244 The American Geologist. April, iso;? 
diameter. These boulders the author thought to have been swept 
down the Mattawan valley by a powerful river outflowing from 
lake Huron, as the debacle of discharge from lake Bonneville 
brought a similar multitude of boulders to Pocatello, where it 
entered the broad Snake river valley. 
In discussion, President Gilbert stated that he had examined 
this divide and the course of the Mattawan to lake Talon, finding 
the avenue of drainage to consist largely of boulders, with scared}- 
any till or stratified beds. 
Dr. Bell thought that all the reported observations fall far 
short of proving an outlet from Lake Huron to the Ottawa. The 
tract of so plentiful boulders at the mouth of the Mattawan he 
<jonsidered as part of a morainic belt. 
On certain features in the distrihution of the Columbia formation 
on the middle Atlantic slope. By N. H. Darton. This paper, 
read in the absence of the author by Mr. McGee, directed atten- 
tion to evidences of an interval of erosion between the two divis- 
ions of the Columbia formation in Marjland and northward to 
New Jersey. The lower division comprises beds of loam and fine 
and coarse gravel, all of which enclose occasional ice-borne boul- 
ders; and the upper division, more rarely- containing boulders, con- 
sists chiefl}' of loam or clay, resembling loess, to which the name 
Philadelphia brick cla}- was given by Lewis. Both parts are re- 
ferred to estuarine deposition, and the interval of emergence and 
erosion dividing them is thought to correspond to the inter-glacial 
epoch and forest bed of Iowa between deposits of till, which ap- 
pear correllative with these lower and upper parts of the Colum- 
bia formation. 
Mr. Upham, in discussion, thought the Columbia gravel and 
loam to be not marine or estuarine sediments, but the deposits of 
river floods bearing ice-rafts when the land in the earl}' part of the 
{jlacial period was higher than now, with much greater suppl}' of 
detritus and of water from rains and snow melting each spring 
^nd summer. 
Note on the geology of Middleton Island., Alaska. By George 
M. Dawson. (The author being absent, this paper was read b}- 
Mr. R. W. Ells.) On Middleton Island, situated near the verge 
of the submarine continental plateau, about 50 miles distant from 
the coast, typical till is found, enclosing abundant argillite frag- 
jnents and broken marine shells. A microscopic examination also 
