16 



remarks on these problems will be made. 



So far, the transient cavities and the steady-state cavity flows have been discussed as 

 entirely unrelated problems from a hydrodynamical point of view. One of the interesting and 

 practically important questions which has not been considered to any extent is that of the 

 conditions (both geometrical and physical) under which cavitation will occur in the form of a 

 large steady-state cavity or a mass of small, oscillating bubbles. In Reference 3, the writer 

 expressed the view that the appearance of the cavitation is associated in part with the pres- 

 sure gradients, but no satisfactory criteria are available as to the initial appearance or transi- 

 tion from transient cavities to a large, steady-state cavity. A cavitated region made up entire- 

 ly of transient bubbles may exhibit the properties of a steady cavity in that the average envel- 

 ope of such a region does not vary with time. Another case of such "steady-state" cavities 

 in which the average envelope remained unchanged, but in which rapid surface oscillations 

 were observed without clear evidence of individual bubbles was discussed in Reference 3 in 

 connection with the study of the development of cavitation in wakes. 



This case of cavitation in wakes raised the question of the processes involved in re- 

 moving the liquid phase from the cavity and the spread of cavitation from the wake boundaries 

 into the interior. It was pointed out in Reference 28 (and also discussed in Reference 3) that 

 in the studies reported therein the cavities contained large quantities of liquid phase and that 

 the liquid was not removed even at the lowest cavitation number reached (0.116). An example 

 of such a flow is shown in Figure 4, which was obtained during the experiments reported in 

 Reference 28. Subsequent experiments have disclosed certain cases in which the liquid phase 

 was removed from the cavity. Under conditions not yet defined or even understood the liquid 

 could be seen to move rearward out of the cavity. This process occurred at a speed slow 

 enough to be followed by eye, leaving a fairly transparent cavity wall. This phenomenon has 

 also been observed in the water tunnel of the Ordnance Research Laboratory at the Pennsyl- 

 vania State College. Just what conditions determine the point at which equilibrium is upset 

 between vaporization and possibly shearing motion at the boundary and the replenishment 

 with re-entrant fluid from the tail of the cavity are not clear and require further investigation. 



Figure 4 - "Steady-State" Cavity 

 behind a Hemisphere 



The photograph was taken with an exposvire time 

 of 1/10,000 second. 



