THE RAINFALL OF GUERNSEY, 285 



The nine stations have given fairly consistent results, 

 the peculiarities of former years having been maintained. 

 The feature which stands out as noticeable is the small rain- 

 fall of Claire Mare. This station has been reporting for eight 

 years, and its results have been persistently and consistently 

 lower than any other station. It is my intention in this paper 

 to confine the discussion to the distribution of rain over the 

 island, its peculiarities having been brought to light largely 

 by the extraordinary difference between the Town rainfall and 

 that of Perelle. There is now no doubt that much heavier 

 rainfall is measured on the elevated portions of the island. 

 There is also no doubt that the rainfall increases in amount 

 from west to east. These facts have been proved by the new 

 gauges and enable us to form an opinion on the distribution 

 which may be wrong as to details, but will no doubt prove to 

 be correct in its main features. There is a difference in the 

 eight years' average between the Town and Perelle of, roughly, 

 6 inches, the figures being 34*65 and 28*31 inches. Such 

 stations as Cobo and FAncresse give intermediate quantities. 

 The results from the two new Town stations, Couture and 

 Rohais, for this their first year go to prove that they are out- 

 side the area of greatest rainfall, as is also St. Sampson's, but 

 these stations must report for a longer time before we are able 

 to place them correctly in the scheme. 



As far as the evidence goes at present the area of great- 

 est rainfall is limited to the Town proper and the high land to 

 the south and west of the Town. 



It is now becoming possible to provide corrections of the 

 rainfall of different parts of the island which, when applied to 

 the readings, will show consistency where apparently there is 

 contradiction. 



The first correction is one for differences of elevation. 



For the moment I have adopted one inch in 100 feet as 

 representing the probable correction, and in Table III. I have 

 corrected the readings by this proportionate amount, but 

 instead of using the elevations of the stations, I have used the 

 elevations of the high ground in their neighbourhood, because 

 it is the true cause of the difference. In this investigation I 

 correct with Brooklyn as a standard because there is, so far, 

 no possibility of fixing a mean height for the island. 



The second correction is an estimated one. Having 

 found that the rainfall diminishes from W. to E. by over three 

 inches after being corrected for elevation, this amount is added 

 (or a proportionate amount) to the western stations. 



