19'4-] 



ATMOSPHERIC PRESSURE. 



73 



of 15 cm. (5.9 in.) apart. The vertical test-wire was then placed at 

 the spot where the wind was to be measured, and heated by electric 

 current. It thus served to measure horizontal wind-velocity in any 

 direction. 



Fig. 12. Connections of Test-Wire for Indirect Measurements of Wind 



Velocity. 



Two methods were used, one indirect-reading, the other direct- 

 reading. In the indirect method, the connections were as shown in 

 Fig. 12. Here the test-wire, shunted by a voltmeter, is placed in a 

 Wheatstone bridge of unequal arms, so that the current in the test- 

 wire side is ten times stronger than in the opposite side bd. The 

 bridge is set for balance at a predetermined resistance and temper- 

 ature of the test-wire. Whatever may be the horizontal velocity of 

 the wind blov^ing over the test-wire, there is some current strength 

 supplied to the bridge through a meter A, which will restore balance 

 and zero current in the galvanometer G. When this balance is ob- 

 tained, the readings of the ammeter A, and voltmeter V, are noted. 

 Their product VA; or the voltage square V^, is proportional to the 

 power dissipated in the 15 cm. of test- wire, from which the velocity 

 of the wind can be deduced with the aid of suitably prepared tables 

 or curves. The advantage of this method is its relatively high pre- 

 cision. Its disadvantages are that it requires to be adjusted for each 

 observation, and in gusty winds, it is not possible to secure a Wheat- 

 stone-bridge balance long enough to obtain readings of either V or A. 



