io6 SCIENCE PROGRESS 



of the part it plays in the Acheson carbon used as a lubricant. 

 The preparation of colloidal carbon has been recently investi- 

 gated by Thorne {Trans. Chem. Soc. 109, 202, 191 6), who 

 employed Sabbatani's method. Thorne concludes that al- 

 though some colloidal carbon is produced by this method, its 

 properties are obscured by more complex products of decom- 

 position, the presence of which is obvious from the odour 

 of burnt sugar. Further, these products do not act as stabilisers 

 for the true colloidal form. The conclusion reached is that 

 in pure water, stable colloidal carbon cannot be produced. The 

 presence of alkali aids the stabilisation. Certain organic sub- 

 stances also assist the stable production. In the Acheson 

 process, tannic acid is the stabilising agent, but its precise 

 action has not yet been determined. 



Whilst considering disperse systems it may not be out of 

 place to refer to an important technical application of the 

 property possessed by fine particles, namely their electric 

 charge. This is exhibited not only by colloidal metals and 

 non-metals in liquid media, but also by dust particles and 

 fumes in air. By making use of this fact, the precipitation of 

 smoke, fume, and dust can be successfully carried out on a 

 considerable scale. Details will be found in two papers, viz. 

 L. Bradley (Met. Chem. Eng. 13, 911, 191 5), and F. G. Cottrell 

 {Eng. and Min. Journ. 101, 385, 19 16). 



Continuity of state. — A valuable service has been rendered 

 by F. II. McDougall (/. Amer. Chem. Soc. 38, 528, 1916) to 

 the study of the continuity of state by his exposition of 

 Dieterici's equation of state. Up to the present time practically 

 only one equation, that of van der Waals, has been subjected 

 to close and searching examination, with the result that almost 

 too much stress has been laid upon this particular form of 

 representing the behaviour of liquids and vapours. As a matter 

 of fact Dieterici's exponential expression (or rather both of his 

 equations, for he has proposed two forms) reproduce certain 

 experimental relations with much greater fidelity than does 

 van der Waals' expression. This is particular^ true of the 

 ratio RTc/PcV,. where the suffix c denotes critical values. On 

 the other hand the Dieterici relation v c = 26, is hard to reconcile 

 with experiment, being even farther from the truth than van 

 der Waals' v c = 36. The new examination of Dieterici's equa- 

 tion deals largely with the significance of the term b, which is 



