from this assumption cannot be tolerated. Depending upon tlie condition of the liquid (air 

 content and air or vapor nuclei which may exist in a stable condition or be stabilized on 

 solid particles, etc.), cavitation may begin above or below the vapor pressure. The role of 

 air and vapor nuclei in the inception of cavitation in both pure and contaminated liquids was 

 discussed briefly in Reference 3. Some effects associated with turbulence in a free stream 

 and in a boundary layer, as well as conditions in a boundary layer that are of some importance 

 in cavitation problems were also discussed. In the present report, these remarks will be elab- 

 orated and extended somewhat, and some recently reported work relating to scaling of incep- 

 tion will be reviewed. 



The problem of the limiting tensions which can be developed in liquids will not be 

 discussed in this report— some of the recent ideas and methods having been reviewed briefly 

 in Reference 3. Although these ideas, based on so-called "hole" and "nucleation" theories, 

 lead to predictions that are of the order of magnitude of available experimental results, the 

 theoretical models and approximations are not complete and, at best, can only result in prob- 

 able predictions (in the statistical sense). In view of the experimental complications, e.g., 

 the ever-present uncertainties associated with the behavior of the liquid at a wall, and of the 

 properties of the wall itself, and the lack of sufficiently detailed theory needed for design of 

 critical experiments, a completely satisfactory prediction procedure cannot as yet be estab- 

 lished. Nevertheless, these ideas and methods are of immediate importance in helping to 

 gain insight into the processes involved in cavitation even in "technical" liquids, and of 

 continuing interest from the standpoint of eventual contribution to a unified theory of the 

 mechanism of inception in any liquid, pure or contaminated. 



FURTHER REMARKS ON THE ROLE OF TURBULENCE AND 

 BOUNDARY LAYER EFFECTS IN CAVITATING FLOWS 



In Reference 3, a rough calculation was carried out to illustrate the possibility of 

 cavitation associated with the pressure fluctuations in a turbulent flow. It was shown, that 

 in a boundary layer, the velocity fluctuations are such that the instantaneous pressures may 

 vary sufficiently from the average value to result in a "microscale" cavitation which would 

 occur in a more-or-less random fashion well before it could be observed visually. Thus, it is 

 of some interest to be able to estimate the magnitude of the pressure fluctuations in turbulent 

 flows of both the boundary layer type and wake and jet type. In these cases, however, the 

 complete relations of turbulence phenomena to cavitation inception cannot be established 

 until more is known of the turbulence mechanism itself. Although a great deal of work has 

 been done in describing the velocity fluctuations, little effort has been devoted to investiga- 

 tions of the pressure fluctuations. The extreme difficulty of the latter problem has restricted 

 the work so far to isotropic turbulence and only statistical averages can so far be obtained. 

 For the cavitation problem, it will probably be important to know in somewhat more detail the 

 values of the pressure fluctuations, the scale(length), and ultimately a time scale. 



