THE DRIFTLESS REGION OF WISCONSIN 833 
der beds. The general slope of the beds is only about one- 
third that of the ravine, so that they soon disappear from view, 
but further up the ravine and about thirty feet above the calcu- 
lated horizon of the highest of the lower beds another one 
occurs about four feet thick, the parting being all clay. There 
appears to be another bed still further up. 
In all the sections examined the extreme abruptness of the 
transition is noteworthy, the clay up to the very line of junction 
being absolutely free from stones. 
Owing to the great labor of obtaining sections, and their 
small extent when obtained, it is impossible to answer many 
important questions regarding the form, extent, and relations of 
the beds. 
Section 7 is fairly representative of the aspect of the beds in 
sections parallel to the valley. The relations shown at p were 
worked out carefully, and are of considerable interest. The 
upper bed, which terminates rather abruptly at that point, has a 
matrix composed mainly of sand, while in the lower bed the 
matrix is a compact clay. The transition from one to the other 
is quite abrupt. 
Causes. — Although I should not like to express a final opin- 
ion as yet, I will, nevertheless, say that so far as the facts are 
known they seem to very strongly indicate one agent, while 
equally excluding others. 
The agents capable of transporting material of the weight 
above described are few. Some of these may be dismissed 
with few words. Simple gravity, such as forms the talus of 
the hills, is excluded, since all the typical examples lie far 
outside its range of action. Wave action is also excluded. 
Neither can shore-ice offer any adequate explanation, although 
it can scarcely be doubted that it existed, and certain widely 
scattered bowlders, as well as certain sharply defined small 
pockets of local pebbles occasionally found embedded in the 
loess, may, with great probability, be assigned to this agent. 
Practically there are but two agents which need be considered. 
The one is torrential, the other, glacial action, in either case 
