On the Boiling -Points of Homologous Compounds. 371 



The whole subject is one that merits an extended in- 

 vestigation. I have not yet examined the impact of the 

 vapour of one metal against a film composed of another 

 metal, nor have I studied the behaviour of non-metallic 

 surfaces other than glass. 



Some very extraordinary results were obtained in the case 

 of sodium condensed on surfaces cooled with liquid air, the 

 colour of the film changing as soon as the liquid air was 

 removed. I believe that there is a change in the structure 

 of the film with rising temperature, but the phenomena were 

 too complicated and not sufficiently studied to make their 

 discussion worth while at the present time. 



It is my plan to continue the study of the subject with 

 improved apparatus ; for it is obvious that we must be able 

 to control the temperature of the metal, the distillation- tube, 

 and the bulb. It will also be necessary to arrange matters so 

 that the behaviour of superheated vapours can be studied. 



XLI. On the Boiling -Points of Homologous Compounds. 

 By H. C. Plummer *. 



1. TN discussing the relations which subsist among the 

 JL boiling-points of series of homologous compounds, 

 without the guidance of any physical theory, the difficulty 

 is to avoid the representation of a large mass of data by a 

 number of unconnected formulae on the one hand, and on the 

 other the too hasty generalization of some particular type 

 which happens to hold for some special class of substances. 

 The first difficulty is the greater because the observations 

 relate to very short series in many cases, and these are 

 naturally affected by more or less considerable errors. The 

 second difficulty is well illustrated by the early experience 

 of Kopp. 



As an example of the way in which a short series may be 

 represented by a very simple formula, the isoparaffins may 

 be chosen. The formula in this case is 



T=430°'51ogrc, 



where n is the number of carbon atoms. The comparison 

 with observation is shown in Table I. This is satisfactory 



* Communicated by the Author. 

 2 C2 



