dp:velopment of the valleys 397 



deed a fortunate circumstance^ esjDecially under such conditions as those 

 of the late war. 



The writer's theor}' as to the histor}^ of such streams is that in the 

 early stages of the development of the present drainage system their loca- 

 tion was determined, like that of other surface streams (as they then 

 were), by the general slope of the country and irregularities of surface 

 due to slight folding during uplift, and other causes. 



The streams, having cut through the superficial soil or drift, would 

 continue to deepen their valleys through the chalk in the ordinary way 

 unless the ground-water level in the chalk were much below the bottoms 

 of the streams. If such were the case, there is no doubt that the smaller 

 streams would be absorbed by seepage through the chalk and descend 

 until they encountered the surface of the water-table beneath. Experi- 

 mental demonstration of this contention was furnished during the war by 

 the efficiency of sumps cut into the chalk in disposing of surface drainage 

 of camps, trenches, etcetera, by seepage. Having reached the surface of 

 the water-table, the stream would no doubt tend somewhat to coalesce 

 with it and spread out over it; but, the descending volume here being 

 much greater than the seepage from the surrounding country (from the 

 surface run-off of which the stream is formed), the surface of the water- 

 table along the locus of the stream would tend to be kept at a higher 

 level than on either side. The water-table dips toward the outlet of the 

 surface streams; hence the waters of this elevated ridge or mound, espe- 

 cially at its downstream end, in seeking their level would flow much more 

 quickly and exercise their solvent power more rapidly along the line of 

 that dip than in other directions — that is, the main movement would be 

 parallel to the original flow of the surface stream. As the fissures were 

 more rapidly enlarged along this axis, they would afford the easiest outlet 

 for an increasingly greater proportion of the descending volume until an 

 open channel was formed sufficient to accommodate most of it. As this 

 channel deepens by erosion of its floor the surface of the water in it 

 would sink until, like that of a surface stream, it became the lowest part 

 of the surrounding water-table, and the relations of the underground 

 stream to the water-table on either side would be similar also in other 

 respects to those of ordinary streams. The motion of ground-water 

 Avould now be toward the stream and downstream, instead of away from 

 the stream axis and downstream. The foregoing argument is based on 

 the assumption that the fissuring of the chalk permits the passage of 

 water with equal facility in all directions, and this seems borne out by 

 facts of observation. 



As the ground-water surface would be lowest near the seacoast or the 



