266 ANNUM. REPORT SMITHSONIAN INSTITUTION, 1913. 



We know therefore that a jfihn of oil at its maximum extension is 

 formed of only a single layer of molecules.^ This remarkable fact 

 is true of other films than those of oil, nor is it limited to liquid 

 films. I have found it to be equally true for various solid substances, 

 with this difference: it is the solid state itself which disappears at 

 the critical thickness and not the surface tension, as with liquids.^ 

 I therefore derive this general conclusion : the characteristic mechan- 

 ical properties corresponding to certain states of a body, the sur- 

 face tension of a liquid or the rigidity of a solid, persist almost 

 intact down to molecular thicknesses, disappearing abruptly the 

 minute we go further. 



This fact has a general significance Avhich we should appreciate. 

 For the present, however^ we will be content in seeing a new and 

 direct demonstration of the discontinuity of matter and the realit}^ 

 of molecules; it is indeed a new method alloM'ing us to measure the 

 dimensions of molecules with a precision comparable with that of the 

 best methods we have, 



III. THE EVOLUTION OF LARGE DROPS OF OIL UPON WATER. 



Instead of placing upon the water a very minute trace of oil, let 

 us put there an ordinary drop of one to three hundredths of a cubic 

 centimeter. We will now watch a series of phenomena as interesting 

 as what we have just seen. Scarcely does the drop touch the water 

 when it spreads out and covers the whole surface. But the film, of 

 course, is very thick. It is hundreds of molecules thick and clearly 

 visible, because it reflects light better than does water. Generally 

 we see interference colors, at least at one phase of its extension. 

 But this phase is always fugitive, especially with nondrying and 

 fresh oils and when the surface of the water is very clean. This is 

 the case with the present film.^ The evolution of a film lasts but 

 10 to 15 seconds; indeed the principal phases take place in the first 

 3 seconds. However, on water already oily, the formation is very 

 much retarded and the film appears with a sharp circular border, as 

 in plate 7. Soon its brilliant surface is pierced with black, circular 

 spots looking like holes, where the water appears as if free from oil. 

 These spots, more or less numerous according to the kind of oil, 

 gradually grow in size, and each one is finally surrounded by a band 

 of small droplets similar to pearls (pi. 3). 



The first of these spots appears near the edge of the fihn, where it 

 is thinner than at the center. They grow very rapidly and soon run 

 together. The spots over the rest of the film subsequenth^ behave in 



1 Lord Rayleigh in the research cited above discussed this question, but the knowledge 

 then of the value of Avogadro's constant was not so accurate. 



= Devaux, 1. c, 1904. 



•""The process of fixation of these films is peculiar and has l)een described in a special 

 communication (1. c, Oct., 1912). 



