Glacial Geology in America.— Fairchild. i8i 
contains considerable matter upon the subject. Down to 
1857 the most voluminous writer was Edward Hitchcock. As 
early as 1833 he explained the Connecticut river terraces by a 
down-cutting of the river through beds deposited in higher 
stages of flood. In his paper of 1857 he recognized the com- 
plexity of forces and thought that the valley terraces were 
formed in different ways; those of the Connecticut valley 
chiefly by a slow lifting of the land with local changes or shift- 
ing of the streams. The recognition of the valley drift of New 
England as derived from glacial debris and deposited by glacial 
floods was made by Dana in 1855. As early as 1858 M. Tuo- 
mey suggested that the Mississippi valley drift was deposited 
by floods from the sudden melting of the northern glaciers, 
and in 1859 E. B. Andrews correlated the terraces of the 
southern Ohio valley with the glacial drift. 
Concerning details of the terrace formation in New Eng- 
land there have been divergence and changes of opinion. In 
the earlier editions of his Manual Dana held that the terrace 
drift was accumulated during a time of land depression and 
slack drainage, and the terraces excavated during pauses in 
the reelevation of the land. But in the edition of 1879 he 
admitted that the height of the upper terraces marked the 
height of the glacial flood, and that change of land altitude was 
not essential; thus granting the early contention of Hitchcock. 
Loess. — The resemblance of certain superficial deposits 
throughout the interior portions of the United States to the 
"Loess" of Germany and of China was recognized early in 
this century, but the writer is not certain of the earliest sug- 
gestion of connection of the deposit with glacial phenomena. 
Such suggestion was made as early, at least, as 1866, by Whit- 
tlesey, who attributed the "loess-like" deposits of Illinois, 
southern Iowa and Missouri to floods from the melting ice 
sheet. From that date to the present the numerous writers 
upon the loess have been almost unanimous in regarding the 
American deposits as aqueous, and as having some relation 
to glacial conditions, although there were differences of opin- 
ion as to the precise conditions of deposition. One notable 
exception to this view was published in 1879 by R. Pumpelly 
advocating the eolian origin of nmch of the Mississippian 
loess. A paper by J. E. Todd before this association in 1878 
