Israel C. (Israel Cook) Russell.

Glaciers of North America; a reading lesson for students of geography and geology online

. (page 19 of 24)
Online LibraryIsrael C. (Israel Cook) RussellGlaciers of North America; a reading lesson for students of geography and geology → online text (page 19 of 24)
Font size
QR-code for this ebook

transverse to the prism, the weight sunk sharply into the upper face of the prism and a corre-
sponding protrusion appeared below. As the process continued, the protrusion below kept
closely parallel to the indentation above, both widening somewhat until a section of the prism
had been pushed entirely out. Optical examination showed that the optic axis remained par-
allel to itself throughout. The block remained transparent and free from fracture. The
weight appeared to have simply slipped the plates over their neighbors, carrying the adjacent
ones forward with them to some extent by dragging, but not visibly affecting the more
remote ones.

" If the prism be made of square cards and placed on its side, and a transverse force
applied, the result will illustrate the apparent method of movement within the ice crystal in
this last case. If such a prism be p'ressed at right angles to the cards, it will illustrate the
bending of the first case, and if the cards be placed on edge, they will illustrate the effectual
resistance to deformation of the first case. Variations of temperature through 10 were not
found to produce notable differences of result.

" Not to mention other significant points, the investigation seems to warrant the impor-
tant conclusion that ice crystals yield to deforming forces by the sliding or shearing of the
crystalline layers at right angles to the principal axis. No analogy to the motion of a viscous
fluid appeared. Dr. Miigge had previously found a similar method of deformation in other
minerals, including gypsum, stilbite, and vivianite. In respect to its mode of internal motion,
ice is therefore to be classed with these minerals, rather than bodies properly called


regions and in the ice of the lower portions of glaciers, also sustains the
same conclusion. As in the case of the compression of ice, its ability to
stretch under tension is lowered by a decrease of temperature.

Without attempting to review at this time all that has been written
concerning the plasticity of ice, it seems safe to conclude, on the strength
of the experiments and observations just cited, as well as many others
which might be given, that ice under pressure slowly applied can be made
to flow. That ice is truly plastic, however, is not sustained by the experi-
ments. The question how much pressure would be required to produce
the flow observed in glaciers still remains. This is a more difficult
question, and perhaps does not admit of precise determination.

The conclusion reached by Forbes and others, that glaciers flow by
reason of their own weight, that is, ice is sufficiently plastic to shear under
the influence of gravity, has, in the opinion of many physicists, been
demonstrated by the bending of planks of ice when supported at their
extremities. That planks of ice will bend under such conditions has
been shown by experiments, and also by observing the behavior of blocks
of ice in Arctic regions, which slowly sag when supported at opposite
edges, even when the temperature is continuously far below freezing. In
the bending of a plank of ice under its own weight, there must of neces-
sity be a compression of the material on the upper side and an extension
on the lower side. In other words, the behavior of ice under the influence
of gravity is similar to that of plastic bodies under like conditions. 1

Certain experiments cited by Henry Moseley, however, have been
claimed to show that the weight of an ice mass is insufficient to cause it
to change its shape, or shear, in the manner in which many plastic solids
are known to do under the influence of their own weight. But in the
experiments referred to the difference in the behavior of ice under pres-
sure when slowly applied and when applied with comparative rapidity is,
in the opinion of several competent judges, not sufficiently recognized.
As shown by Moseley, the force necessary to shear a column of ice one
square inch in area of cross section, when applied in the course of 15
or 20 minutes, is about 75 pounds, or 34 times the computed portion of
gravity available in producing glacial flow, and, therefore, these glaciers
cannot move by reason of their own weight.

Moseley's experiments have been repeated by other physicists, and also,
in part, by the present writer, with essentially the same results, so far as

1 This conclusion is held by William Mathews and others. Philosophical Magazine, vol.
212, 1871, pp. 332-334.


the amount of force required to shear ice under the conditions observed
in the experiments are concerned. That the conclusions reached from
these experiments can be applied in explaining glacier motion cannot be
so readily accepted.

