142 The American Geologist. March, 1903. 
gion, as evidence is not wanting that its cle]>osition immedi- 
ately preceded glaciation. 
Overlying the gravel is a heterogeneous agglomeration of 
small boulders, cobbles and pebbles embedded in a rather stiff 
clay. It is not distinctly stratified and does not show the sort- 
ing action of water; nor is its constitution of the local charac- 
ter of a landslide deposit. I was at first unwilling to recognize 
it as a glacial deposit, although Mr. Maitland had inforp.ied 
me that Mr. E. B. Preston of the State Alining Bureau had 
so declared it years ago. A search soon presented luimistake- 
able evidence of glacial action. Many of the included pebljles 
and small boulders are of serpentine, a rock peculiarly favor- 
able to the preservation of glacial scratches. Not a few of 
them are unmistakalily striated and flattened. These scratches 
were not produced in the operation of mining as I made sure 
by digging pebbles from the undisturbed bank and finding 
striae on their banl<ward sides. These stride are seemingly not 
as fresh appearing as usual in the glacial deposits which I 
have formerly described from these mountains, but the der- 
ivation of this idea may have been somewhat influenced by 
other considerations. 
The constitution of the deposit shows that tlic glacier 
which formed it did not come down the Coffee creek valley, 
because in that case the till should be abundantly supplied with 
hornblende schist, the predominating formation of the main 
valley, whereas the included rock fragments are chiefly of ser- 
pentine, the vastly predominating formation of the Union 
creek valley. At the junction of the two valleys there is a pro- 
jecting narrow point of mica schist. The Union creek glaciej" 
over-rode this ridge and plunged down into the narrow val- 
ley of Coffee creek. The ice abraded the upper portion of the 
creek gravels and incorporated them into its own peculiar till. 
In this manner some of the Courtney granite boulders be- 
came included in the glacial deposit. Near the junction be- 
tween the creek gravel and the till, the former is much broken 
up and at only a few places is there a sharp line between them. 
The till varies in thickness from five to fifty feet, and has 
been piled up into a slight moraine marking the former extent 
of the glacier. In fact, it is the end of a discontinuous lateral 
moraine. Near the base the till preserves its original difll 
