A PLEA FOR THE EXACT MEASUREMENT OF 

 RAINFALL. 



By Frank Flowers, C.E., F.R.A.S., F.R.G.S. 



The detailed study of rainfall should be a national question 

 in every civilized covuitry, and the economic importance of it is 

 so tremendous that there is no need for an apolog^y in introduc- 

 ing' a paper, which has for its purpose a brief consideration of 

 the method of rainfall measurements. 



It is unnecessary here to tiilate on the influence of rainfall 

 ujion climate and then upon immobile and mobile life. Its 

 importance upon the water supply to towns is apparent. Many 

 industrial concerns depend upon a favourable rainfall. Besides, 

 the character of its occurrence determines the safety of life and 

 property, the formation of dongas and a best of other con- 

 siderations into which the subject enters. 



From the day, in 1662, when Sir Christopher "VVren fir.st 

 designed the rain-gauge, little was done in the British Isles until 

 i860 to establish a co-ordinated system of rainfall observations. 

 Since the latter date, however, an organized body of observers 

 has been working', and a large amount of valuable information 

 has thus been secured. In South Africa, for manv years past, 

 the Cape and Natal Provinces have been attending- to this national 

 necessitv, while in the Transvaal we are just celebrating- the 

 tenth anniversary of the establishment of its Meteorological 

 (Observatory. The success achieved by the latter institution 

 during^ the past decade is a matter of gratification to all interested 

 in the subject of climate. Its Annual Reports, which have been 

 regularly issued in the past, have been much -rippreciated and of 

 considerable value. It is hoped that the new Union Depart- 

 ment of Meteorolog-A- will continue the excellent practice. 



In countries where the characteristic rainfalls are slow and 

 continuous, and where the total annual precipitation is such ss 

 to adequately replenish both natural and artificial reservoirs the 

 question of the exact measurement of rainfall is not so impera- 

 tive and the ordinarv eaug^es of the '' Snowden " or Indian 

 pattern give sufficient data upon which to design hydraulic 

 schemes. In the zone of torrential rainfalls, however, it is con- 

 tended that the ordinary gauge does not g-ive sufficient informa- 

 tion, and that tlie automatic instruments possess manv advan- 

 tag^e's over the older forms. 



For the ag^riculturist, in areas where the characteristic rain- 

 fall is torrential, the determination of the ratio of beneficial to 

 non-beneficial rainfalls of the total annual precipitation is a sub- 

 ject of great importance, while to the engineer :i knowledg-e of 

 the intensity of rainfalls, tiie area covered by same, and the run- 

 ofif for any particular locality is absolutelv necessary for the 

 correct desig^ning of public and private works. 



The British standard unit of rainfall is taken as the number 

 of inches precipitated per hour, and the run-ofif is calculated in 

 cubic feet per second from any g^iven catcliment area; conse- 

 q4.iently, the more clearly the rainfall measurement shows the 



