204 REPORTS ON THE STATE OP SClENCi!. 



' protective colloids ' which prevent precipitation by electrolytes. I'hiis 

 silver halides may be obtained in colloidal solution in presence of gelatine, 

 a fact of great importance in photographic-plate making. An interesting 

 point is that in many cases the nature of a sol may be changed by 

 oxidation, reduction, or chemical substitution without destroying its 

 colloidal character. 



An important method of producing metallic sols is that of Bredig,' 

 which differs in many respects from those just described. He employs 

 the metal as cathode of an electric arc formed under water, which forms 

 a sol with the particles projected from the cathode. Svedberg^ has 

 applied the same method to the productioji of so's in ether, isobutyl 

 alcohol, and other organic liquids, and of carbon, silicon, sulphur, and 

 phosphorus in water.^ Bredig and Haber ' have also shown that under 

 certain conditions sols can be produced by electrolysis, especially in 

 alkaline solution. 



The ultra-microscope of Siedentopf and Zsigmondy '^ has rendered 

 many details of colloid sols accessible to direct observation. In its 

 original and most complete form it consists of a rectangular cell on the 

 stage of the microscope, through the side of which a powerful horizontal 

 beam of light is concentrated by appropriate lenses and adjustable slits ; 

 but several other types have been constructed for special purposes. No 

 light can reach the observer but that from the brilliantly illuminated 

 particles, which are thus rendered visible irrespective of size, as stars are 

 visible to the naked eye on which the most powerful telescope can raise 

 no measurable disc. As the field of view, and the vertical thickness of 

 the beam can be accurately measured, it is possible to count the visible 

 particles in a given volume of liquid, and as the total concentration is 

 known, their weight (and on certain assumptions as to form, their size), 

 can be calculated. While the diameter of the smallest particles which 

 could be seen by ordinary illumination, even with a magnifying power 

 of 2250 diameters, is about 140 /i/t (mm. xlO'^') Zsigmondy has ob- 

 served particles in gold sols of from 20 to 80 *(/i (submicrons) while still 

 smaller particles (amicrons) exist, which scatter and polarise light, but 

 cannot be separately distinguished. Zsigmondy estimates the size of the 

 smaller of these amicrons as from 1-7 to 3 /</j (0-0000017 to 0-000003 mm.). 

 Ostwald '' gives the probable size of a hydrogen molecule of molecular 

 weight 2 as about 0-16 fifi. As organic substances certainly exist having 

 a molecular weight of several thousand, it seems quite probable that 

 some of them may be visible to the ultra microscope. Vanino and 

 Hartl ^ have shown that if a small quantity of a ready-formed gold sol 

 be added to a solution of a gold salt and a reducing agent, the colloid 

 particles act as nuclei for the reducing gold. It is thus possible to build 

 up amicrons to sub-microns which can be seen and counted. Some similar 

 action takes place in gold ruby glass, which has been shown by Zsigmondy 

 to owe its colour to colloid gold particles, but which is sometimes colour- 

 less until, by reheating, the gold germs have been allowed to grow liy 

 absorbing amicrons or reducing gold salts. 



' ^./. Anffew. Ch., 1898, 951-954; Z.f. Elektr., 1898, 4, 514-515. 



- Ber., 1905, 38, 3616-3620, 



» LAd., 19U6, 39, 1713. * Ibid., 1898, 31, 1741. 



'■" ZuT JE-rhenntniss der Eollmde, Jena, 1905. 



"■ Grmidius, p. 82. 



' Bcr., 1906, 39, 1696 -1700. 



