OIL FILMS ON WATER AND MEECUEY DEVAUX. 269 



loses its bright aspect, becoming leaden and as if covered with an 

 exceeding!}^ fine foam. Microscopic examination assures us that the 

 oil has changed into a multitude of droplets of various sizes, 10 [x, 

 5 [;l, 1 [X, and less. Working in the sunlight, I have seen the foam ap- 

 pear before the film has been reduced to one-half its maximum 

 extension. 



This is a new and direct proof of what we have just learned, that 

 as soon as an oil film is so much reduced in surface that it is more 

 than one molecule thick, nearly all the excess of oil forms into 

 globules. 



(e) Variation of molecular distances : This extraordinary fact gives 

 a new and interesting insight into the field of molecular action. It 

 shows particularly that the forces which stretch out these films of 

 liquids are due almost wholly to a single layer of molecules and that 

 the surface layer. It is evident further that a film, if it is uniform, 

 must be greater than one and less than two molecules in thickness. 

 Now, everything indicates that a film is really uniform and homo- 

 geneous, since the least variation in its thickness gives rise to con- 

 siderable differences of tensions which tend to reestablish everywhere 

 a perfect homogeneity, and especially the equality of molecular dis- 

 tances. The difference between the states of least and greatest exten- 

 sion can be dependent then only on the distances between the mole- 

 cules ; if they are compact in the first case, they can not be so in the 

 second. At any rate, that is the interpretation given by M. Brillouin 

 in a discussion which followed my communication.^ The distance 

 apart of the molecules in such films will be inversely as the square 

 root of the surface. Accordingly, the square roots of the preceding 

 ratios give the relative molecular distances. This ratio ranges 

 between 1.1 and 1.2. 



It follows that as soon as the molecules of a mono-molecular oil 

 film are separated by from 1.1 to 1.2 their normal distances, they lose 

 all power of lowering the surface tension of water. Conversely, as 

 soon as the molecules are brought together, so that they are separated 

 by 1.1 to 1.2 of their normal distances, they cause an abrupt and con- 

 siderable fall in the surface tension of the water, making it practi- 

 cally the same as if it were a large body of oil. For beyond this 

 limit the oil gathers into globules. 



(/) Correction to the value of the normal molecular distance : The 

 measure of the molecular distance 1.10 pijx, given above, corresponds to 

 films at their greatest extension. The true distance in normal oil will 

 be somewhat smaller, say 1.10/1.1 to 1.10/1.2 or 1.10 to 0.92 [jl^l. This 

 corrected distance differs decidedly from the theoretical value, 1.13 ]x^, 

 deduced from the measures of Perrin. Some day we will examine 

 the cause of this difference. 



Meeting of the Societe de Physique, May 3, 1911 



