Nipher — Method of Measuring Pressure on a Structure. 19 



result by more than a fraction of one per cent. After several 

 unhappy days, in which the advisability of publishing any- 

 thing concerning the disk collector was under serious consid- 

 eration, it was finally discovered that in reducing from 

 grammes per square centimeter to lbs. per square foot, a 

 factor 0.1174 had been used, instead of 0.1024. 



An inspection of Fig. 3 will show that there is some evi- 

 dence of a minimum of pressure at the center of the front 

 face. This may be due to the effect of the hinges which cover 

 the horizontal rows b and h, although this is not regarded as 

 very probable. It is not probable that this is wholly due to 

 errors in observation. 



The results shown in Fig. 3 also furnish a method of deter- 

 mining a value of considerable historical and theoretical inter- 

 est. The average pressure on a board held normally in any 

 fluid stream, due to the compression on the front and rarefac- 

 tion on the back side, is 



F = k\v> (3) 



instead of being represented by (1). The value K is inde- 

 pendent of any friction or viscosity in the fluid. It is the 

 same for water as for air. It involves simply inertia in resist- 

 ing a change of direction in passing the edge of the plate. 

 Many determinations of this value have been made. In some 

 cases in which the stream was water, the pressures before and 

 behind the board have been measured by pressure columns.* 



The values obtained by various experimenters are given in 

 a table. The rotation experiments of Borda, Hutton and 

 Thibault are said by Unwin to have been made on a long arm, 

 but the lengths are not given. In Langley's experiments, 

 the radius was 9 meters. His result is the mean of 14 deter- 

 minations made with an automatic recorder. 



If we may assume the value 7.17 lbs. per square foot at 

 the middle of the front side of our pressure board (Fig. 3) 

 to be the pressure due to the velocity with which the board 

 moves through the air, computed by Newton's theorem, we 



* Hydromechanics. End. Brit. 517.— Report Chief Signal Officer, 1887. 

 2 : 223. 



