32 



C. L. SADRON 

 , [7,]= 0795 A (p) 



Fig. 16. There is no simple correlation between M w and fol; O = category Ci and 

 # = category C» . 



values of M, if there is a univocal relationship between M and the morpho- 

 logical parameters on which S is dependent, that is to say between M and 

 S. It is clear that if S is a function of M, the polydispersity of S is due to the 

 polydispersity of M, and that the law of distribution g(M) of M can be 

 derived from the experimental law of distribution of S. 



b. Comparison with Other Results 



(1) Light- Scattering and Viscosity. If there exists such a relationship be- 

 tween M and the shape and dimensions of the particles, the measured values 

 of [??] for different samples should be a simple function of the measured 

 weight. 



Figure 16 shows that such a correlation apparently does not exist. The 

 points giving [77] as a function of M w are dispersed in the part of space con- 

 tained between the two curves Ci and C 2 . Such a dispersion appears still 

 more accentuated from the results collected by J. Hermans. e It is difficult 

 to avoid the conclusion that particles of different shape and dimensions may 

 correspond to a given value of M. 



(2) Sedimentation and Viscosity. This conclusion should lead us to think 

 that the use of hydrodynamical methods is, at the present moment not 

 profitable and that we have to wait for further developments before being 

 able to derive from it more precise conclusions. However, it is worthwhile 

 to mention some interesting attempts made by different authors, espe- 



23 P. Doty, J. Cellular Comp. Physiol. 49, 27 (1957). 