The experiments made by Moseley have been discussed by J. Ball 1
and others, who have shown that the time element is important, and that
there are reasons for doubting if the results reached can be applied in
explanation of the flow of glaciers.

Although the conclusion that ice will flow under the influence of its
own weight may not be established to the satisfaction of all who are inter-
ested in the problem, yet if applied provisionally to glaciers it is found to
explain many of their movements, and enables one to predict what will
occur under given conditions.

By comparing the movements of glaciers with the movements of pitch
and other similar bodies under the influence of gravity, striking analogies
m&y be obtained. If pitch of the proper consistency be placed in a gently
inclined trough, having so far as practicable the proportions of a represen-
tative glacier-filled valley, it will slowly descend in the same manner that
glaciers move, even though it is extremely brittle at the temperature at
which the experiment is conducted. The flow of the pitch is most rapid
in the central part of the surface of the stream, but decreases gradually in
rate of flow towards the sides and bottom. If two streams of pitch be
made to unite so as to form a trunk stream, representing a compound gla-
cier, lines of debris on the borders of the tributaries, representing lateral
moraines, will unite and form a " medial moraine." Where inequalities
in the bottom of the trough exist, crevasses will appear in the pitch, etc.
In this and other ways the flow of a glacier may be shown to correspond
in a most striking manner with the flow of a plastic substance which is
urged forward solely by the influence of its own weight. Strong as are
the arguments tending to show that glaciers are urged forward in the same
manner that truly plastic substances flow under the influence of gravity,
this explanation has not been unanimously accepted.

Many forcible objections to the hypothesis of plasticity have recently
been advanced by T. C. Chamberlin. 2 These objections, however, are
based principally on observations made on the glaciers of Greenland, where
the generally low temperature may be considered as reducing the plasticity

x "0n the Cause of the Motion of Glaciers," Philosophical Magazine, vol. 42, 1871,
pp. 81-87.

2 "Kecent Glacial Studies in Greenland," Bull. Geol. Soc. Am., vol. 6, 1894, pp. 199-220.


of ice to its lowest limit as exhibited in glaciers. Whether this condition,
however, may render the observations less valuable for determining the
existence of plasticity, or make the test more searching, might be differently

Chamberlin states that his observations seem to be adverse to anything
which can be termed viscous fluency. In some instances the surfaces of
glaciers were found to rise in the direction of the motion of the ice, so
that surface streams flowed backward. Similar changes in surface slope
were observed by the present writer in several instances in the neves of
Alaska, and is evidently not exceptional. This phenomenon may, how-
ever, not be opposed to the hypothesis of plasticity, since the energy which
urges forward a given molecule within a glacier is the resultant pressure
of molecules at higher levels. It is principally the surface gradient that
determines the rate of flow. The formation of elevations analogous to
anticlinal folds may be due to the pressure of ice at higher elevations,
acting as a thrust on the edges of layers, the cohesion of which is sufficient
to allow them to bend upward and reverse the normal surface gradient.

The breaking of glacial ice under moderate and slowly acting tension
was also observed by Chamberlin, who concludes, as others have done,
that if the ice could stretch even in a slight degree, crevasses would in many
instances be avoided in situations where they are found in abundance.
As the ability of ice to stretch decreases with temperature, it is to be ex-
pected that in the far north .the conditions are unfavorable for the study
of such phenomena. On the whole, the observations thus far made on the
breaking of glacial ice and the formation of crevasses, do not seem to
controvert the results of experiments which show that ice does yield to
slowly applied tension, without rupture. What has been shown by gla-
cialists in many countries, is that the limits to which glacial ice can yield
to tension under certain conditions is frequently exceeded. It was also
noticed that boulders resting on the glaciers in Greenland, or inclosed
within them, showed no tendency to descend through the ice as heavy
substances descend through viscous bodies. This, as is well known, is
true of boulders carried by glaciers in all countries, and offers an objection
to the hypothesis of plasticity that cannot be easily removed.

