Copyright
Israel C. (Israel Cook) Russell.

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

. (page 18 of 24)
Online LibraryIsrael C. (Israel Cook) RussellGlaciers of North America; a reading lesson for students of geography and geology → online text (page 18 of 24)
Font size
QR-code for this ebook


the amount of debris carried, would undergo the changes outlined above at
different rates. If their percentages of debris were the same and other
conditions varied, their periods of halt and advance or retreat would again
vary. So diverse are the conditions controlling the flow of glaciers that
in no two instances would their pulsations be synchronous, even under the
same meteorological environment.

While the greater changes exhibited by glaciers can only be accounted
for by variation in the supply of snow on their neves, or changes in the
rate of melting, or both of these causes combined, due to meteorological
fluctuations, it seems evident that the minor advances and retreats of
their extremities may be due in part to the effects of debris on the flow
and on the melting of the ice as outlined above.



CHAPTER IX.
HOW AND WHY GLACIERS MOVE.

A REVIEW of the various hypotheses that have been advanced to account
for the movements characteristic of glaciers, necessitates an extension of
the geographical limits set for this book, since the critical investigation
of glacial phenomena was well advanced in Europe before the fact that
glaciers existed in North America, with the exception of the remote
Greenland region, was known. It is only in recent years that the detailed
study of existing glaciers has been undertaken in this country. Although
Agassiz adopted America as his home, the investigations that placed him
in the foremost rank of glacialists were carried out in Europe. It is to
the physicists and geologists of Europe that we are indebted for nearly all
that has been done respecting the philosophy of glacial motion.

The Nature of Glacial Flow. Many observations have been made
which show that glaciers have motion, and in probably all instances exhibit
a well-defined flow during some portion of their history. Although this
matter has claimed a large amount of attention, it is important to remem-
ber that glacial ice frequently does not flow, and that probably some part
of every glacier is stagnant.

The fact that glaciers move was first determined in a qualitative way
by noting the changes that take place in the moraines on their surfaces.
Conspicuous boulders resting on the ice were observed to slowly change
their positions from year to year with reference to fixed points on adjacent
cliffs. These crude observations lead to measurements of the rate of flow.
In the case of several glaciers of the alpine type, rows of stakes have been
placed in the ice at right angles to the direction of movement, and their
displacement observed by means of surveying instruments from the adja-
cent banks. In this manner changes in the positions of the stakes have
been measured from day to day, and in some instances even from hour to
hour. Many observations of this nature have shown that in the case of a
river-like glacier, the most convenient variety for study, the maximum
motion is in the central part and decreases toward the borders. It has
also been found that the rate of flow is greatest at the surface, and decreases
toward the bottom. When a glacier follows a sinuous course, the thread



HOW AND WHY GLACIERS MOVE. 161

of maximum current is deflected to the right and left of a medial line, in
the same manner that the swift central current of a winding river is thrown
first against one bank and then against the other ; but the bends in the
sluggish ice current are less abrupt than in the case of the more flexible
water current,

The rate of flow of a glacier varies from locality to locality, through-
out its length, in response to irregularities in the valley it occupies,
unequal distribution of the debris it carries, and other reasons. The rate
of flow also varies with the seasons, being greatest in summer and least in
winter. Similar but much less pronounced variations occur between day
and night. These seasonal and daily changes coincide with variations of
temperature, the rate of flow increasing with an increase in the amount
of heat that reaches a glacier. It is important to note, however, that
although glaciers are sluggish in cold weather, they continue to flow, in
many instances at least, even ill the depth of winter. A flowing motion
in the ice of piedmont and continental glaciers is frequently evident from
the arrangement of debris on their surfaces. The same fact is shown, also,
by the presence of lobes about their margins which many times become
well-defined streams. Although no measurements have been made of the
manner in which the currents in great ice sheets move, there is no reason
for supposing that the causes producing them are of a different nature
from those which urge an alpine glacier through a narrow valley. If an
explanation can be found for the flow of a mountain ice stream, it is evident
that it should explain the movements of other types of glaciers as well.

