252 
viously smooth and uncrevassed; the lower 
portion of the glacier was greatly thick- 
ened; where unconfined between mountain 
walls there was a notable spreading at the 
margins; and the free ends of the glaciers 
were bodily moved forward. In all cases 
the transformation was rapid and even 
spasmodie, requiring a period of but a few 
months for the complete cycle; and in all 
cases the advance was quickly followed by 
relapse into the previous state. In other 
words, a wave spread down through the 
glaciers with accompanying thickening, 
spreading, advance and breaking of the 
rigid upper ice; but after passage the gla- 
cier was left in essentially the same state 
of activity as before, even though that state 
had been complete stagnation in parts of 
the affected area. 
In some cases the wave spent its effects 
in breaking, thickening and spreading a 
piedmont bulb, with little actual advance; 
in others, the effects of the thrust being 
confined by bordering mountain walls, and 
thereby concentrated on the frontal end, 
there was notable advance of the terminus. 
Such an advance is best illustrated in the 
Hidden Glacier, whose front was pushed 
forward about two miles; and where the 
ice front stood in 1906 the glacier was 
1,100 feet thick after the advance. During 
a brief, spasmodic advance, at least a third 
of a cubic mile of ice moved beyond the 
1906 front; and great volumes of ice were 
also added to the glacier back of the old 
front, for in 1909 the glacier surface rose 
to a far greater height than in 1905 and 
1906. 
The theory put forward to account for 
this series of glacier advances is that the 
vigorous earthquakes of September, 1899 
shook down such great avalances of snow, 
ice and rock in the glacier reservoirs as to 
necessitate a wave of advance that swept 
down through the glaciers, reaching the 
SCIENCE 
[N.S. Vou. XXXV. No. 894 
terminus of the smaller ice tongues very 
quickly, and the larger ones at later dates, 
while up to the period of our last observa- 
tions, in 1910, the very largest glaciers had 
not yet responded. Since the cause was a 
sudden and concentrated addition of large 
supplies to the glacier reservoirs, the re- 
sulting wave was naturally rapid in its 
passage, and it quickly subsided, while its 
effects in passing were both spasmodic and 
extreme. 
A study of four seasons discovers only 
evidence favoring this theory, and since it 
is an efficient cause, known to have been 
actually present, while no facts are known 
to oppose it and a great number favor it, 
I feel convinced that the earthquake aval- 
anche theory merits the wide acceptance 
that it has received. It adds a new, and, 
in favorable regions, probably a very im- 
portant cause for fluctuations in glacier 
margins. How widely it may be extended 
in explanation of other glacier advances re- 
mains to be established by future studies; 
it is not to be expected that it will replace 
the theory of climatic cause for glacier 
fluctuations; but it may well be expected to 
supplement it and perhaps in part replace 
it in regions of frequent earthquakes. 
Local Nature of Recent Great Advances 
It is too early to attempt to explain all 
the known variations in Alaskan glaciers, 
for as yet the body of fact is limited both 
as to time and as to area. Yet there are 
some significant features that are well 
worthy of consideration. Attention has al- 
ready been directed to the fact that there 
has been a very great recent recession of the 
ice fronts in Glacier Bay and a similar re- 
cession in the Yakutat Bay region 150 miles 
to the northwest. This recession, which 
has been in progress for the past century 
or more, is really but part of a cycle in 
which the glaciers are still receding toward 
