274 The ylmerican Geologist. October, 1897 
sense a petrology, the pages devoted to rocks being regarded as a neces- 
sary introduction to that part of the book which deals with rock-weatii- 
ering, still it seems that much more space is given to the discussion of 
the origin, structure and classfication of rocks than is essential for the 
I^urposes of a cori-ect understanding of the processes and results of 
weathering. At the same time this petrological part of the work will 
be of assistance to many students who desire a brief introduction to the 
study of rocks and who cannot use and appreciate the larger treatises 
of Rosen busch and Zirkel. 
Part III begins with a consideration of the principles involved in rock- 
weathering and this is followed by a discussion of special cases. Here the 
chemical composition of a number of fresh and partially or completely 
weathered rocks is given, the different analyses of each rock being made 
from exposures which show the fresh and decayed rock immediately asso- 
ciated. Many of the analyses were made by the author, or under his di- 
rection, from material collected by himself with the special object of the 
book in mind. The value of these analyses for the work in hand is thus 
apparent. It becomes a matter of some importance in studying these 
analyses to determine the percentages of the various parts of the rocks 
lost by solution in the process of weathering. In order to attain a basis 
for the calculations a certain chemical component of the rock is assum- 
ed to have remained undissolved in the weathering process. The com- 
ponent thvis taken is usually alumina (AI2O3), but in some cases iron 
sesquioxide (FeeOa) is found to remain practically undissolved. In lime- 
stones the silica is thought to remain constant. With such a basis to 
start on the relative amounts of the various components lost is readily 
calculated. 
Part IV is devoted to transportation and i-edeposition of rock dt^bris, 
and part V to a discussion of the regolith, a term applied to the earth's 
mantle of unconsolidated material. Here is included a discussion, with 
many analyses, of the chemical nature and the various x:»roperties of dif- 
ferent kinds of soils. u. s. g. 
CORRESPONDENCE. 
Origin of the Loess. In the September number of the American 
Geologist I find it reported that in my paper on the origin of the loess, 
presented at the meeting of the Geol. Soc. of America in Detroit, I 
" took the wide view of the loess, including in it all forms and conditions 
of this material — that in the river valleys and that at all altitudes — and 
maintained that it was all due to wind-action and was an accumulation 
of dust carried in the atmosphere." In the paper referred to, the fol- 
lowing statement was made : " The waters of these rivers (the larger 
drainage channels of the loess region) may have added a part of the ma- 
terial (of the loess) in some places along their courses. This is indicated 
by occasional stratified phases of the loess in such places." 
I also find it stated that I should have "argued that its (the loess') 
lack of assortment was conclusive proof that it was not of aqueous ori- 
gin." What I said with regard to this was the following : "An aqueous 
