924 PROFESSOR J. PRBSTWICH ON THE EVIDENCES OF A SUBMERGENCE 
sediment with which the surrounding waters were charged became, as the flood 
waters advanced, gradually greater; while the height which the Loess reaches in the 
valleys of the great Alpine-born rivers would indicate that the submergence was 
probably not less than from 1500 to 2000 feet. It was, in any case, sufficient to cover 
all the lesser inequalities of surface over which the Loess spreads, and accords with the 
assumption that, whatever the cause was, it acted equally in all directions round 
certain central areas. It is also in accordance with the singularly homogeneous 
character of the Loess, which is such as would be due to the agency of a common fluid 
medium. 
As the waters, on the upheaval of the land, retreated, they carried with them a 
portion of the sediment deposited on the heights, which again would be re-deposited 
at lower levels on the slopes and plains that were least under the influence of the 
effluent currents. I do not imagine that at any time this deposition of this Loess 
was sufficent to fill the great valleys of the Rhine and Danube, but only to drape 
their contours, while the scour of the effluent waters kept the main channels free. 
Not only have great valleys been thus scoured, but large tracts in the midst of 
Loess-laden countries have been swept bare by the great volume and more rapid flow 
of the retreating waters. For wherever the velocity of the effluent currents was less 
than 8 feet per second, the precipitation of fine sediments may have taken place, 
but where the channels were more contracted, and the currents exceeded 8 feet per 
second, no such deposit, or but little, need have been left on the land, so that under 
similar flood conditions, these deposits might cover certain large tracts, whilst adjacent 
tracts might remain bare of them. 
It may be objected that the deposit of Loess is so great that it could not have been 
formed in the short time I would allow for the submergence. But it must be under¬ 
stood that I look ujoon the Loess, as a whole, as belonging to more than one period, in 
the sense that it was, in the main, primarily a valley deposit in time of floods; and, 
secondly, as just described, a reconstruction in great part of a more general character 
—the one the accumulation of long ages, and the other of brief time. The first is a case 
of ordinary sedimentation ; the other, one of wider sweep and mainly of redistribution. 
It is well known how one heavy rainfall will remove and re-deposit many feet of an 
alluvial deposit like the Loess. Striking evidence of this is afforded by the changes in 
the valley of the Ganges and its tributaries, described by Mr. James Fergusson.* Old 
channels have there been filled up and new ones formed with extraordinary rapidity; 
and, on the banks of the Brahmapootra, “hundreds of miles” of alluvial land are swept 
away every year and re-deposited lower down the river course. A few seasons suffice 
to form deposits 40 feet thick, whilst little addition is made in other places. We 
may suppose, therefore, that as the ocean waters, charged with silt, advanced up the 
great river valleys, and met the river waters laden with glacial mud, a vast homoge¬ 
neous mass of sediment would be the result, and this would necessarily lodge over 
* ‘Qaarfc. Jonrn. Geol. Soc.,’ vol. 1,9 p. 321. 
