aps be rendered
more distinct on the Glacier du Geant by the cascade, and necessarily
must be so, if the rents coincide with the limit at which the annual
snow-line is nearly straight across the glacier. In the region of the
Aar glacier, however, where my own investigations were made, all the
tributaries entering into the larger glacier are ribbed across in this
way, and most of them join the main trunk over uniform slopes, without
the slightest cascade.

It must be remembered that these surface-phenomena of the glacier are
not to be seen at all times, nor under all conditions. During the first
year of my sojourn on the glacier of the Aar, I was not aware that the
stratification of its tributaries was so universal as I afterward found
it to be; the primitive lines of the strata are often so far erased that
they are not perceptible, except under the most favorable circumstances.
But when the glacier has been washed clean by rain, and the light
strikes upon it in the right direction, these lines become perfectly
distinct, where, under different conditions, they could not be discerned
at all. After passing many summers on the same glacier, renewing my
observations year after year over the same localities, I can confidently
state that not only do the lines of stratification exist throughout the
great glacier of the Aar, but in all its tributaries also. Of course,
they are greatly modified in the lower part of the glacier by the
intimate fusion of its tributaries, and by the circumstance that their
movement, primarily independent, is merged in the movement of the main
glacier embracing them all. We have seen that not only does the centre
of a glacier move more rapidly than its sides, but that the deeper mass
of the glacier also moves at a different rate from its more superficial
portion. My own observations (for the details of which I would again
refer the reader to my "Systeme Glaciaire ") show that in the higher
part of the glacier, especially in the region of the _neve_, the bottom
of the mass seems to move more rapidly than the surface, while lower
down, toward the terminus of the glacier, the surface, on the contrary,
moves faster than the bottom. The annexed wood-cut exhibits a
longitudinal section of the glacier, in which this difference in the
motion of the upper and lower portions of the mass is represented, the
beds being almost horizontal in the upper snow-fields, while their lower
portion slopes move rapidly down