1905.] Distribution of Chlorides in Nerve Cells and Fibres. 189 



which specially distinguish it from solutions of inorganic compounds in 

 respect to osmotic pressure, electrical conductivity, cryoscopy and the effect 

 exercised on transmitted light. These properties, and particularly the facts 

 obtained by the use of the ultramicroscopic apparatus on " solutions " of 

 colloids, compel, as it were, the conclusion that the necessary feature of such 

 " solutions " is a state of suspension of particles of a magnitude beyond the 

 limits of microscopic vision.* It is not necessary to discuss the cause of such 

 a state of suspension or why it is indefinitely maintained without any trace of 

 sedimentation, as the force or forces concerned, whether electrical, as Bredig, 

 Hardy and Billitzer hold, or otherwise, do not immediately affect the question 

 of the physical division of the colloid in the fluid, and we have now only to 

 deal with the problem of the structural character of the particles in suspension. 

 How these particles in their contact relations with the molecules of water 

 differ from inorganic compounds in the same situation is not determined, but 

 the dominant view is that the colloid particles are not single molecules but 

 aggregates of such molecules and, of course, this excludes the possibility of an 

 intimate association of the molecules with those of water. 



It is, however, not to be supposed that the molecules of water do not 

 penetrate the aggregates, for it is not, otherwise, possible to understand why 

 the particles in organosols should exhibit such a degree of transparency in the 

 ultramicroscopic apparatus, or why the addition of certain non-electrolytes 

 should bring about precipitation of the particles. Indeed, one of the two 

 phases possible in the relations of colloids to water necessarily postulates the 

 presence of water molecules in the interior of the colloid particles. 



The superficial portions of such particles must be regarded as exercising a 

 considerable check on the process of diffusion from, or into, their interior, as 

 the results of efforts made to obtain chloride-free globulins and albumins 

 show. As already pointed out.f at least eight precipitations with ammonium 

 sulphate are required to free the proteids of egg-albumen and serum from 

 the associated chlorides, and this number suffices only if the precipitate after 

 each precipitation is dissolved and kept in solution for a day at least before 

 the next precipitate is brought about. When a much shorter time was 

 allowed, for example, a couple of hours, the proteid of the twentieth 

 precipitate still contained demonstrable quantities of chlorides. 



These results seem to indicate that a part of the chlorides present is in 

 such intimate association with the colloid particles that when these are 



* Zsigmondy, " Ueber Kolloidale Losungen," ' Zeit. fur Elektrochemie,' 1902, No. 32 ; 

 Zsigmondy und SiedeDtopf, ' Annalen der Physik.' vol. 10, p. 1, 1903; also Gatin- 

 Gruzewska and Biltz, ' Arch, fur die ges. Physiol.,' vol. 105, 1904, p. 115. 



t Macallum, 1 Roy. Soc. Proc., ! B, vol. 76, p. 217, 1905. 