Everywhere, as stated by Chamberlin, the ice of the Greenland glaciers
appeared to behave as a rigid rather than as a plastic substance. The
rigidity did not prevent contortions and foldings of the laminated ice, but
faults and vein structure also occurred, and there seemed to be no more
occasion to assume plasticity in the one case than in the other.


The same author also remarks " that there is a theoretical objection to
the assumption of viscous flowage in the very fact of crystallization itself.
The property of viscous flowage rests upon the relative indifference of a
particle as to its special point of adhesion to its neighbor particles. The
property of crystallization rests upon the strongest preferences respecting
such relationships. Particles of water in their fluid condition lie against
and cohere to each other indifferently. When they take on a crystalline
form they arrange themselves in specific relationships by the exercise of a
force of the highest order. In the presence of this very forceful disposi-
tion of the particles to retain fixed relationships to each other, it would
seem little less than a contradiction of terms to attribute to them viscous
flowage. The crystalline body may readily be made to change its form by
the removal of particles from one portion by melting and their attachment
at other points by congelation, but not, I think, by the flowing of crys-
tallized particles over each other while in their crystallized condition."

While some of the objections to the hypothesis of plasticity advanced
by Chamberlin are at present unanswerable, and his general conclusion in
reference to the rigidity of the ice at the north of much weight, yet the
theoretical considerations just quoted would seem to be more than counter-
balanced by the experiments which show that ice composed of many
crystals does yield continuously without fracture both to compression and
tension. If a slab of ice supported at its ends does gradually sag under
the influence of its own weight, simply, and at temperatures that do not
admit of melting and refreezing, it seems unnecessary to argue that on
account of its crystalline structure it is impossible for it so to yield.

The discussion that has been carried on for half a century respecting
the hypothesis of plasticity has been ably advocated on each side, and
some of the arguments against it remain unanswered ; but to-day many able
investigators, and especially many of those who are familiar with glaciers
from actual contact with them, hold that it more nearly meets the re-
quired conditions than any other hypothesis that has been proposed. That
it is not a complete and sufficient explanation of all the phenomena asso-
ciated with the flow of glaciers, however, will appear still more forcibly,
I think, from a review of other explanations that have been proposed.

The Hypothesis of Reg-elation. It is now well known, thanks to
Faraday, Tyndall, and others, that when two pieces of ice having a tem-
perature of about 32 F. are brought in contact they freeze together.
This property, now termed regelation, was studied especially by Tyndall,


and by him first used in attempting to explain glacier motion. Under the
hypothesis of regelation the ice of glaciers is thought to be crushed and
the fragments reunited by refreezing after a change of position. "It is
easy, therefore," says Tyndall, "to understand how a substance so endowed
can be squeezed through the gorges of the Alps, can bend so as to accom-
modate itself to the flexures of the Alpine valleys, and can permit a dif-
ferential motion of its parts without at the same time possessing a single
trace of viscosity."

In illustration of the process of regelation numerous experiments have
been made by placing fragments of ice in moulds of various forms, and
subjecting them to pressure. When thus treated the ice is crushed and
the fragments move past each other so as to take new positions, and are
thus adjusted to the shape of the cavity containing them, but freeze to-
gether in their new positions and form a solid body. In this manner ice
has been made to assume almost any desired shape. When the pressure
is slowly applied rude fracture is avoided and the ice changes its shape in
apparently the same manner as many plastic substances would if experi-
mented with in a similar way.

In applying the principle of regelation to account for the flow of
glaciers, it is assumed that the ice is crushed and that the fragments are
made to move past each other and are refrozen in new positions. That
rude fractures may be healed by regelation is abundantly attested. When
a glacier passes down a steep descent it is greatly crevassed, but below
such an ice fall the fissures frequently close, their walls freeze together,
and the ice is possibly even more compact and homogeneous than before it
was fractured. The conclusion, however, that the characteristic flow of
glacial ice is accomplished in the same manner, but by incipient fractures,
has been seriously questioned.

