Sharpness of a liquid surface

This happens to be a test of such sensitiveness that it will detect layers of the order one molecule thick. Jamin and others had found that the light reflected from most solid and liquid surfaces usually deviates considerably from this law, but Rayleigh1 showed that when the accidental layer of grease was carefully cleaned away from a water surface practically no trace of ellipticity remains. It is known now that the layer of grease on water is only one molecule thick (see Chap. II) hence the practical absence of ellipticity indicates that a water surface is a definite transition, within the thickness of about one molecule. 

Raman and Ramdas2 found that there is still some small amount of residual ellipticity in their cleanest surfaces of water, and also that these scatter light to some extent. There appears to be a slight discrepancy between these results and Rayleigh s, but both agree that the transitional layer is about one molecule thick the slight residual ellipticity is ascribed to the thermal agitation of the water molecules at the surface. 

Further evidence is given by the very dense nature of the films of fatty substances which can be supported on the surface of water. These have practically the same density as the substances in bulk, and are very strongly attached to the surface of the water (Chap. II). This could scarcely occur unless the surface layers of water were of nearly the normal density. 

Edser3 has worked out a theory of the internal cohesion and surface tension of liquids, based on the assumption that the molecules may be regarded as occupying a space spherical in shape (not because the molecules themselves are spherical, but because they will probably sweep out a space more or less spherical during any considerable period of time), and concludes that about 94 per cent, of the free energy of the surface resides 

Theories of Capillarity have been proposed, in which a gradual transition between liquid and vapour is assumed. These cannot now be considered as resting on a basis of fact. 

---