The movements of a glacier are usually greatly modified by local con-
ditions. In seeking for general laws applicable to all glaciers, it will be
of assistance if we can determine what would be the behavior under normal
conditions of an ideal glacier composed of clear ice, flowing down a straight,
even channel of uniform width and with a uniform gradient both of the
valley and of the glacier's surface. Let us assume also that in cross sec-
tion our ideal glacier has the form of half an ellipse, the division being
along the longer axis. From what is known concerning the behavior of
glaciers we may determine in what manner our ideal ice stream would flow.

All students of glacial phenomena will agree, I think, that in the ideal
example before us the thread of maximum current would be at the center
of the surface, and that the rate of surface flow would decrease uniformly
toward each bank ; also, that the rate of flow would decrease uniformly from
the center of the half-ellipse along any of its radii. The minor axis of
the ellipse would divide a cross section of the glacier into two equal and



162 GLACIERS OF NORTH AMERICA.

similar figures, and similar points in each half would have a corresponding
motion which would differ from the motion of all other points in the cross
section.

Any theory of glacial motion that is applicable to all glaciers should
account for the flow of our ideal glacier, and also explain all changes in
its movements resulting from alterations in the assumed conditions. With
a true and sufficient explanation of glacial motion in mind, we should be
able to predict what modifications in the flow of our ideal glacier would
result : if, for example, its channel was no longer straight ; if changes
should be made in the gradient \of either its bottom or surface ; if the sides
or bottom of its channel should be made uneven ; if its width should
vary ; if the temperature to which it is exposed should change either grad-
ually from its source to its extremity, or fluctuate irregularly ; if the ice
should become unevenly charged with debris ; and in fact any other
changes in environment to which actual glaciers are subject.

The student who has in mind the movements in our ideal glacier, and
attempts to trace the changes that would result from such a combination
of conditions as exert an influence on even the simplest existing alpine
glacier, will no doubt be willing to accept the conclusion reached by gla-
cialists that in any existing glacier, no two points in any cross section, and
in fact no two points in any portion of the ice stream, move at the same
rate for any considerable time. In other words, adjacent molecules or
other small divisions into which ice may be assumed to be divided
throughout a glacier are moving at different rates. 1 That is, the flow of
glacial ice, in a general way, at least, is analogous to the flow of a liquid,

1 Observations by Messrs. R. H. Koch and Fr. Klocke (Philosophical Magazine, Fifth
Series, vol. 9, 1880, pp. 274-277) on the movements of a point in a vertical plane in a glacier
indicate that the movements of such a point are much more complex than has been generally
supposed. In the observations referred to it was found that a given point at one time moved
toward the source of the glacier, and at another time toward its extremity. Two adjacent
points were found to move in opposite directions at the same time. The maximum move-
ments occurred in the forenoon, beginning with the irridation of the glacier by the sun.
These morning movements were irregular, but on the whole tended down the valley ; while
at night the resultant in the direction of movement was toward the mountains.

So far as I am aware, these delicate observations have not been repeated, and conclusions
based on them may not apply to glaciers in general.

Messrs. Koch and Klocke do not state what precautions were taken to shield their
instrument from changes of temperature, and it is possible that their observations are in error
from this cause. The difficulty of keeping a transit or other similar instrument in adjust-
ment, when exposed to changes of temperature, makes it desirable that the observations
referred to should be repeated with an instrument so arranged, perhaps with a prism, that a
fixed point on land can be seen at the same time that the movements of a point on a glacier
are measured.



HOW AND WHY GLACIERS MOVE. 163

or more nearly to the flow of a viscous fluid. It must be remembered that
in the above attempts to describe the flow of an ideal glacier, and to sug-
gest the nature of the changes that would result from modifications in the
conditions to which it is subjected, the aim has been to obtain a graphic
idea of how a glacier flows, without attempting to explain why it flows. In
order to learn if possible the nature and mode of action of the natural
forces which cause ice to move, and at the same time become more familiar
with glacial phenomena in general, let us review briefly some of the
principal explanations that have been offered of glacial motion.

