1880.] 469 [Stone. 
must have been far deeper yet. In general during the kame period 
the thickness of the ice was three hundred feet or less. 
RELATIONS BETWEEN THE KAMES AND THE Upper TILL. 
1. They were contemporaneous formations. 
2. Both were derived from the same morainal matter scattered 
through the ice, the difference being that the kames have been classi- 
fied by water. Yet kames sometimes contain large unclassified 
masses of upper till which probably slid down into the bed of the 
stream and defied classification by the water. The word classification 
is not here used as synonymous with stratification. The pell-mell 
kame has been classified by water but is not now stratified. 
3. A part of both formations were glaciated, but in the kame the 
glaciation has been more or less effaced by water wear. 
How wERE THE MATERIALS OF THE KAMES COLLECTED ? 
Evidently a part was already in the ice at the place where the 
channel of the kame river was formed, and was laid bare as the ice 
melted or was eroded. ‘This will account for the presence of many 
of the large pebbles in the kames. Another part was brought to the 
main channel by tributaries from the sides. This was composed for 
the most part of fine materials. Occasionally small gravel beds are 
found not far from the kames, which I have thought might be the 
gravels of such tributaries, but their fewness is surprising. ‘The nar- 
rowness of the channels of many of the kame-rivers as compared with 
their depth has often made me suspect that these channels were so 
deep that the sides flowed inward toward each other, the stream 
eroding the ice as it advanced. The flow of the ice as a whole must, 
over a large part of the state, have ceased at the time of the depo- 
sition of the kames, and such a flow of the walls inward could not 
have been very great in amount. If such a flow occurred, it would 
help to bring together kame material. 
