VISCOSITY OF PROTOPLASM 297 



exhibit. In fact, whereas doubling the viscosity of a solution of 

 Sodium Caseinate by the addition of protein does not measurably affect 

 its conductivity, doubling its viscosity by the addition of forty per cent, 

 of alcohol reduces the mobility of the caseinate ions to one-half, and the 

 conductivity of the solution to a still smaller proportion. In estimating 

 the influence of viscosity upon the mobilities of protein ions we can 

 entirely disregard that portion of the viscosity of the solution which, 

 although comparable in magnitude with the viscosity of the solvent, 

 is attributable to the protein itself. 



There are thus two kinds of viscosity which may be displayed by 

 solutions, the one which impedes the motion of molecules or ions, and 

 the other which does not hinder the motion of such small particles, 

 although it does very greatly impede the passage of the fluid through 

 a narrow tube or the rate of oscillation of a rotating disc suspended 

 within the fluid. The former type of viscosity is displayed by solutions 

 of inorganic substances and the simpler organic substances, the latter 

 type of viscosity by solutions of the proteins. 



The customary method of measuring viscosity, such as the measure- 

 ment of the time taken by a given volume of fluid to pass, under the 

 force of gravity, through a specified length of a narrow tube, all involve 

 deformation of the fluid, whereas the estimation of viscosity which 

 depends upon the diffusion of molecules or ions through it, does not 

 require any displacement of the particles of the solvent in which the 

 diffusion is occurring. Deformation is especially resisted by protein 

 solutions, but internal molecular motions are not impeded. This fact 

 strongly suggests the existence of a Structure within solutions of the 

 proteins. It appears very probable that the molecules of protein in 

 solution are loosely connected with one another so as to form a mesh- 

 work or three-dimensional net throughout the body of the solution. 

 Such a net, which, in two-dimensional section, we may picture as some- 

 thing analogous to a tennis-net with microscopic or ultramicroscopic 

 meshes, would offer no hindrance to the passage through it of a quickly- 

 moving body which is much smaller than its meshes, but to any force 

 involving deformation of its structure, for instance to a force tending 

 to drag it through a small tube, it would offer a very considerable 

 resistance. In measuring the resistance which a protein solution 

 offers to passage through a capillary tube, we are not measuring true 

 viscosity or internal friction between adjacent molecules, therefore, 

 but the resistance of the structure of the solution to deformation. 



A common method of measuring the viscosity of fluid consists in 

 suspending a disc within the fluid and causing it to oscillate, the decrease 

 of the rate of oscillation being a measure of the viscosity. 



When this method is applied to protein solutions, however, it is 

 found that the decrease in the rate of oscillation of the disc is abnormally 

 rapid, but if the liquid be slightly shaken or the disc taken out and 

 replaced, the decrease in the rate of oscillation becomes normal again 

 for a brief period. Evidently the protein network adheres to the disc, 