In the hypothesis of regelation, as in the hypothesis of plasticity, the
force which causes motion is assumed to be the weight of the ice. Instead
of floAving as a plastic substance, however, the ice is considered as behav-
ing as a brittle substance, under the conditions to which it is subjected,
and as being crushed and having the fragments reunited by freezing after
a change in their relative positions. In all of the experiments that have
been made to illustrate the process of regelation a force greater than the
weight of the ice experimented on has been applied. Much discussion has
been carried on in reference to the regelation of ice once fractured and
having its fragments brought in contact at the proper temperature ; but
little has been said, however, in reference to the manner in which glaciers


might be crushed so as to make regelation possible. As has been shown
by Moseley, a pressure of about 75 pounds per square inch is necessary
to shear ice if applied with comparative rapidity. Although it seems
impossible to apply these experiments in a quantitative way in explain-
ing the movement of glaciers, they indicate that certain general con-
clusions may be valid. From the experiments referred to it has been
computed that a column of ice in order to begin to crush at its base
would have to be over 700 feet high. Evidently, then, glacier ice
cannot be crushed under its own weight unless at least 700 feet thick,
and then the fracturing would be confined to the bottom layer ; we should,
therefore, under the hypothesis of regelation, expect the greatest freedom
of movement to occur in the basal position of a glacier. Yet, as is well
known, the maximum movement is at the surface. How, then, can the
principle of regelation be applied in explaining the surface flow, especially
of a glacier with a low surface gradient ? Again, regelation takes place at
a temperature of about 32 F., and cannot occur much below that temper-
ature unless the ice is under pressure. The rate at which the melting-
point of ice is lowered by pressure is so. small that practically it may be
ignored in this discussion. Besides, the rate of surface flow of a glacier is
greater than the rate below the surface, even in winter, when the tempera-
ture of the ice is frequently far below the point where regelation is possible.
It seems, therefore, that the regelation hypothesis fails to meet several
important features of the problem of glacier motion.

The principle of regelation is not to be entirely discarded in seeking
an explanation of the behavior of glaciers, however, as the healing of
fractures, as already noticed, may be satisfactorily explained' in this
manner. The principle of regelation apparently assists one in under-
standing how the granular snow of neves becomes consolidated under
pressure into compact ice. As a neve becomes deeper and deeper, the
granules of which it is composed, but which originate and increase in
size from other causes, are brought in contact at the proper temperature
and freeze together.' The granules formed from light, porous snow may
by this process be converted into compact ice.

It will undoubtedly occur to the reader that the question whether a
glacier flows by reason of its plasticity or on account of fracture and rege-
lation, could be decided by a study of the intimate structure of the ice of
which it is composed. From a geological point of view glacier ice may be
considered as a "rock" and investigated by petrographical methods. That
is, it may be cut into thin sections and examined by means of the micro-



scope and polariscope. As already described, glacier ice has a peculiar
grain, which is frequently so pronounced and characteristic that even a
small fragment may in many instances be readily distinguished even by
the unaided eye from lake and other ice. If the flow of glaciers is due
to plasticity one would expect that the grain of the ice would exhibit
something of a fibrous structure, similar perhaps in a general way to the
structure of wrought iron, or to the structure of certain schistose rocks
which have passed through a plastic condition. Nothing resembling this
structure, however, is revealed in the grain of glacier ice. The struc-
ture of a characteristic sample of glacier ice, when examined by means
of a polariscope, is shown in the accompanying figures, 1 one of which


A two-thirds natural size. The section is vertical and at right angles to the direction of flow.
B natural size. The section is vertical and parallel with the direction of flow.

exhibits the appearance of a thin section cut parallel to, and the other at
right angles to, the direction of flow. Although the grains in the section
parallel with the direction of flow are perhaps slightly flattened, nothing
resembling a fibrous structure or a marked elongation of the granules is

Experiments by A. Heim, 2 have shown that the peculiar grain of glacier
ice is accurately imitated when ordinary lake ice is crushed and again con-
solidated by regelation. So far as the study of the intimate structure of
glacier ice bears on the explanations of glacier motion -already considered,

1 These diagrams are copied from a paper on " The Structure of Glacier Ice and its Bearing
upon Glacier Motion," by R. M. Deeley and George Fletcher, Geological Magazine (London),
Decade 4, vol. 2, 1895, pp. 152-162.

