35 119 
(23) Measurements of the Velocity of the Pressure Wave. 
The velocity of the pressure wave was measured by means of the spark 
chronograph described in Section 27. As the pressure wave passed two contacts at 
measured distances from the charge it closed each in turn and operated the chrono- 
graph. The lay-out of one of these experiments is shown in Fig. 43, and the circuits 
iu Fig. 44. Each contact consisted of a pair of brass springs built up with rubber 
strips into a watertight unit (Fig. 45). ‘he time for tiie contact to close under a 
pressure of l’ton per square inch was estimated to be about 3 x 10° second. ‘The 
two contacts were lashed to a hemp line at distances of 50 feet and 60 feet from the 
charge. It will be seen from the lay-out diagram that as the ‘‘ Malapert”’ drifted the 
resistance of the charge kept the hemp line taut. Above and between the contacts 
was a floating box containing the chronograph and its associated sparking-circuits. 
This box was connected to the “ Malapert” by a 4-wire cable, so that the charging of 
the spark condensers and the speed of the chronograph could be controlled from the 
ship. ‘The firing of the shot was delayed until the condensers had been charged and 
the chronograph brought to correct speed. Each condenser charging circuit included 
a high resistance, which gave it a charging period of several seconds, so that the 
chronograph would be unaffected by the rapid reclosing of a contact, such as might 
be occasioned by the reflected wave from the bottom. For the sake of simplicity only 
two contacts have been shown in Fig. 44, but actually these contacts and their 
associated circuits were in duplicate, so that as the pressure wave passed each contact 
position it operated two contacts and made a double record on the chronograph. The 
results of four shots, in all of which the contacts were at 50 feet and 60 feet from a 
300-Ib. amatol charge, were as follows. The mean result is probably within 5 per 
cent. of the truth. : 
Shot. Velocity (feet per second). 
72 & : = 2 - - 5,180 
74 - - * - = - 4,820 
105 - - - - - - 4,830 
106 - - - - - - 4,730 
Mean 4,890 
The velocity of sound in water was determined by Colladon and Sturm in the 
Lake of Geneva in 1826 as 4,700 feet per second. The modulus of compressional 
elasticity— 
pdp 
dp 
of sea-water is 5} per cent. higher than that of fresh water \ Tait, Challenger Reports, 
18838) and its density (p) is 3 per cent. higher, so that the velocity of sound— 
dp 
dp 
should be 2? per cent. higher, or 4,830 feet per second. The experiments of 
Mr. Boulding, off Culver, Isle of Wight, in 1917 gave the velocity as 4,940 feet per 
second (B.1.R. 34555,17). The water in the Clyde estuary between Arran and the 
mainland has nearly the full salinity of the opeu sea (three samples collected on 
different days gave 5°18, 3°16, and 3°20 grams NaCl per 100 c.c.), so that the velocity 
there may be taken as about 4,9U0 feet per second. This is practically identical with 
the figure found for the velocity of tle pressure wave. 
The velocity of sound in a medium is the velocity of an infinitely weak pressure 
wave. The theory of the velocity of a strong pressure wave has been considered by 
Rayleigh (Theory of Sound, § 251), whose observations may be paraphrased as 
follows :---Consider a small portion of the medium occupied by the pressure ware 
at a given moment, and imagine an infinitely weak secondary wave to be superposed ; 
the secondary wave will travel, relatively to the medium, with the velocity a’ of 
sound—not the velocity of sound in the undisturbed medium but the velocity of 
sound in the medium.as modified by the pressure of the main wave; but the medium 
itself is in motion with a velocity u, depending on the pressure of the main wave, 
so that the whole velocity of the secondary wave is a’ + uw; what has been said of 
the secondary wave applies also to the parts of the main wave, so that the velocity 
of the main wave at the point considered may be taken as a’ + u. We have to apply 
E2 
