
the Loess of the Lower Mississippi passes laterally and impercepti- 
bly into a clayey loam, undistinguishable in its landward form from 
that which constitutes the general soil of the South-western States ; 
and that ‘all observers testify to its dead uniformity over entire 
areas.” These “bedding planes” often shown by the Loess seem 
to me to correspond to the “layers” in which, Dall says, the Alaska 
clay accumulates, where it occupies depressions in the ice mass it 
covers. General Warren says, “that in many places the Loess is nearly 
uniform in thickness, conforming with the previous irregularities 
of surface, as snow lies where it has not drifted.” Also that it is 
‘Ca fine material deposited over all other formations, including the 
Glacial deposits near the river, but not upon the river sand and 
gravel terraces.” If this last be so, it seems only reconcileable with 
Lyell’s statement, that the Loess overlies the river beds at Natchez, 
by this being a terrestrial formation synchronous with those terraces, 
and which slid but partially over their edges, as these became land 
before the formation ceased. 
In his 6th Report of the Geological Survey of Minnesota, Win- 
chell says that the Loess overlies “the boulders and gravelly clay of 
the real drift ;”’ in respect of which I may observe that this does not 
make it later than the minor. glaciation, the moraine of which may 
not have reached to where this occurs, and the Loess may here have 
been generated during this glaciation on the more advanced moraine 
of the major. It may, indeed, have been generated on this, here, 
during the latter part of the major also, as the ice receded ; for 
Winchell adds that there is a gradual change from stony Boulder- 
clay to Loess horizontally. He also says that it is more sandy or 
clayey according to locality, but that while very thick in some con- 
fined valleys, it is thin or wholly absent on some rolling gravelly 
tracts. This latter agrees with what I have mentioned as to the 
action in which the material in question has I consider arisen, porous 
sand and gravel being adverse to it; the material where it covers 
beds of this kind, as, for instance, in the Somme valley, having slid 
from contiguous areas favourable to this action, just as the clay 
over the Alaska ice cannot have arisen from that ice, but must have 
slid from some contiguous rock area. | 
Although I am clear that the Picardy loam, the intrusive cave- 
earth, and the other formations described under the letter y in my 
Newer Pliocene memoir, originated in the process I have described ; 
and I think that the Limon Hesbayen, and the Rhenish and American 
Loess must have originated in a similar way, during both glaciations, 
yet I venture no opinion as to the extraordinary formation of North 
China, also called Loess. The accounts of its thickness are so stag- 
gering, and its origin in this amazing thickness (whether from dust 
or any other form of terrestrial accumulation, or from freshwater 
