1259 
neglect entropy changes. It is then possible to write 
for certain fluids the following common equation of speea’ 
p ve =k (Vel 
Here p=p +r. p is an effective pressure, p is the 
overpressure, and WT is a constant, characteristic of the 
fluid (for example, in perfect gases 1 = 1 atmosphere 
and in water, Tr=3 x 10° atmospheres) ; v is the 
specific volume; y is also characteristic of the fluid 
(for gases it is the adiabatic constant, lying between 1 
and 5/3, and for water it is approximately 7.1 5), k 
is another constant characteristic of the fluid. The 
constants appearing in (7.1) are actually functions of 
the entropy. Here it is assumed that the entropy is 
@ constant on both sides of the shock front and that it 
is continuous across this front. It will be shown ieter 
(par.9) that this anproximation is adequate for the problem 
at hand. Hence (7.1) is assumed to hold everywhere in 
this report. 
8. In this report two methods are used. One method 1) 
is numerical; it is described in part III. The other 
method is analytical; it is described in part II. The 
numerical method allows the entropy to change across the 
shock front, but nowhere else. Except for this approexi- 
mation it is completely rigorous. In the analytical 
méthod, on the other hand, there are several approximations 
which are discussed in part II. These approximations are 
