234 PHENOMENA, ATOMS, AND MOLECULES 



matter. Consider, for example, jellies made by amounts of gelatin or soaps 

 in concentrations less than i per cent by weight. The elasticity proves the 

 existence of a continuous frame-work of molecules in contact, extending 

 throughout the liquid. The time of relaxation of such jellies serves as a 

 measure of the rate at which the molecules which form the chains separate 

 or evaporate from one another. Guided by these views, I made some ex- 

 periments, over ten years ago, to determine the diameters of the cross 

 sections of the fibers or rods that must be the elements of the rigid frame- 

 work. For this purpose I made up some dilute gelatin jellies on filter paper 

 and measured the rate at which water could be forced through the jellies 

 by applying a definite pressure. By a modification of Stokes' law which 

 gives the rate of fall of small spheres in liquids, it was possible to calculate 

 the force necessary to move small cylinders of various diameters through a 

 liquid. This law was then checked experimentally by measuring the rate at 

 which water passed through a column of glass-wool having fibers of known 

 size. Applying this law to the case of the motion of water through a gelatin 

 jelly, it was thus possible to calculate the size of the fibers. Only rough ex- 

 periments were made, but the results showed clearly that the diameter of 

 the fibers was approximately lo""^ cm. 



I believe that there is real justification for dealing with molecules of this 

 character, at least as a first approximation, as though they followed the 

 same laws as bodies of large size, such as the fibers of glass-wool. Einstein 

 showed, years ago, that Stokes' law could be used, approximately, for the 

 study of the rate of migration of ions through water solutions. I believe 

 that a thorough quantitative study of the forces necessary to drive water 

 through various jellies should give much valuable information as to the 

 structure of these jellies. 



Very useful pictures of the mechanism involved in the viscosity of 

 liquids and of diffusion in liquids and solids may be had by considering 

 that molecules in contact exist in two states : one in which the surfaces are 

 rigidly connected, the other in which they are entirely free to move. The 

 behavior of the molecules is thus analogous to that of a gas condensing on 

 a solid ; the molecules of the gas strike the surface, remain adsorbed for a 

 certain time, and then evaporate off again. Applying this conception to 

 molecules in liquids, we see that the motion of the molecules past each 

 other, involving viscosity and diffusion, depends upon the relative times 

 during which the molecules are in this rigid, or in this mobile contact. 

 These times can be calculated from an equation of the Boltzmann type in 

 terms of the energy difference between the two states. This theory accounts 

 for the frequent occurrence of temperature coefficients of viscosity and 

 diffusion, which agree with the Boltzmann equation. It seems also to ac- 

 count for the fact that the viscosity of different members of a series of 



