632 Dr. Allan Ferguson on the Drop- Weight 



of high atomic weight on account of the comparatively 

 greater scattering. 



The results, though incomplete, seem to support the theory, 

 the values of I/3 3 for tin being consistently slightly lower 

 than those for mica ; whilst the values for any one substance 

 are approximately constant, but increase somewhat for the 

 high velocity lines. 



In conclusion, I wish to express my thanks to Dr. Bohr 

 for suggesting these experiments, and to Professor Sir 

 Ernest Rutherford for his interest and advice throughout the 

 course of the work. 



The University, Manchester, 

 28th July, 1915. 



LXIV. On the Drop- Weight Method for determining Surface- 

 Tensions. By Allan Ferguson, D.Sc. (Lond.), Assistant 

 Lecturer in Phi/sics in the University College of North 

 Wales, Bangor*. 



IN a recent article f I have given a critical account of 

 some twenty methods for the determination of surface- 

 tensions, and have endeavoured to discriminate between them. 

 It was there pointed out that only those methods which were 

 quite independent of the contact-angle could safely be used 

 as instruments for research purposes, and therefore that the 

 results obtained by the use of the classic " capillary-rise " 

 method were of very variable value. It is quite possible that 

 the great majority of organic liquids in contact with air or 

 with their own vapour may have zero contact-angles with 

 glass; but the fact requires exact experimental verification, 

 and should certainly not be assumed on the strength of more 

 or less doubtful "glass-wetting" phenomena. Otherwise 

 there is an obvious danger that in discussing, say, the 

 question of molecular complexity, the results may be quite 

 vitiated by the presence of an unlooked-for cos term. 



Take, for example, the equation of Eotvos, in the simple 

 form as originally used by Ramsay and Shields J, viz.: — 



In the capillary-rise experiments the value obtained for the 

 surface-tension is really T cos 6 ; and if 6 be finite the value 

 of k as found must be divided by cos 6 in order to obtain its 



* Communicated by Prof. E. Taylor Jones, D.Sc. 

 t ' Science Progress,' Jan. 1915, p. 428. 

 t Phil. Trans. A. p. 647 (1893) 



