OF WESTEEJvT EUROPE AND THE MEDITERRANEAN COASTS. 
923 
causes and not to original structure. It is where a homogeneous bed of Loess, say 
from 10 to 15 feet thick, has, by the long continued action of the rainfall, been so 
modified in its upper part by the removal of all soluble matter, that the residual 
insoluble argillaceous and siliceous matter, with the iron oxide in an altered state, 
assumes the appearance of a separate bed. This upper division is known as the 
terre d briques, and the lower is the ergeron of the Belgian geologists. There are, 
however, cases in which the distinction is real, and the lower bed seems to be of a 
different age from the upper and may be due to an earlier glacial drift. 
M. A. Briart'”' also points out that in Belgium, besides the ordinary river-valley Loess, 
there is a mid-level Loess, which he names Limon cles plaines moyennes, as well as a 
high-level Loess, or the Limon des hauts q^lateaux; and, though he sees in these a 
great lithological analogy, he considers them distinct in jDoint of time. He says that, 
whereas the plateau Loess is unfossiliferous, the Loess of the valleys and slopes 
contains abundant remains of the ordinary Quaternary Mammalia, together with 
land shells {Helix, Pupa, and Succineaf). He further states that this mid-level 
Loess frequently rests on an angular and sandy drift of debris derived from the 
adjacent local rocks. As these are the characters of one phase of the Bubble-drift, it 
is not improbable that the deposit to which M. Briart refers may be the equivalent 
of the Faversham, Upchurch, and similar beds in the Thames Valley. 
I take this mid-level rubbly Loess to have been formed, like the “head” and the 
angular rubble on slopes, during the emergence of the land. The results during 
submergence were different. As the waters then gradually crept up the valleys they 
would have abraded portions of the softer strata and the older drift beds, including 
the river-valley Loess, and so become charged with sediment, which, as the turbid 
waters rose, would be carried to higher grounds, and there be deposited in intervals 
of rest and where the currents were weakest. 
South of the Alps Professor F. SACCoj has also recognized similar divisions of 
the Loess in Piedmont. He describes (l) a Loess of the plains of fluviatile origin; 
(2) a Loess of the hills of meteorological origin ; the latter, w'hich attains a height of 
400 metres and contains sixty-two species of land shells, he assigns to the close of 
the Glacial Period ; (3) a high-level Loess of the mountains with remains of the 
Mammoth. In Spain, south of the Eastern Pyrenees, there is also a large develop¬ 
ment of Loess. The general conditions seem, in many cases, to point to some 
connection betw'een the high-level Loess and Alpine ranges. 
Let us now see what the consequences of this gradual submergence wmuld be. The 
mountains of Europe were still loaded wdth the ice of the Glacial period, and the rivers 
surcharged with the silt from the great glaciers and abraded lands. The volume of 
* ‘Ann. Soc. Geol. de Belgique,’ vol. 18, Memoires, 1892 
t M. Briaet adds Limnea and Flanorlis, on tbe authority of M. LADRitRE, but M. LADRitRE’s correla¬ 
tions bave yet to be confirmed. 
f ‘Bull. Soc. Geol. France,’ 3rd ser., vol. 16, p. 229, 1888. 
6 B 2 
