18 Mr Stevenson on (he Defects ofBain-Gauges. 



The mouth of the instrument which is recommended, is in- 

 tended, as shewn in the diagram, fig. 1 (representing a gauge 

 of the common size), to be sunk to a level with the ground {hf 

 g i), which is generally considered the best position by those 

 who have experimented on the subject. On the top of the fun- 

 nel {a b c), which has a larger mouth than that of the receiving 

 area {d e), there is placed a zone (a d e c) projecting inwards 

 about six or nine inches, thus reducing the area of the receiving 

 surface to the required dimensions. There is therefore exposed 

 to the wind a very small rim (at d and e), while the shelving 

 sides of the funnel come only as high as the bottom of the zone, 

 and are therefore so far below the aperture of the gauge as to 

 render it quite impossible for even hail to recoil in such a way 

 as to escape out of it. To prevent rain from without being 

 blown up the inclining side (as at a d or c e) of the zone, so 

 as to pass over the edge into the gauge, as well as to fill 

 up the space between the edge and the surrounding grass, 

 a circular brush f d e g^ about 3 inches wide, is placed so 

 as to present its bristles to the impinging rain, which sinks 

 among them instead of recoiling and being scattered about. 

 The wind, in passing from the grass to the receiving-hole, 

 will, of course, meet with no inequality, as the top of the 

 bristles is on a level at once with the surrounding grass (hfg i) 

 and the edge of the zone. 



Fig. 1. 



K V, ., ,„,.y d 



Fig. 2 shews the adaptation of this plan to a larger scale, 

 in which case the brush is supplanted by bristles loosely ar- 

 ranged on the horizontal copper plate, and fixed either with 

 wires or resin. A thinly cut sod or turf might probably an- 

 swer every purpose quite as well as the bristles. It i» ad vis- 



