Mr Stevenson ow the Defects of Bain-Gauges. 13 



country is no doubt chiefly important with reference to the 

 interesting but uncertain science of meteorology, yet it often 

 forms an indispensable element in practical questions of fre- 

 quent occurrence. Thus it is interesting and important in 

 relation to agriculture; while its intimate connection with 

 some questions of drainage and waterworks is well known to 

 every one who is conversant with these departments of civil 

 engineering. 



Having been incidentally led some time ago to inquire into 

 the inconsistent results of rain-gauge experiments, I came to 

 the conclusion, that the great sources of error were the small- 

 ness of the recipient surface, the rim, and the position of the 

 funnel, which offering resistance to the wind, produce dis- 

 turbing eddies, and cause also resilience and dispersion of the 

 rain-drops. 



In order the better to contrast rain-gauges of the size now 

 generally used with instruments presenting greater surface, I 

 shall enumerate the different sources of error, pointing out as 

 I go along the diminution of error which would in some cases 

 result from an enlargement of the area, and adding such ge- 

 neral remarks as may occur. I shall thereafter describe a 

 form of instrument which seems to me to possess several ad- 

 vantages over those now in use. 



It may in the first place, however, be proper to observe, 

 that although there is some diversity of opinion among the 

 best authorities regarding the proper size for gauges, yet 

 these instruments are almost invariably made of from 6 to 12 

 inches diameter, I am not prepared to recommend any par- 

 ticular size to the notice of the Society, more especially as 

 convenience must in many cases be consulted; but it seems 

 probable, that the larger they are made the better, and for 

 ordinary use they could be conveniently enough constructed 

 of from 2 to 4 feet diameter, and of the form hereafter de- 

 scribed. 



I. The first error which may be noticed, is that resulting 

 from an inaccurate discharge of the rain, hail, &c., actually 

 in motion, arising from the contour of the adjacent ground or 

 from the altitude and exposed position of the gauge itself. 



