1260 
apparently satisfactory, however, because agreement bexween 
the methods is good, as one can see in the next paragraph. 
9. In fig. 2 = 4 five examples 4) of a pressure= 
time curve are given. These curves show the pressure at the 
wall during reflection. The three linear cases were 
worked out by both the numerical and the analytical 
methods. The two exponential cases, however, were computed 
by the numerical method only. The exponential cases 
were calculated because the experimental pressure-time 
curves are more nearly exponential than straight. The 
underwater shocks were chosen of the strengths indicated, 
because their strengths probably bound the range of most 
interest. A shock of 300 atmospheres is of about the 
violente ordinarily measured by underwater gauges. On the 
other hand, it would be difficult to realize the reflection 
of a shock of strength exceeding 2000 atmospheres from a 
rigid surface, because no surface would remain rigic under 
such a blow for long. (The elastic limit of mild steel is 
not more than 2000 atmospheres.) Calculations like the 
ones in this report for shocks of greater strengths than 
2000 atmospheres can be interpreted, however, as relating 
to collisions of equal and opposite shocks, rather than 
to impacts between a single shock and a rigid surface. 
The curves in fig. 2-4 are described in the following 
table. 
== 
