172 The American Geologist. March, i8£9 
line rocks which at some depth in the drift are fresh and hard, 
thoroughly disintegrated at the contact between the Kansan 
and the loess. Often they may be crushed between the fingers. 
It is not very unusual to find rotted bowlders in any por- 
tion of the older drifts as would be suspected from the fact 
that the older drift sheets had much rotted material to work 
with. But it is frecjuently. possible to prove that the disinte- 
gration of the bowlders at the surface of the drift took place 
after they had been shaped by the ice and furthermore, the 
number of rotted bowlders usually increases with the nearness 
to the surface. Accordingly this phenomenon accords with 
the decalcification and the development of ferretto in indicat- 
ing a period of considerable exposure to weathering agencies. 
The extra-morainic drift of Carroll county occurs in two 
facies; one, the normal Kan'san, showing ferretto, leaching, 
rotted bowlders, etc., and the other an abnormal type in which 
these phenomena are lacking. The outcrops of these two 
phases are inextricably mixed. Each occurs in areas wholly 
surrounded by ordinary, simple erosion. The field relations 
indicate clearly that they are but different phases of the one 
drift. The abnormal phase of the drift shows no trace of fer- 
retto and is wholly unleached. It effervesces up to the loess 
contact when tested with acid. It contains more rotted bowl- 
ders but they are not usually more abundant at the top than 
the bottom. In many particulars it is indistinguishable from 
a voung drift and yet the most careful search in the field has so 
far failed to reveal any dividing line either vertical or horizon- 
tal^ between it and the normal Kansan. If there were two 
drift sheets in the region, the one fresh and unleached and the 
other old and ferretto-covered the younger drift could hardly 
have the patchy geographical distribution necessitated by the 
facts in the present case except upon the hypothesis of its be- 
ing thin and much eroded. Single exposures of more than 
thirty feet are, however, known and there is no evidence 
whatever of more than one period of erosion between the drift 
and the loess. No cases of superposition have been detected 
nor are there accounts of forest beds, buried loess sheets or 
other evidences of an interglacial period. Both sorts of drift 
have exactly the same relations to the loess which in turn 
shows no evidence of being anything exce])t a homogeneous 
