242 H. A. Hazen — Wind Velocity and Pressure. 



or any other object, the resistance of the air against which ho 

 wished to learn, and observed the uniform velocities of the fly- 

 wheel under the action of different weights. He used plates 

 9-59, 6-39 and 4'26 in. on the edge, and weights of 8*8, 44, 2*2, 

 1*1 and "55 pounds, in generating velocities. The observations 

 may be very satisfactorily united into the following formula : 



^=(-0031 + -00035c) SV 2 



In which p — pressure in pounds; C= contour of plate in ft.; 

 S= surface in sq. ft.; V= velocity in miles per hour. This 

 notation will be preserved throughout this paper. The formula 

 may be converted into velocity in ft. per second by multiply- 

 ing by (22-15) 2 . 



Hageii's experiments. — By far the most careful experiments 

 with a whirling machine were those of Hagen in 1873. The 

 arm in this case was 8 ft. in length. The velocity ranged from 

 1 to about 4 miles per hour. " The room seems to have been 

 quite small, as, with the latter velocity, the air was set in feeble 

 rotation. The plates ranged in size from 4 to 40 sq. in. area. 

 He found, as in the case of Borda, that the pressure per sq. ft. 

 increased with the larger plates. His formula is : 



_p=(0029+-00014c) SV 2 . 



These experiments were conducted with so great care, and so 

 complete an elimination of all apparent sources of error, that 

 they alone were depended on by Prof. Ferrel in his recent dis- 

 cussion of this question. See Annual Report Chief Signal 

 Officer, 1885, Part ii, pp. 314 and 407. 



Experiments at Metz, France. — MM. Piobert, Didion and 

 Morin, at Metz, used a plate of 10*764 sq. ft. area, allowing it to 

 drop vertically a distance of nearty 47 ft. in a foundry. The 

 times, spaces, and velocity of fall were measured electrically, 

 and the following approximate pressure of the air was obtained : 



p= (-002 + -.0035) SV 2 . 



To test the effect of the size of the plate, another series of results 

 was obtained with the above plate, and another of ^ its size. 

 The larger plate required a little longer time in falling through 

 any given space than the smaller, hence the pressure per sq. ft. 

 upon it must have been greater than on the smaller. This is 

 in the same direction as the results of Borda and Hagen, but 

 the difference was very slight and the conditions were quite un- 

 satisfactory. The short distance and time of fall preclude any 

 but the most general conclusions. 



Experiments in Washington. — In Nov., 1886, it was decided 

 to make a short investigation on this subject. It was necessary 

 to adopt the simplest and cheapest possible apparatus. At- 



