T.HE 



LONDON, EDINBURGH, and DUBLIN 



PHILOSOPHICAL MAGAZINE 



AND 



JOURNAL OF SCIENCE 



[FIFTH SERIES.] 



NOVEMBER 1897. 



^*fr 





V 



XLIX. The Surface- Tension of certain Dilute Aqueous Solu- 

 tions, determined by the Method of Rijiples. By N. Ernest 

 Dorset, Ph.D.* 



Introduction. 



PREVIOUS observers have found that the surface-tension 

 of most aqueous solutions is a linear function of the 

 concentration ; but they have worked on no solutions more 

 dilute than about one-half normal. (A normal solution is one 

 that contains in one litre of the solution as many grams of the 

 salt as there are units in the molecular weight of the salt, the 

 molecular weight of hydrogen being taken as two.) It seems 

 therefore desirable to determine, if possible, the surface-tension 

 of more dilute solutions ; and this work was undertaken at the 

 suggestion of Prof. J. S. Ames in the spring of 1895, for the 

 purpose of studying solutions as dilute as possible. Most of 

 the time since then has been spent in studying the various 

 methods, and in perfecting the method of ripples, which 

 appeared to offer fewer theoretical objections than any other. 

 The arrangement of apparatus finally adopted gives an 

 average departure from the mean of several observations of 

 not more than one-seventh of one per cent. Though different 

 samples of water may give results that differ among themselves 

 by more than this amount, still in the entire series of twenty- 

 one samples of water the average departure of a single obser- 

 vation from the mean was but one- fifth of one per cent. This 

 compares very favourably with the results obtained by any 

 other method. 



* Communicated by the Author. 

 Phil. Mag. S. 5. Vol. 44. No. 270. Nov. 1897. 2 E 



