14 Mr Stevenson on (he Defects of Bain-Gauges. 



All recent experiments have shewn that the results given 

 by a rain gauge are intimately connected with its height 

 above the surface of the ground, and have fully confirmed the 

 somewhat unlooked for fact, that on the same place rain- 

 gauges placed at different elevations above the ground indi- 

 cate very different falls of rain.* The higher the instrument 

 is placed above the ground, the less rain it collects. It is 

 needless to enter on a consideration of this subject ; but I 

 may observe that Professor Stevelly's theory of an increase in 

 the velocity of the wind as we ascend, appears to furnish the 

 most satisfactory explanation o£ the phenomenon, the more 

 so as from experiments which I have made, it seems to follow 

 that the rvind is the great origin of error ; that the stronger it 

 is, the less fain is collected; and that in calm nieather^ the ele- 

 vation, as well as form of instrument, are of little consequence. 

 But in whatever way this question may be settled, there is at 

 least every inducement for; placing the gauge on the surface 

 of the ground. 



Many opinions have been given as to the best locality for a 

 gauge, and I suppose it is now generally admitted that an open 

 plain is that which is most suitable. But as many who are most 

 anxious to undertake these meteorological experiments have 

 no champaign country to resort to, but are obliged to have re- 

 course to small garden plots, encumbered with trees and shrubs, 

 assistance in determining the most suitable position for the 

 gauge may in such cases be derived from observing the fall of 

 snow. This observation should not, however, be made when 

 the wind is strong enough to disturb the snow after being de- 

 posited. The spot to be adopted for the gauge is of course 

 that on which there is generally, with the prevalent winds, the 

 same depth as has fallen in the neighbourhood. Observa- 

 tions on the depths of snow at irregularities in the ground, in 

 valleys, and on the sides and tops of mountains, seem likely to 

 give us more accurate notions regarding the atmospheric cur- 

 rents, on a knowledge of which so much depends. If, after 

 falls of snow unaccompanied with wind strong enough to cause 



* See Professor Phillips' Reports and Professor Forbes' contributions to 

 Meteorology. 



