RATE OF DIFFUSION OF IODINE IN KI 501 



Taylor and Rankin (1903). Their results, so far as comparable, agree 

 fairty well with ours; but as he worked with only three concentrations his 

 data was not sufficient for our purposes. Getman (1906-1908) also 

 determined the viscosity of KI solutions of different concentrations; but 

 his work was done at 18°. 



In drawing the curves for fluidity, density, and diffusion, the axes were 

 so shifted as to give a set of curves that could all be drawn on the same 

 sheet for the sake of comparison. A full discussion of these curves will be 

 found in the proper place. 



DISCUSSION OF RESULTS. 



General confirmation of the work of Van Name and Edgar. If we exam- 

 ine table IX, we see that there is no apparent discrepancy between the 

 relative increase in the rate of reaction velocity with increasing concen- 

 tration of KI and that of diffusion under the same conditions. This 

 becomes clearer if we examine the curves of these quantities. In fact we 

 may say that within experimental error the curves for diffusion and for 

 reaction velocity are parallel. (It is to be noted that the scale on which 

 these curves are drawn magnifies the discrepancy.) It is to be regretted 

 that the work done by Van Name and Edgar, and by Van Name and 

 Bosworth, do not give us any points on our curve beyond a concentration 

 of 2.5 N, and hence, we can only guess as to the probable form of the curve 

 at higher concentrations. It would be very interesting and instructive to 

 obtain a curve for reaction velocities covering all concentrations covered 

 by the diffusion or viscosity data. As it is, there is no way to know if 

 the reaction velocity curve continues to rise or Uke the diffusion curve 

 falls or becomes flat at higher concentrations. In the absence of evidence 

 the latter seems the more probable, for two reasons; (a) the velocity curve 

 though still rising, is rising much more slowly near the end, (b) the re- 

 action velocity curve has approximately the same form as the diffusion 

 curve so far as it is kno'ma, and hence it seems reasonable to assume that 

 it will probably continue to have the same form. 



General theory of diffusion. The phenomenon of diffusion may be 

 defined in three ways according to the point of view, viz. : 



1. As an empirical fact. 



2. As a consequence of the kinetic theory. 



3. As a result of the second law of thermodynamics, i.e., from the view 

 point of energy relations. 



1. Diffusion may be defined as the mixing which takes place when two 

 dissimilar but miscible liquids are left in contact for some time under 