2 "On Glaciers,'' Philosophical Magazine, vol. 41, 1871, pp. 485-508. Translated from
Poggendorff s Annalen, Erganzungsband, 1870, pp. 30-63.


it favors the hypothesis of regelation rather than that of plasticity. Yet,
the observed uniformity in the size of granules composing glacier ice at
various localities and their gradual increase in size from near the source
of a glacier but below the lower limit of the neve to its extremity, are not
accounted for on the supposition that continual crushing and refreezing
take place.

The Hypothesis of Expansion and Contraction. Geologists are
familiar with the fact that talus slopes, as the piles of loose rock fragments
at the bases of steep escarpments are termed, experience a slow down-
ward creep, due to the alternate expansion and contraction of the frag-
ments composing them, with changes of temperature assisted by gravity.
In a similar way sheets of lead, as observed by Moseley, will slowly creep
down an inclined surface when exposed to variations of temperature.
Glacier ice is exposed to changes of temperature and subject to similar
variations in volume, but owing to the fact that the same change of tem-
perature will produce greater changes in ice than in rocks or lead, that is,
owing to its greater coefficient of expansion, 1 its movements under the
same fluctuations of temperature will be greater.

It is claimed by the advocates of the hypothesis under review, that in
the case of an alpine glacier, for example, the ice in alternately contract-
ing and expanding will slowly creep down a valley, since movement in
that direction is assisted by gravity and in the opposite direction is opposed
to gravity. The same argument has been applied, also, to continental
glaciers originating on a plain and flowing in all directions from a
center of accumulation, since on account of the rise of the surface
gradient from the periphery toward the center of the mass, the weight
of ice acting on any point in the glacier is greater in one direction than
in others.

This hypothesis of alternate dilation and contraction was advanced
by Moseley 2 and sustained by strong arguments and suggestive experi-
ments, but has been severely criticised by Ball 3 and others. Among the
objections suggested in reference to it are the following :

1 The coefficient of expansion of ice is nearly twice that of lead, and more than twice that
of any other solid.

2 " On the Motion of a Plate of Metal on an Inclined Plane, when Dilated and Contracted ;
and on the Descent of Glaciers," Philosophical Magazine, Fourth Series, vol. 23, 1862, pp.

3 " On the Cause of the Descent of Glaciers," Philosophical Magazine, Fourth Series, vol.
40, 1870, pp. 1-10.


A glacier lying in a high-grade mountain valley or flowing from a
center of accumulation on a plain, would, if it experienced changes of
temperature, alternately contract and expand, and these changes in
volume should produce a resultant motion in the direction of least
resistance. The direction of least resistance at nearly all points in a
glacier is upward, hence in general the movements in a glacier result-
ing from contraction and expansion would be in a direction normal to
the surface of the ice.

The changes of temperature which might be expected to cause a
glacier to " creep " are such as affect it below the melting-point of ice, for
if raised above that temperature it will melt. Observations have shown
that the internal temperature of glaciers is uniformly 32 F., but extended
measurements in this connection, especially in winter, are wanting. We
know, however, that so long as the interstices of ice are occupied by water
the temperature of the mass cannot vary sensibly from that just stated,
the effect of pressure being disregarded ; and as glaciers, at least in tem-
perate latitudes, are as a rule saturated with water in summer, they must
have a uniform temperature at that season of 32 F. In winter the tem-
perature of the air above a glacier may fall far below freezing, and if such
a change should be continued long enough the temperature of the entire
glacier would be correspondingly lowered.

With the above considerations in mind it is evident that under the
" creeping hypothesis " the rate of flow of a glacier should be greater
in winter than in summer, and should also be more rapid by night

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 19 21 22 23 24

Online LibraryIsrael C. (Israel Cook) RussellGlaciers of North America; a reading lesson for students of geography and geology → online text (page 19 of 24)