232 § ANNUAL REPORT SMITHSONIAN INSTITUTION, 1944 
of various sizes are filled in advance with TNT compacted firmly. A 
hole is punched in the center of the lid of each can. Fuses of various 
lengths are attached to the detonators and this junction of the fuse 
and detonator must be watertight. Just before an R. A. R. position is 
required, the detonator is inserted through the hole in the lid of the 
can and pushed down into the TNT. The fuse is lighted on an electric 
heating element and the bomb is thrown overboard well clear of the 
ship, which is under way. 
The bombs must not be exploded too close to the survey ship. Also 
they should not explode too close to the surface of the water, for then 
part of the sound energy is dissipated into the air. Best results are 
apparently obtained with explosions at a depth of about 7 fathoms. 
To achieve these two results, the bombs are weighted to make them 
sink at the required speed, and the fuse is cut in lengths to provide the 
required delay in time. 
In 1940 the cost of a 1-pint bomb with fuse and detonator was about 
30 cents. 
VELOCITY OF SOUND IN SHA WATER 
To measure distances by subaqueous sound transmission, one needs 
to know not only the elapsed time intervals but the effective horizontal 
velocity at which the sound travels through the water. The velocity 
of propagation of sound in sea water may be calculated from the tem- 
perature and salinity of the water and the hydrostatic pressure. 
Tables have been prepared based on these three variables. The velocity 
of sound varies with these three characteristics by the following ap- 
proximate percentages: 
(a) Each 1° C. increase in temperature increases the velocity 0.2 
percent. 
(b) Each 1 part per 1,000 increase in salinity increases the velocity 
0.1 percent. 
(c) Each additional 100 fathoms (183 m.) of depth increases the 
velocity 0.22 percent. The velocity of sound in water is approximately 
1,500 meters per second at a temperature of 14° C., salinity 35 parts 
per 1,000, and at surface atmospheric pressure. ‘ 
To determine the velocity of sound from the tables, the temperature 
and salinity of the water must be measured, and the depth must be 
known, for pressure varies almost exactly with depth. The variation 
of salinity in sea water is small, and its effect on velocity, as compared 
with the effect of variation in temperature, is relatively unimportant. 
The temperature varies not only from place to place, but ordinarily 
decreases with the depth. For the average R. A. R. survey, the velocity 
of sound must be known within 4 meters per second, and to attain this 
