258 The American Geologist. April, 1895 
Witliout attempting to solve the puzzle, we may only add 
the reason for the assertion that the ice observed beneath the 
drift and mammoth remains b}' Dall, and by others before 
him, on the shore of Esehscholtz bay in Alaska,* and by 
Baron Toll on Lyakhoft' island of the New Siberia group, f is 
due to the accumulation of neve snows and their consolidation 
into ice, rather than to the freezing of shallow lakes or infil- 
trating surface waters, as Russell, Dall, and Howorth have 
supposed. Russell's explanations may probably be true for 
many parts of the tundras of Alaska and Siberia; but in the 
two localities noted, where abundant animal remains occur in 
the drift above the ice, the structure of the ice itself proves 
it to be remnants of ancient ice-sheets. 
The researches of Grad, Klocke, Forel, and others, have 
shown that all glacier ice preserves the granular structure 
which has its beginning in the change of the neve from snow 
to ice. Lake ice, on the other hand, has a prismatic or co- 
lumnar structure. The glacier granules, var^'ing in size and 
occasionally so large as from one to two or three inches in di- 
ameter, are irregular or without parallelism in the directions 
of their crystalline axes; but the prisms of lake ice are 
perpendicular to the freezing surface. Glacier ice in the pro- 
cess of melting reveals its granular condition, and this is dis- 
tinctly noted by both Dall and Toll. 
Writing of the ice-clitfs of F^schscholtz bay in northwestern 
Alaska, Dall remarks : 
The ice in "ienei-al had a st-mi-st rati tied appearance, as if it still re- 
tained the iiorizoutal phine in which it originally congealed. The sur- 
face was always soiled by dirty water from the earth above. This dirt 
was, however, merely superficial. The outer inch or two of the ice 
seemed graiuilar. like comi)acted hail, and was sometimes whitish. 
On the New Siberia islands, about 1,400 miles west-north- 
west from the Alaskan ice-cliffs, the same structure is noted 
{Nature, Jan. 31. 1895): 
The chief geological result is the settling of the real positions of the 
layers which contain relics of the mammoth. They are undoubtedly 
Post-Ulacial, as they overlie the masses of underground ice which form 
*Am. .lour. Sci., III. vol. xxi. i)p. 104-111. Feb., 1881; U. S. Geol. Sur- 
vey, Bulletin 84, 1802, pp. :2(IO-"20S. (Compare comments by N. H. AVin- 
ch'ell, Am. .lour. Sci., Ill, vol. xxr, pji. :!.")8-a60. May, 1881.") 
fGeol. Magazine. III. vol. x. pp. 107-111, March, 1893: Nature, vol.i.i, 
p. ;527. .Ian. :!1. IS!!.'). 
