middle chalk at its junction with this chalk marl, must, in flowing to the lower levels, pass over this formation. The drainage from the adjacent Gault very rarely passes over the Upper Greensand, the former being almost invariably on a lower level than the latter, and the flow of water consequently determined in a contrary direction (see fig. 18, p. 94).

On the whole it is probable that the Upper Greensand receives considerable additional supplies of water from the above-mentioned sources; for the sloping surface of the chalk marl forms, in fact, a gathering ground of equal, if not greater, breadth than that of the Upper Greensand itself; whilst the outcrop of the latter constantly forms with the Lower Chalk a trough in which the surface drainage lodges. In this case, therefore, the extra supply is evidently an important element in the calculation of the water-value of this formation.

96. The position of the Lower Greensand, with respect to its receiving any additional supplies of water from adjoining grounds, is unfavourable. At its boundary with the underlying deposits it forms, in Kent and Surrey, an escarpment only second in prominence to that of the chalk, with which it runs parallel; and the general drainage is from this escarpment southward towards the Weald. In the same way

in Buckinghamshire and Bedfordshire, the surface drainage is from the Lower Greensand northward towards the Kimmeridge Clay and Oolites. Again, along its boundary with the overlying gault, this latter formation occupies the lower level (with the exception of a limited tract near Farnham, and another near Devizes), and therefore retains the rain-water which falls upon it, and also intercepts any that may drain from off any of the strata above it. The Lower Greensand consequently appears to be restricted for its water-supply, almost exclusively to the rain falling within its own area.

§ 11. On the Fall of Rain,—on the Quantity infiltrating beneath the Surface, on the Power of Absorption, and on the Permeability, of the Water-bearing Strata.

97. BEFORE proceeding further with the geological question, it is necessary to determine the quantity of rain which falls on the exposed surfaces of the water-bearing deposits, and to inquire what proportion of it may probably pass into the interior of the strata.

The annual rain-fall in any one district varies so materially from year to year, that in order to obtain a true mean, it is indispensable that it should be taken from the results of many years. At present exact records of this description are confined to a very few places in the kingdom; but it is to be hoped that the more numerous and very complete meteorological observations, commenced within the last few years, will, at no distant period, supply this want. In the mean time

we can only take the data as they exist.

98. The geographical configuration of a country leads to such differences in the amount of rain-fall even in districts nearly adjacent, that it is essential to have independent data for the area occupied by each separate geological formation. It is true that in the Tertiary district, where the physical features of the surface are not very variable, the rain-fall at a central position, as at London, might give a mean applicable nearly to the whole area; but where we have a tract of country rising to the height of the chalk escarpments, and ranging for many miles in nearly rectilinear and unbroken lines, then we may expect to find, in the Greensand districts beyond these ranges of hills, a rain-fall different to that of the Tertiary area on the opposite side of them. The evidence is at present very limited, but it points generally to an amount of rain considerably

greater in the Greensand districts of Kent, Surrey, and Wiltshire, than in the Tertiary area around London.

99. As few observations have actually been made immediately within the boundaries of the Greensands, it may be expedient for the present to take the rain-falls at the nearest adjacent places in the same geographical tract. This has been done in the annexed list.*

4 Mr. Bland.

1 Luke Howard's "Climate of London." 2 For these lists I am indebted to the kindness of Mr. Glaisher of the Greenwich Observatory.

"Hydraulic Tables."

3 Beardmore's

7 University Observatory.

5 Mr. Dickinson. 8 Mr. J. Osborn.

6 Mr. Lucas. Radcliff Observatory. 12 Dr. Fielding.

* I have since received the following rain-falls for 1850,-Greenwich, 19.7; Hitchin, 20-56; Cardington, 185; Linslade, 20.91; Uckfield, 28'8; Maidstone, 23.36. The fall this year was generally much below the average. The fall in 1849 may be taken as a fair average; at Greenwich it amounted to 23.9.

+ Since continued to 1849, see Proc. Inst. Civ. Eng. Vol. IX. p. 158.

This quantity is, for a mean of 9 years, so large, that there is probably some error.

I have therefore omitted it in the general averages.

The observations are lost, and the rain-fall is given from recollection. It is stated to have been rather above 30 inches.

From these observations the following approximate mean results may be deduced, by taking an average of the rain-fall at stations in, or surrounding, the several geological districts,— those nearest on either side being combined to form a mean.* Thus the rain-fall at and near Devizes is taken as intermediate between the fall at Hungerford and Beckington; while that at and round Wallingford, is assumed to be intermediate between the fall at Oxford and that at Reading. For the Tertiary district and the chalk, the average is that of all the places named taken together. In the Greensand districts, as the several areas are detached and wide apart, a separate calculation is made for each. This must be viewed only as a temporary measure until more complete data are obtainable.

1 Position of the zone of outcrop with reference to London.

2 In round numbers.

100. We have next to inquire what proportion of the rain-water may pass from the surface into the interior of the strata. This is a subject involved in much obscurity; very few experiments have been made, and these at first sight appear rather contradictory. In 1796-8 Dr. Dalton made, at Manchester, the first series on which any dependence could be placed. The

* In Johnson and Berghaus's Physical Atlas (Map, Meteorology), the line of 25 inch rain-fall runs by Herne Bay, passes at a short distance to the south of London, and then sweeps round by Oxford; whilst the 30 inch line of rain-fall proceeds from Dover, by the north of Portsmouth, to Bristol.

66

Experiments and Observations to determine whether the quantity of rain and dew is equal to the quantity of water carried off by rivers and raised by evaporation,

apparatus he used was a vase in the form of a pluviometer, 10 inches across at top and three feet deep, into which he put "a few inches of sand and gravel," and then filled it up with "good fresh soil." A pipe at the bottom of the vase conducted the water which filtered through these materials into a graduated receiver. This was placed in the ground with an ordinary pluviometer by its side, and left undisturbed for three years the grass being allowed to grow over the

surface.

The next, and more complete series of experiments, were those performed by Mr. Dickinson at Abbot's Hill, near Hemel Hempstead.* They extended over a period of eight years (1835-43), and were conducted on the same plan as those by Dr. Dalton. The cylinder was filled with the "soil of the country, a sandy gravelly loam.”

Mr. Charnock's experiments were made near Ferrybridge, and continued during five years. The cylinder was filled with two feet of "gravel and sand, capped by soil."

101. The following are the results of these several experiments:

It must be observed in explanation of the apparent discrepancies between these results, that Dr. Dalton used the common surface soil of the district (New Red Sandstone), which contains a good deal of clay, and would probably be very

with an inquiry into the origin of springs." Memoirs of the Literary and Philosophical Society of Manchester, Vol. V. part II. p. 346.

*

Proc. Inst. Civil Eng. Vol. II. p. 168; 1843. Since these pages were written a further account, bringing the experiments down to 1849, has been published in the 9th vol. of the same work, p. 158. The infiltration during this latter period was rather + Journ. Agric. Soc. Vol. X. p. 516.

less.