1863.] PKESTWICH MOULIN QUTGNON. 503 



river, when it flowed in a channel higher by nearly 100 feet than the 

 present river-bed. In the diagram-section, fig. 3, I have given 

 what I consider may have been the section of the valley at that 

 period, showing a shallow and broad river with numerous, generally 

 dry, shoals and shingle -banks, but during floods, arising from the 

 melting of the winter snows and a greater rainfall than at the pre- 

 sent day, rising to a height of 40 to 50 feet above its ordinary level, 

 flooding the adjacent country, and depositing, out of the course of the 

 main current, the fine silt now forming the Loess. Mixed with this 

 high Loess are angular fragments of flints washed down from the ad- 

 jacent heights or carried from the shores by floating ice. As I have 

 entered into this question fully in another paper*, I will not go fur- 

 ther into it here than is necessary to apply it to this particular case. 

 As the valley was by degrees excavated, such portions of the old river- 

 bed as escaped denudation emerged gradually from the level, first of 

 the liver, and later of the river-floods, as shown in fig. 4. At this 

 time the high-level gravels of Mautort and Moulin Quignon (A' and 

 B') would have acquired their full thickness, but would, during floods, 

 be subject to disturbance by ice-floes, and would gradually be co- 

 vered up by Loess. At the same time the draining-off of the water 

 and the percolation of the rain-water through the permeable gravel- 

 beds, acting through a long period on the chalk, would attack and 

 dissolve the weakest parts, and gradually give rise to the numerous 

 gravel-pipes we find on this level, both at Moulin Quignon, Mautort, 

 and elsewhere ; whilst the depression formed on the top of the gravel 

 thus let down would in process of time be filled up with Loess by 

 the successive floods, levelling the irregularities of surface thus pro- 

 duced. 



Fig. 5 represents the valley at the time of the formation of the 

 low-level gravels, when the excavation had attained nearly its present 

 depth, and when the deposits of Mautort and of Menchecourt, with the 

 many fluviatile and estuarine shells and the abundant Mammalian 

 remains of the latter place, were in course of formation. The section 

 does not traverse Menchecourt, but it intersects the low-level beds of 

 the " Carree de Six " at the gates of Abbeville, where several fine 

 molars of Elephas primigenius and some Flint Implements were found 

 during the making of the moat by M. Boucher de Perthes. Here the 

 high-level gravels, A" and B", are dry and out of the reach of the 

 floods which formed the lower-level Loess of Mautort and Menche- 

 court. 



Fig. 6 gives the section of the valley at the present time, show- 

 ing the relation of the high-level gravels of Moulin Quignon, B'", 

 and Mautort, A'", to the lower-level gravels of the valley, and 

 of these to the recent alluvium with peat which has filled up and 

 covered the irregularities left by the latter, the two showing an 

 origin entirely separate, and an existence perfectly independent. 



Whatever may be the conclusions drawn respecting the Jaw and the 

 Flint Implements recently discovered at Moulin Quignon, the age of 

 the beds is to me perfectly well determined, as belonging to an early 

 * Proc. Roy. Soc. vol. xii. No. 49, March 18C2, p. 38. 