HYPOTHESES or GLACIAL MOTION.*

Several hypotheses in explanation of the characteristic movements of
glaciers have been advanced, but no one of them has thus far met with
general acceptance. It is manifestly the duty of the geologist and geog-
rapher, however, to examine these proposed explanations, and ascertain so
far as is possible how much of each can be safely accepted, even if it does
not afford a sufficient reason for all of the movements known to occur in
large ice masses. By so doing we may, perhaps, clear the way for renewed
study, or possibly be able to frame an eclectic theory from portions of
previous explanations which will be satisfactory.

The Sliding- Hypothesis. As far back as 1760, as stated by Tyndall,
Altman and Griiner proposed the view that glaciers move by sliding over
their beds. Nearly forty years subsequently this notion was revived by
De Saussure, and has therefore frequently been called " De Saussure's
theory," but is more frequently, perhaps, designated as the "sliding
theory" of glacial motion. Subsequently the hypothesis was ably discussed
by W. Hopkins. 2

Under this hypothesis glaciers are supposed to slide bodily down the
valleys they occupy, in obedience to gravity, and grind away the rocks
over which they pass, by means of sand and stones frozen into their under
surfaces.

1 Many reviews of glacial hypotheses have been published ; among those that the student
will find most interesting and instructive are : " The Physical Cause of the Motion of Glaciers,"
in " Climate and Time," by James Croll, first edition, chapters 30, 31 ; " The Great Ice Age,"
by James Geikie, second edition, chapters 3, 4; "Illustrations of the Earth's Surface
Glaciers," by N. S. Shaler and W. M. Davis, chapter 12. The last-named book also contains
a useful list of works on glaciers, published previous to 1881.

2 "On the Motions of Glaciers," Philosophical Magazine, vol. 24, 1895, pp. 607-609 :
Also, "On the Theory of the Motion of Glaciers," ibid., vol. 25, 1863, p. 224.



164 GLACIERS OF NORTH AMERICA.

It is unnecessary to discuss the sliding hypothesis, since, as is now
unanimously conceded, the normal movements to be observed in glacial
ice are of the nature of an onward flow, accomplished by the mutual dis-
placement of molecules of ice. It is of interest to note, however, that
this early and now totally rejected hypothesis does contain an element
of truth.

When glaciers descend steep slopes they become broken, and even
large masses sometimes move short distances by sliding bodily downward.
This, it will be understood, is an exception to the normal movements
characteristic of glaciers, and is referred to simply to show that even the
crudest of the hypotheses we are considering contains a grain of truth.

The Hypothesis of Dilatation. As the movements of glaciers
became better known, the sliding hypothesis just referred to was sup-
planted by the "hypothesis of dilatation," advocated especially by Char-
pentier and Agassiz. The basis of this proposed explanation is that water
on freezing expands and, if confined, will exert a pressure on the walls
retaining it. As water penetrates freely into a glacier through fissures
and capillary passages, it was concluded that it would freeze in such situ-
ations, and thus exert a pressure on the ice containing it, and that a
movement of the ice would thus originate which would take the direction
of least resistance ; and, as alpine glaciers were alone considered, it was
concluded that the direction of least resistance would be down the valleys
they occupy.

This hypothesis, like the one it displaced, met with opposition and lead
to much discussion. Better still, it awakened fresh interest in glacial
studies and led to renewed observations of glacial phenomena.

Among the able opponents of the dilatation hypothesis was W. Hop-
kins, who showed, by means of mathematical demonstrations, that the
direction of least resistance to expansion at most points within a glacier
would be vertically upward, and that the assumed cause of glacial flow, if
really in action, would cause a glacier to increase in thickness rather than
advance down a valley.

Many other objections to the hypothesis under review have been
advanced from time to time. It has been shown, for example, that the
changes of temperature to which glaciers are ordinarily subjected do not
penetrate far beneath the surface ; and besides, if glacial flow is due
solely to the freezing of water within the ice, it should be greater by
night than by day, and greater in winter than in summer, which, as



HOW AND WHY GLACIEKS MOVE. 165

we know, is the reverse of the truth. These and other objections to
the hypothesis of dilatation have led to the conclusion that it is in-
adequate as a complete explanation of the normal movements of even
alpine glaciers.

