196 Z.H. D. La Touche—Relics of Ice Age in India. 
rotting away, if I may so express myself, and breaking up the rocks 
with great rapidity, in comparison with tracts where the roots of grass 
and trees hold together and form a protective covering of soil. At 
these altitudes many of even the highest peaks are almost buried 
under an accumulation of their own debris, so much so that it is 
sometimes possible to ascend to heights of 20,000 feet or so with no 
more discomfort than is caused by the difficulty of breathing in 
a rarified atmosphere. This loose material is always in a state of 
more or less unstable equilibrium, and when lying on steep hill-slopes 
is constantly liable to be precipitated into the valleys, especially when 
it is saturated with moisture by melting snow or rain. When this 
happens in a tributary valley, the semi-liquid mass moves irresistibly 
downwards as an avalanche of mud mingled with boulders and 
fragments of rock, and on reaching the valley of the main stream or 
river, spreads itself out as a conical fan, radiating from the mouth of 
the tributary. In all the valleys of the higher Himalaya these fans 
can be seen in actual process of formation, and they may sometimes be 
met with even at elevations well below the snow-line, when the head 
of the tributary stream reaches up to a sufficiently high altitude. 
When the side-streams are close together the fans often coalesce, 
forming a more or less continuous terrace along the banks of the 
main river. 
If we now follow the larger valleys downwards, we find that, when 
we reach altitudes of say 7000 feet or less, exactly similar cones or 
fans occur along each bank wherever the main stream is joined by 
a tributary, but that these fans bear every sign of great antiquity, in 
proportion to their distance from the snows. Not only are they 
covered by vegetation and usually cultivated, but they are often 
selected: as sites for villages, on account probably of the ease with 
which water can be obtained by leading channels from the side- 
stream. (I have also noticed this propensity in the Alps, where in 
the main valleys each fan will have its well-watered little town built 
upon it.) This shows that no fresh material has been spread over the 
fans within recent times. Moreover, the stream that originally 
formed them, instead of flowing in devious rivulets over their surface 
as it did at first, has now cut a single deep channel directly through 
the centre, so profound in most cases that it is evident that a very 
long time has elapsed since the stream ceased to construct the fans. 
Lower down still the whole of the valley is found to be choked with 
a great thickness, often as much as three or four hundred feet, of loose 
deposits continuous with and similar to those forming the cones, out 
of which the river has cut great terraces, now on one side and now on 
the other, with a perpendicular face towards the river. Sometimes 
only small remnants of these terraces are now left, but at other 
times they form large level plateaus, on which considerable towns 
may be built. 
The origin of these terraces has aroused some discussion among 
eeologists, for it is evident that they are now being washed away by 
the rivers that once deposited them, and that, being high above the 
present. flood-level, there must have been a great change of conditions 
of some kind since that time. Changes of level, either resulting from 
