442 Proceedings of the Royal Irish Academy. 



If we might assume the a observed as above to represent that be- 

 longing to real wind (and especially when found with a low velocity 

 of whirl when the eddies must be small), we could obtain /? and y by 

 its action. Let the Ajiemometer above described be detached fi'om the 

 whirling machine and placed in open air with its axle vertical, and 

 near it another as a standard, having its cups and arms of the same 

 dimensions. They will have the same a, jB and y, and the same un- 

 known V may be supposed to act on both. 



The observations of v with the two must be synchronous, and calling 

 m' that belonging to the standard, m" and m'" those belonging to the 

 other with two different values of/ obtained by hanging weights to 

 the longer arm of the brake lever, we have, as Fis given, 



= nm' ; m'" = n'ni', 



and by (YII.) 



2/3 F F' 



— = m'{n 4 1) - -J- = m{n' + 1) - -j—, — -, 



a m\n-\) m\n'-\) 



whence 



.r-=(^-^ix^. (Yin.) 



Knowing m' we have 



r, ^, and ^. 

 a a 



The difference of these values from those given by the whirling 

 machine will show how far the latter method can be relied on in this 

 inquiry. 



To avoid the irregularities already referred to as caused by local 

 circumstances, the cups of the two instruments should be at the same 

 height above the ground, their placement should be quite open, and 

 their position occasionally interchanged. But the wind itself is irre- 

 gular to an extent of which I had no idea till I examined the pressure 

 curves from which I deduced the a of my Anemometer. In one of 

 them whose time was only 99 seconds, the force at the axle varied 

 from 26 to 9 pounds, andin none of them was it nearly uniform. Such 

 variation will affect mf and m" differently ; but possibly the mean 

 results, especially if each experiment lasts for several minutes, will 

 not be much astray. 



I think it may be expected, with some confidence, that the line 

 of research which I have indicated will lead to useful results, both in 

 theory and practice, and give values for these constants, which, if not 

 absolutely exact, will be a close approximation to the truth. The 

 experiments which I propose would not be very costly if the appara- 

 tus were not constructed for permanent use ; and I am not withoxit 

 hopes that I may myself be enabled to execute them. ■ 



