REPORT OF THE DIRECTOR, 1922 oe alos 
have been to the south and over the belt of stagnant ice which 
lay in the Utica-Little Falls section of the [Mohawk] valley.” 
Aside from these and a few other scattered references which 
might be quoted, the literature indicates how slight an impres- 
sion the evidence has made upon the ciassical assumption of a 
retreating wall of ice. 
Finally: if everything in New England and eastern New York 
which has been, rightly or wrongly, thought by some author to 
be an ice-front recessional moraine were mapped together with 
the recessional moraines of the Erie Lobe on any reasonable 
scale (say 20 miles to the inch), the comparison would still be a 
striking contrast. 
Lacking the direct evidence of recessional moraines, some 
indirect evidence ought to be furnished by an application of the 
ice-front hypothesis to interference with the normal land drain- 
age; there should be found: the records of ponded waters with 
lowering outlets at predictable points over cols or across land 
salients as the assumed ice front withdrew. In critical localities 
this hypothesis breaks down conspicuously, for example, in the 
following north-sloping hydrographic basins: Wallkill River, 
Shawangunk Creek, Rondout Creek, Moodna Kill, (Woodbury 
Creek branch) and Schoharie Creek. 
On the other hand the evidence of stagnant ice melting im siiu 
is widespread and abundant. Its expression is so varied that 
generalizations are difficult, while detailed descriptions would 
expand this paper far beyond the limits contemplated for it. The 
statements which follow should be accompanied by a running 
commentary on the published conclusions of others who have 
investigated the several fields mentioned, but this also is ren- 
dered impossible by space limitations. 
As has been long known, the Adirondack mountain mass 
appeared above the lowering surface of the ice sheet as an 
island. On the principle that the comparatively clean upper ice 
melted more rapidly than the basal ice which carried the bulk 
of the drift, the earlier stages of ablation tended to produce an 
irregular surface roughly corresponding to the upper limit of the 
portion which was heavily drift-laden. There appears to be little 
to suggest the action of meteoric waters, and the movement and 
redistribution of the drift as it was released from the frozen mat- 
rix was accomplished almost wholly by the waters-of-melting, 
the higher uncovered basal ice draining down onto the lower ice 
