of the adjacent upper greensand, gives another good example. Here the villages always lie thickly along a definite line. There is a well-marked and constant relation between the outcrop of porous strata and the parish or township boundaries, the longer axes of the parishes crossing the outcrops more or less at right angles. A careful study of the distribution of the villages, and of the relation of their parish boundaries to the main physical features, throws much light upon the past history of the country, and often enables us to determine the relative ages of the settlements. This branch of the subject does not now concern us: we need only note that the arrangement of the parish boundaries depends upon the sites of the settlements, and that these are controlled by the outcrops of waterbearing beds.

The early settlements in England were nearly always controlled by such circumstances as have been here referred to; but the later development of special towns and districts has depended upon a variety of circumstances. In early times it was around some shrine of special fame or sanctity, or under the shadow of the castle of some powerful noble, that the population clustered and the town increased. A little later it was also in places especially well suited for various manufactures. Within the last 200 years the great development of our mineral wealth (especially of coal and iron) has entirely transformed the country. Large towns have sprung up over the coal-fields, often on wide tracts of clay, where few settlements would otherwise have taken place. The natural surface water supply of such places is often bad and small, and the mining operations frequently drain even this.

The water supply of modern towns is, in nearly all cases, either (a) obtained from a neighbouring river, (6) brought from a distance, or (c) obtained from deep wells beneath the town. It thus, except in the first case, differs from that

*For discussion on this question, see a paper by the author, in "Journal of the Anthropological Institute," vol. iii. pp. 32-55 (with maps), 1873

of the original settlement, which always obtained its water from streams, springs, or shallow wells. In far too many cases the primitive source of supply has been continued in use long after the time when it should have been abandoned; and the local source of water supply, essential to the early development of a town, has become a source of danger as the population has increased.

Of the points just mentioned, London affords an excellent example. The old parts of London and its suburbs are built upon gravel resting on London clay (Fig. 5). Where small valleys (such as the Fleet) cut through the gravel, there are natural springs; but everywhere water can be obtained in shallow wells sunk through the gravel. So long as the inhabitants were dependent entirely upon these springs and wells, the houses were confined to the gravel; when a general system of water supply was introduced, the population extended over the intervening area of clay. Meanwhile, the increasing population, without any adequate system of drainage, fouled the shallow wells, and rendered them all more or less impure. It is only within the last few years that some of these have been closed by authority.

Below the superficial deposit of gravel, there are other sources of water supply for London. The strata beneath lie in a basin-shaped form, and thus favour the accumulation of water. Underneath the London clay there are the lower tertiary sands, holding water which rises in the wells when these are sunk through the clay. Still lower, there is the great mass of chalk in which there is an enormous store of water. Still lower, and separated from the chalk by a bed of clay (gault), is the lower greensand. This, on the south side of London, may yet yield some water, but it can never be the great source of supply which was once hoped for.

There are, then, with the river, four different sources of water supply at or beneath London, each giving a different quality of water. Probably no large city in Europe is better situated than London for supplying itself with water from within its own area; but so vast has London now VOL. VIII.—H. C.

2 C

become, that all these taken together are insufficient, or inefficient.

It is a curious circumstance that some others of the great capitals of Europe are built on "basins " like that of London, and hence are able to obtain deep well-water from beneath. Paris, Berlin, and Vienna, are good examples. This is a circumstance that could not have been known to the early settlers, who concerned themselves only with the surface sources of water supply.

ON A POSSIBLE INCREASE OF UNDERGROUND WATER SUPPLY.

By CHAS. E. DE RANCE, A.I.C.E., F.G.S., F.R.G.S.

Secretary of the British Association Underground Water Committee.

FOR nearly a century the subject of water supply has been constantly before the public, and with the growth of population has become a question of vital importance to the community. The amount of information that has been accumulated is very large, but, investigated by Royal Commissions, inquired into by committees appointed by scientific societies, it is spread over a wide range of literature. It is difficult for any one individual to focus the stores of information already available, still more for him to follow up the numerous lines of investigation these inquiries suggest.

In my work on 'The Water Supply of England and Wales,' published in 1882, I made an attempt to show what was the probable supply of water available in all the river basins of England and Wales, and what amount was required to satisfy the demands upon that supply, with the result that it appears to be amply demonstrated that the rainfall this country receives is more than sufficient to meet all the requirements of human consumption, manufacturing interests, and the purpose of canalisation; and yet, with

these resources, large districts still suffer from all the ills due to a polluted water supply, whilst other large areas are devastated by floods, representing unproductive rainfall passing to the sea.

It is to this unproductive rainfall that I would chiefly wish to call attention in the present communication. Much has been written on underground water supply since the year 1841, when the Rev. James Clutterbuck stated "the extent of the supply must necessarily be regulated by the quantity of rain falling upon the surface, the rapidity with which it is absorbed, and the reduction to which it is subject by evaporation;" but this definition still expresses the knowledge we have on the subject. Rainfall being the sole source of supply to the waters beneath the surface as well as those flowing upon it, accurate information on the amount of rain falling on a district is a matter of the first importance, and this, to be of any value, must represent the observations of an extended period of years, so that not only may the minimum supply to be expected be ascertained, but the average or mean of several successive dry years also. Happily, through the voluntary labour of Mr. Symons, F.R.S., we have now more than 2500 stations at which rainfall is recorded, and we are able, by consulting his annual volumes, to obtain the necessary information for a large number of localities; but it is obvious that, considering the direct bearing such observations have on engineering, agricultural, and sanitary questions, the scope of the inquiry should be enlarged by placing it under a Government department, which could enlarge the scope and usefulness of the inquiry without being, as now, partially crippled for want of sufficient funds to carry out the necessary details.

Pervious or permeable formations, by gradually absorbing waters falling on their surface, which slowly percolate through them, act at once as filter-beds and reservoirs, the capacity of which is limited by the area of absorption, and the thickness of the pervious bed. When rain falls upon a perfectly pervious rock, underlaid by impermeable deposits

the water line is generally near the surface; this line or plane of complete saturation, in sandstone and limestone hills intersected by valleys, is found to be slightly above the level on which the deepest valley intersects the various strata, constituting the water-bearing rock.

When bands of permeable and impermeable rocks alternate, each porous band contains a separate sheet of water, which flows down the dip planes of the strata, confined by the impermeable layers above and below. Such water flows with the head, due to the difference of vertical level, of the area of outcrop to that of the area of discharge, less the frictional resistance of the fragments of the rock through which it passes. When the facilities for the discharge of a volume are less than the quantity capable of being received, the porous rock will be full up to the impermeable layer above, which is invariably the case when all outlet is stopped by faults throwing in impermeable strata, or by the dip carrying the strata beneath the sea-level.

Such porous rocks may be regarded, when provided with an outlet, as underground conduits, the depth of which is the thickness of the bed, the width of which is the extent of the outcrop or horizontal strike of the bed, and the inclination of which is the dip of the strata. Where the outlet is blocked, the saturation-level remains unchanged, and unless water is artificially removed, so as to provide space for a fresh supply, no additional water can be added to the existing supply.

In sinking wells, or in boring into a mass of porous rock, the plane of saturation is found to vary within certain limits, being governed by the amount of previous rainfall. This level, by excessive pumping, is artificially and locally lowered, but the old "rest-level" is restored after a certain number of hours' cessation from pumping. The difference between the "rest-level" and the pumping-level is, in some wells, in porous strata as much 100 feet. The area of exhaustion resembles an inverted cone, the apex of which rests on the point at which the pumps abstract the water, and the base of which is a circle at the surface around the