566 The Bight Hon. Lord Bayleigh [March 29, 



stance, such as pitch. Under these conditions, which preclude 

 more than a moderate pressure, it seems probable that no pits are 

 formed by the breaking out of fragments, but that the material 

 is worn away (at first, of course, on the eminences) almost mole- 

 cularly. 



The progress of the operation is easily watched with a micro- 

 scope, provided, say, with a -J-inch object-glass. The first few 

 minutes suffice to effect a very visible change. Under the micro- 

 scope it is seen that little facets, parallel to the general plane of the 

 surface, have been formed on all the more prominent eminences.* The 

 facets, although at this stage but a very small fraction of the whole 

 area, are adequate to give a sensible specular reflection, even at per- 

 pendicular incidence. On one occasion five minutes' polishing of a 

 rather finely ground glass surface was enough to qualify it for the 

 formation of interference bands, when brought into juxtaposition 

 with another polished surface, the light being either white or from 

 a soda flame ; so that in this way an optical test can be applied 

 almost before the polishing has begun.f 



As the polishing proceeds, the facets are seen under the micro- 

 scope to increase both in number and in size, until they occupy 

 much the larger part of the area. Somewhat later the parts as yet 

 untouched by the polisher appear as pits, or spots, upon a surface 

 otherwise invisible. Fig. 1 represents a photograph of a surface 

 at this stage taken with the microscope. The completion of the 

 process consists in rubbing away the whole surface down to the level 

 of the deepest pits. The last part of the operation, while it occupies 

 a great deal of time, and entails further risk of losing the " truth " 

 of the surface, adds very little to the effective area, or to the intensity 

 of the light regularly reflected or refracted. 



Perhaps the most important fact taught by the microscope is that 

 the polish of individual parts of the surface does not improve during 

 the process. As soon as they can be observed at all, the facets appear 

 absolutely structureless. In its subsequent action the polishing tool, 

 bearing only upon the parts already polished, extends the boundary 

 of these parts, but does not enhance their quality. Of course, the 

 mere fact that no structure can be perceived does not of itself prove 

 that pittings may not be taking place of a character too fine to be 

 shown by a particular microscope or by any possible microscope. 

 But so much discontinuity, as compared with the grinding action, 

 has to be admitted in any case, that one is inevitably led to the con- 

 clusion that in all probability the operation is a molecular one, and 

 that no coherent fragments containing a large number of molecules 

 are broken out. If this be so, there would be much less difference 



* The interpretation is facilitated by a thin coating of aniline dye which 

 attaches itself mainly to the hollows. 



t With oblique incidence, as in Talbot's experiments (see Phil. Mag., xxviii. 

 p. 191, 1889), achromatic bands may be observed from a surface absolutely un- 

 polished, but tbis disposition would not be favourable for testing purposes. 