In closing this brief review of a long and instructive discussion, I wish
to remind the reader that, although the hypothesis of dilatation as a whole
has been abandoned, the labors of its advocates were not in vain. It not
only served a useful purpose in stimulating inquiry, but that it is based
on a true principle must be conceded even at this time when several
younger and more promising hypotheses are in the field. It cannot be
truthfully denied that water does freeze in cavities and capillary passages
in glaciers, and in so doing does exert a force which tends to move them.
What the opponents of the hypothesis have demonstrated is that the force
appealed to is inadequate to bring about the results observed, and that it
is not the only force that tends to produce glacial motion.

The Hypothesis of Plasticity. The attempt to explain the flow of
glaciers, to which this name has been applied, is based essentially on two
principles: 1. That ice is plastic, and will change its form under pressure.
2. That ice in sufficiently large masses, when unconfined, will flow under
the influence of its own might. Each of these propositions has been vig-
orously assailed, and even at the present day their correctness is not
admitted by eminent physicists. More will be said in this connection in
advance.

It is stated by Tyndall 1 that the first suggestion in reference to
glacial ice behaving as a plastic body was made by Bordier, in 1773.
This germ, however, did not bear fruit.

In 1841 Rendu presented a " Theorie des Glaciers de la Savoie "
before the Royal Academy of Savoy, in which the idea that glacial ice
behaves as a plastic solid is clearly enunciated. This important discussion
not only of the movements of glaciers, but of many other phenomena con-
nected with them, has been republished, together with a translation in
English, and may be found in most scientific libraries. 2

The hypothesis that glaciers owe their movements to the inherent
plasticity of the ice composing them found its chief advocate in J. D.
Forbes, who first applied the term viscous to glacial ice, and by long con-
tinued study and careful experiments sought to establish a "viscous

1 "Forms of Water," 1875, p. 153.

2 "Theory of the Glaciers of Savoy," Macmillan & Co., 1874.



166 GLACIERS OF NORTH AMERICA.

theory " of glacial motion. The formal hypothesis, as stated by Forbes, 1
is : "A glacier is an imperfect fluid or viscous body, which is urged down
a slope of certain inclination by the mutual pressure of its parts." As the
terms viscosity and plasticity are now strictly denned, although applied to
phenomena which in reality merge together, it is advisable to change the
wording of the hypothesis under review, as first stated, without, however,
altering the meaning that was intended to be conveyed. We shall speak
of solids that yield continuous under pressure as a plastic solid, and a fluid
which flows sluggishly as a viscous fluid.

The conclusion that an alpine glacier flows more rapidly in its central
part than at the sides was first definitely established by Agassiz, and after-
ward verified by Forbes. The flow of a glacier was thus shown to be
strikingly analogous to the flow of a river. This fact has been so well
established, and may be so easily verified, that we are justified in saying
that a glacier flows in the same manner as a fluid body, that is, it advances
owing to differential molecular motion in essentially all its parts. Whether
the flow of glacial ice is due to its plasticity under the influence of its own
weight, or is owing either wholly or in part to other causes, I trust will
appear as we advance.

The propositions on which the hypothesis of plasticity are based must
necessarily be verified before they can be applied in explanation of the
behavior of glaciers. Let us see first of all whether conclusions have been
reached in reference to the plasticity of ice.

Ice, as we ordinarily see it, is a hard, brittle substance, which may be
shattered into angular fragments by a sharp blow. At the first glance it
would seem that scarcely any statement could be farther from the truth
than to say that such a substance is plastic, that is, will yield continuously
to pressure without fracture. Many substances, however, like pitch,
asphaltum, etc., which at ordinary atmospheric temperatures are as brittle
as ice, and like it may be broken into angular fragments by a force sud-
denly applied, will, if time be allowed, slowly change their form, or flow,
under the influence of their own weight. Many experiments have been
made which demonstrate that ice under sufficient pressure, if slowly
applied, will also change its shape without being broken. Without digress-
ing too far from the main subject in hand, I may state that perhaps the

1 " Travels through the Alps of Savoy and Other Parts of the Pennine Chain, with
Observations on the Phenomena of Glaciers," Edinburgh, 1845, second edition, p. 365.
''Norway and its Glaciers," Edinburgh, 1853. "Occasional Papers on the Theory of
Glaciers," Edinburgh, 1859.



