881 
where Ps is the response time of the gauge. This condition 
also enables us to ignore pressure waves which are propagated in 
toward the piston from water beyond the edge of the baffle; since 
these arrive too late to influence the gauge. Finally, it is not 
difficult to construct a gauge which satisfies (19.3). 
20. If the condition (19.3) 1s not satisfied, there is a 
further reduction in pressure due to rarefactions travelling in 
from the region beyond the edge of the block. This region has 
been denoted by C, (see Fig. 3). Let the incident wave be plane 
and neglect the verturbation caused by the block in the region 
Co. Then the pressure there is Pe» the vressure of the incident 
wave. Let the corresnonding velocity be Wee If we assume that 
the velocity of flow across Cy is constant between the edge of 
the block and infinity, the relief pressure due to this moving 
water is 
(20.1) pec [ v(t -00) - Ye (t= a,)] =-Fc v(t - %), 
juet ae the pressure due to the finite baffle is given by (19.1). 
Since the incident wave is assumed to be plane, the material 
velocity is p,/pc. By (20.1) the relief pressure due to the 
moving water becomes 
(20.2) = oil - @) 
The correction (20.2) has been made by Goranson in hie exneri- 
ments (13), There is, of course, more uncertainty in using a 
emall baffle and applying the correction (90.2) than in using a 
beffle large enough to satisfy (19.3). 
eis ee 
