78 J. HOPKINSON THE CHADWELL SPiJING 



second we know that it did not, although, from the heavy winter's 

 rainfall and rather large year's fall, it was at least likely that it 

 might do so. 



It may be objected that the mean rainfall in Hertfordshire does 

 not represent that which falls in the valley of the Bourne. I have 

 therefore compiled another table (Table II, p. 79) showing the 

 mean rainfall at two places, one a short distance above Eourne End 

 and the other a short distance below it — Berkhamated, and Nash 

 Mills near Hemel Hempstead — for each year preceding the flowing 

 of the Bourne from 1853 to 1897. In this table the year is divided 

 into quarters, and the rainfall of the nine months ending 31st 

 March, and year's rainfall ending the same date, are also given. 

 For the year 1852-53 it is that of Nash Mills only, observa- 

 tions not having then been commenced at Berkhamsted. It may 

 be observed that, although in the last year in which the Bourne 

 appeared there was less rainfall than in any other of the series, it 

 was probably quite as effectual as in several other years, for the 

 deficiency is more than accounted for by the small fall in the 

 second quarter of 1896, when, owing to vegetation being in its 

 most active state, but little rain usually gets into the Chalk out of 

 the reach of evaporation. There was an average year's rainfall in 

 the nine months, July, 1896, to March, 1897. 



It may be suggested that the flowing of the Bourne is not alto- 

 gether dependent upon the rain which has fallen during the 

 preceding twelve months. Water percolates through the unfissured 

 Chalk very slowly, and the plane of saturation may and doubtless 

 does stand at a higher inclination one summer, after all percolation 

 has ceased for a time, than at another, so that when a period of 

 excessive rainfall comes, the rain may not always have the same 

 depth of chalk to percolate through, and therefore after a wet 

 year, or series of wet years, a smaller amount of rain would be 

 required to cause the Bourne to flow than after a dry year, or 

 series of dry years. 



The mean rainfall at Berkhamsted and Nash Mills for the year 

 ending 3l8t March, 1892, was half an inch more than it was for 

 the year ending 31st March, 1897. If the flowing of the Bourne 

 depended upon the amount of rain which falls in a single year, 

 irrespective of the time of the year at which it falls, it would be 

 more likely to have flowed in 1892 than in 1897. The mean rainfall 

 at these two places for the year 1890-91 was six inches less than 

 it was for the year 1895-96. (See Table III, p. 80.) This gives 

 five inches and a half more for the two years together at the later 

 period, when the Bourne did flow, than at the earlier, when it did 

 not. This is a strong point in favour of the idea that some portion 

 of one year's rainfall may be stored up in this valley above the 

 normal plane of saturation until it receives further accession from 

 the following year's fall, as it would be in a valley of less 

 inclination with a permanent river, but it is not conclusive. Again, 

 not only was the rainfall of the year 1890-91 much less than was 

 that of the year 1895-96, but that which did fall was much less 