HOW AND WHY GLACIERS MOVE. 167

most conclusive experiments in this direction have been made by J. C.
McConnel and D. A. Kidd, 1 who demonstrated that a bar of glacier or
other ice, if composed of many crystals, will yield continuously and with-
out fracture to both pressure and tension. That is, it apparently behaved
as a plastic solid. The greatest freedom of motion occurred when the
ice experimented on was near the melting point, and decreased with a
decrease of temperature. At 2 C. its " plasticity " was twice as great
as ( at 10 C.

Further experiments, by McConnel, with bars, of ice cut from single
crystals, brought out most interesting results. It was found that when
such a bar with the optic axis of the crystal perpendicular to two of the
side faces was subject to bending stress, it would bend freely in the plane
of the axis either at or below the freezing point, but not at all in a plane
perpendicular to it. In the bent crystal the optic axis in any part was
normal to the bent faces in that part. The crystal behaved as if it was
composed of an infinite number of thin sheets of paper, normal to the optic
axis, attached to each other by some viscous substance which allowed one
to slide over the next with great difficulty.

The results of the experiments just cited seem to show that while ice
composed of many crystals yields to both pressure and tension in a manner
that is strikingly similar to the behavior of plastic substances under similar
conditions, yet the manner in which it yields is decidedly different. The
flow of liquids and of viscous substances is accounted for by the movement
of adjacent molecules past each other, in any direction ; in ice, motion in
response to pressure or tension -takes place along gliding planes which
have a definite relation to the optical axis of the crystals.

The movements observed in ice under pressure is not, therefore, true
plasticity, but, as pointed out by McConnel, is identical in nature with
the displacements along planes observed in rock salt, Iceland spar, and
other substances when subjected to pressure. For this peculiar " plasticity "
no definite name has been proposed. 2

1 " On the Plasticity of Glaciers and Other Ice," Proc. Eoy. Soc., London, vol. 44, 1888,
pp. 331-367. Also, J. C. McConnel, "On the Plasticity of an Ice Crystal," Ibid., vol. 48,
1890, pp. 256-260 ; vol. 49, 1891, pp. 323-343.

2 The experiments by McConnel and Kidd cited above have recently been repeated by
Dr. O. Mtigge, and are described by Chamberlin, in the Journal of Geology, vol. 3, 1895,
pp. 965, 966, as follows :

" Prisms were cut from carefully formed ice in various directions to the principal crys-
tallographic axis, i.e. the optic axis of the crystal, particularly in directions parallel and
transverse to it. These were tested by placing their ends on supports, and weighting them in
the center. In testing the transverse prisms, the optic axis was first placed in a vertical posi-



168 GLACIERS OF NORTH AMERICA.

Experiments by Tyndall on the moulding of ice into various shapes,
by pressure applied with comparative rapidity, will be referred to in
advance in connection with the consideration of another property of ice,
namely, revelation.

The manner in which glacial ice moulds itself to the inequalities of
the rocks over which it flows, so as, in many instances, to polish and striate
the bottoms and sides of narrow trenches, and even the under surfaces of
projecting ledges, is undeniable evidence that it behaves as a plastic body.
One of the objections urged against the idea that ice is plastic is that
although it yields to pressure, it is supposed not to yield to tension, and
hence lacks one of the properties of a plastic substance. The formation
of cracks in glaciers is frequently cited as proof that ice breaks under ten-
sion. This conclusion was held by Tyndall, who apparently demonstrated
by delicate experiments that ice would not yield to tension except by frac-
ture. More recent experiments by McConnel and Kidd, already cited,
have shown, however, that ice does stretch under tension when slowly
applied. The widening of crevasses, frequently to be observed both in neve

tion. The prisms sagged, and their ends were drawn inward. Optical examination showed
that the optic axis remained normal to the bent surface. Subsequent observations on surfaces
fractured for the purpose showed striation and other indications that plates of the crystal
parallel to the basal plane had sheared upon one another.

" When similar prisms were placed so that these gliding planes stood on edge, no
appreciable results followed, even though greater weights and longer times were employed.

" When prisms cut parallel to the principal axis were tested, the gliding planes being


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 18 20 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 18 of 24)