Iviii PEOCEEDINGS OE THE 



to a knife-edge at the top, and sunk to their full depth below the 

 level of the ground in which they are placed, but so that the edge 

 just projects above the ground. Some of these cylinders are filled 

 with the ordinary surface-soil of the neighbourhood, and the others 

 with chalk (from the Upper Chalk) in a fragmentary condition, and 

 with sand ; grass is growing on the upper surface of the contents 

 of the cylinders, and also on the ground surrounding them ; and 

 there are means provided of collecting and measuring the amount 

 of water which finds its way through the cylinders, that is, through 

 three or six feet of soil, chalk, or sand. It is a remarkable thing 

 that, although there may have been a rainfall of 15 or 16 inches in 

 the summer months, not more than 1 inch finds its way through 

 the ordinary soil, and If through the chalk; while, with an average 

 rainfall for 23 years of 14-17 inches in the six winter months, 6'03 

 inches went through the 3 feet of soil, and 8'8 inches through the 

 3 feet of chalk. For the same (winter) period for the 12 years 

 ending 1872, the average rainfall was 12"8 inches, of which 5"1 

 percolated through the soil, and 7*3 through the chalk. "With an 

 increase of about 2 inches in the rainfall in the winter period there 

 was thus an increase of 1 inch in the quantity which went through 

 3 feet of soil, and l-i- in that which went through 3 feet of chalk. 

 In this way it was found out what is about the proportion of the 

 water which percolates into the soil as compared with that which 

 is carried off by evaporation and vegetation. Some idea of the vast 

 amount of water absorbed from the ground by trees may be gathered 

 from Professor Attfield's paper " On the Physics and Chemistry of 

 the Sap of Plants," * which many of those now present heard him 

 read at a meeting of the Society a few days ago. 



These intermittent streams are thus fed from the natural subter- 

 ranean reservoir, which varies in height as there are long periods 

 of dry or wet weather. After a long period of drought, the surface 

 of the underground reservoir falls, becoming very nearly level, in 

 consequence of the water leaving it as springs not being replaced, 

 but after heavy rains the inclination of the surface of the water is 

 again raised. It is thus possible to tell the variation in the height 

 of the water in the wells in the neighbourhood as well as elsewhere 

 by knowing what the fall of rain has been. In a well at Studham, 

 miles away from any stream, the level of the water in two successive 

 years has varied as much as 70 feet. When the subterranean 

 reservoirs receive a great accession of water by a long- continued 

 rainfall, these " bournes," or intermittent streams, make their 

 appearance ; but it might happen that the rain which fell in the 

 course of one winter would not be sufficient to raise the surface of 

 the reservoir sufficiently high for a particular stream to flow, and 

 on the other hand it might flow when there has not been any par- 

 ticularly heav}^ rainfall, for in some years the Bourne has been 

 found flowing when there had been no excessive amount of rain, 

 but this was in consequence of the accumulation of the rain of 

 previous years. 



* 'Trans.,' Vol. II, p. 229. 



