244 Professor A. W. Bikher [March 20, 



the properties of which are different from that of the liquid in the 

 interior, and to which rather than to any ideal geometrical boundary 

 the term surface might be applied. It may, however, prevent con- 

 fusion if it is called the surface-lmjer. 



Many attempts have been made to measure the thickness of the 

 surface-layer. In particular, M. Plateau studied a thinning soap film 

 with the view of determining whether or no the pressure exerted on 

 the enclosed air by the film when very thin is the same as when it is 

 comparatively thick. Had any such difference been observed it might 

 but have been taken as prima fade evidence that the tenuity was so 

 great that all the interior portions of the film had drained away, and 

 that the thickness did not exceed that of the two surface-layers. 



This experiment has been criticised by Prof. Eeinold and mys.elf, 

 but it is not intended in this lecture to enter upon the general question 

 of the thickness of the surface-layer, or the interesting theoretical 

 problems w^hich are closely connected with it, as we are at present 

 engaged in an investigation which we hope may throw further light 

 upon the subject. There are, however, two preliminary questions on 

 which we have arrived at definite conclusions. 



In any experiments which have for their object the detection of 

 small changes in the properties of a soap film as it becomes thinner, it 

 is essential that we should be able to assert with certainty that no 

 causes other than the increasing tenuity have been in play, by w^hich 

 the effect looked for might either be produced or masked. Changes 

 in the temperature or composition of the film, must especially be 

 prevented. 



The liquid ordinarily employed for such investigations is the 

 " liquide glycerique " of M. Plateau. In dry air some of the water of 

 which it is in part composed would evaporate, while in moist air, in 

 consequence of the hygroscopic properties of the glycerine, additional 

 water would be absorbed. Though these facts were well known, and 

 though they are evidently possible sources of error, no attempt (as 

 far as I am aware) had been made before our own to determine 

 what precautions it was necessary to take to prevent the results ol 

 experiments such as M. Plateau's being affected by them. The first 

 question then that we set ourselves to answer, was — to what extent is 

 the composition of a soap film altered by changes in the temperature 

 or hygroscopic state of the air which surrounds it ? 



The method adopted in answering this inquiry was to measure the 

 electrical resistance of soaj) films formed in an inclosed space con- 

 taining a thermometer and hair hygrometer. If the observations led 

 to the conclusion that the resistance of film varied inversely as its 

 thickness, they would jirove that no change in composition had taken 

 place, and that the film at the thinnest had afforded no evidence of an 

 approach to a thickness equal to that of the surface-layers. If the 

 specific resistance was found to vary according to some regular law as 

 the thickness altered, there would be a strong presumption, that the 

 thickness was not much greater than, and was i)0hsibly even less than 



