143 
or to occur on a steep slope, they may have been taken 
advantage of, modified, or imitated by man for economical 
purposes. In the north of England, some of the geological 
surveyors, I believe, are disposed to regard them as the 
effects of land glacial action. For reasons which may be 
gathered from this paper, I am more inclined to look upon 
them as the marks left by currents laden with floating ice 
during the last great denudation of drift-deposits. The fact 
that there are terraces of artificial origin on hill sides in 
many parts of England, is no more a reason why we should 
regard the vast majority of drift- terraces as made-earth, than 
the existence of artificial mounds is a proof that all drift- 
knolls were piled up by man. 
6. Limits of River Action. — Since the glacial submergence, 
rivers have been occupied in cutting well-defined channels in 
drift-deposits. In most parts of the valley of the Aire, they 
have not yet got down to the solid rock, and there is reason 
to believe that between Bingley and Skipton the drift in 
some places fills up hollows which at one time may have 
been lake-basins. The river deposits consist of small pebbles,* 
never spread out continuously for great distances, but, con- 
fined to patches or ridges, and of a nearly stoneless sand and 
loam, deposited over considerable areas. The rivers are not 
giving rise to boulder- deposits, or, with a few exceptions, to 
original drifts of any kind. A little investigation will show 
that where patches of pebbles are found in the river channels 
they have generally been washed out of boulder- drifts in the 
immediate neighbourhood. It may further be remarked that 
the stones and boulders of the valley drifts are often in posi- 
tions where they could not have been left by river action, 
even supposing the rivers to have been formerly much larger 
* These remarks are not intended to apply to mountain-torrents, or streams 
witli considerable inclination of channel, which are capable of moving large stones 
during floods. 
