From various sources, we summarize (Figures 31 and 32) the number of 

 density distributions down to about 10 micrometers diameter. It is clear 

 that these concentrations can vary by several orders of magnitude; several 

 different types are probably represented on these figures. The scattering 

 data contain, undoubtedly, both solid particles and microbubbles, although 

 for the larger sizes it is clear from the reports (Arndt and Keller 1976 

 and Keller and Weitendorf 1976) that air bubbles are the main contributor. 

 Gavrilov (1970) and Medwin (1977) measure microbubbles acoustically. Both 

 particulates and bubbles are seen in the holographic studies of Peterson 

 (1972), Peterson et al. (1975), Gates and Bacon (1977), and Katz (1978). 

 From the latter we present an example showing how it is possible to readily 

 detect particu lates in the holographic reconstruction process (Figure 33). 

 We could add to these plots additional data from the St. Anthony Falls 

 Hydraulic Laboratory (e.g., Schiebe 1969), of ocean particulates measured 

 by the Coulter counter (Peterson 1974) and similar measurements by the 

 Naval Ocean Systems Center.* These findings all fall within the bands of 

 data of Figure 31, and they all have about the same functional behavior 

 for the small sizes (even to less than one micrometer for the particulates). 

 This apparent uniformity seems quite remarkable. 



Now that microbubble and particulate nuclei distributions are be- 

 ginning to appear more frequently in the hydrodynamic literature, one may 

 well ask just how these data are to be used quantitatively in cavitation 

 inception studies and in flows with more fully developed cavitation. The 

 several orders of magnitude difference in number density seen in Figure 31 

 suggest that there may be some considerable differences to be expected in 

 cavitation. We have in fact seen such differences appearing on inception 

 forms already in Figure 6; it seems certain that many of the differences 

 seen there are due to the nuclei content. It is also clear that the 

 Caltech HSWT has a very low bubble nuclei population (Figure 32), and that 

 the numerous particulates of that facility do not serve as nuclei. Other 

 investigators have shown the vital importance of microbubble supply (but 

 unquantif ied) in the unsteady flows of model propeller wakes (e.g., 

 Albrecht and Bjorheden 1975, Noordzij 1976). There are other interactions 



*T.G. Lang, private communication. 



55 



