IMPACT OF JETS 373 



(2) Flat Plate 1*15 inn. diameter, ^r in. thick. 



Velocity . 



Ratio 



(3) Flat Plate 2'0 ins. diameter, ^ in. thick. 



(4) Flat Plate 2*0 ins. diameter, ground 1o knife edge. 



(5) Hemispherical cup, 1 in. diameter. 



In the first three of the flat-plate experiments, the most noteworthy 

 features are the remarkable increase in efficiency in each case with a 

 velocity of about 32 feet per second (an effect which is also marked in the 

 case of the hemispherical cup at about 22 feet per second), and the dis- 

 turbing effect of the capillary action at the periphery. This deflected the 

 escaping stream upward through an angle varying from 3 to 6, the 

 value of 1 sin 6 varying between '9477 and '8955. With a sharp- 

 edged periphery the effect of surface tension was less marked, particularly 

 at low velocities, the angle apparently varying steadily from about 5 to 

 2 ((1 sin 0) from '913 to '965) as the velocity increased. 



It will be noticed that, except at very high velocities and at the " criti- 

 cal " velocity, the smallest was quite as efficient as the two larger plates. 



In the case of the hemispherical cup, the escaping stream, which 

 theoretically should have been vertical, showed the effect of capillary 

 attraction by being inclined at 13 to the vertical. On taking this into 

 account, the ratio of actual to theoretical pressure is increased by about 



