38 J. HOPKINSON — HEETFOEDSHIEE EAINTALL, 



its area to be 630 square miles. If the county be divided along 

 the water-parting of the Colne and Lea, the Colne catchment- 

 basin (including very small portions of the basins of the Thame 

 and the Brent) will be found to have an area of about 220 square 

 miles, and the Lea catchment-basin (including a small portion of 

 the basin of the Great Ouse) an area of about 410 square miles. 

 It is evident that a correction is required for the disparity of areas 

 in order to obtain an approximately true mean for the county, the 

 efPect of its application being to reduce the mean derived directly 

 from the records of the ten stations. Multiplying the inches of 

 rainfall in each area by the square miles it contains, and dividing 

 the product by the number of square miles in the two areas, the 

 resultant mean is 25-80 ins., thus : — 



(28-08 X 220) + (24-58 X 410) 



630 = ^''■^" 



The following results have now been arrived at by three different 

 methods : — 



Mean rainfall derived directly from the records of ten stations : 

 26-33 ins. Mean rainfall corrected by comparing ten calendar 

 years' rainfall with the mean at 21 stations, and fifteen calendar 

 years' rainfall with the mean at a varying number of stations 

 (26 to 36): 26-07 ins. Mean rainfall corrected for disparity of 

 distribution of the ten stations: 25-80 ins. The mean of these 

 results, the first of which is certainly too high, is 26-07 ins. I there- 

 fore think that we should adopt 26 inches as the nearest possible 

 approach in round numbers to our mean annual rainfall, of which 

 12-93 ins. will be our summer, and 13-07 ins. our winter rainfall. 



The availability of rainfall for water-supply depends very much 

 upon how the rain falls. In heavy falls of short duration some 

 of the rain may run off the ground into ditches, and thence into 

 the rivers and cause floods, and if followed by drought much 

 moisture may evaporate, and consequently the percolation may 

 be comparatively small. With gentle rain continuing for a long 

 period, on the other hand, there will be a considerable amount of 

 percolation, but little evaporation, owing to the atmosphere being 

 kept in a more than usually humid state, and no loss by water 

 running off the ground. This is, no doubt, especially the case 

 in a chalk district. If rain fell every day in Hertfordshire, the 

 daily fall would be about 0-07 in. ; but the mean number of days 

 per annum on which at least 0-01 in. falls is about 170, and 

 therefore the average per diem on the days on which it does 

 fall is about 0153 in., the falls recorded varying from 0-0 1 in. 

 (the least quantity measured as a rule) to upwards of 4 inches. 

 iSimilarly, two moderately wet months following one another 

 would probably give a rather greater percolation than would 

 a very wet month followed by a very dry one. 



Temperature also must have a considerable effect upon evapora- 

 tion. The higher the temperature, the brighter (or less obscured) 

 the sun, the more moisture will the atmosphere absorb, and the 



