PHYSICS. 341 



Koenift- has calculated the coefficieut of iuterior friction for several 

 liquids at a meau temperature of about 18°, from comparative ex- 

 periments made by Maxwell's method of oscillating disks, and by the 

 method of flow through capillary tubes. Ether gave by the former 

 method 0.00274 C. G. S. units ; by the latter, 0.001156 ; carbon disulphide, 

 0.00451 and 0.00388; light benzine, 0.00G37 and 0.00G27 and 0.00523; 

 heavy benzine, 0.00862 and0.00GS8 ; distilled water, 0.01.587 and 0.01096 ; 

 and turpentine, 0.02836 and 0.01865. Hence the former method gives 

 somewhat higher results than the latter, and moreover the differences 

 increase with the coefficients. (Wied. Ann., xxv, 618: J. Phys., No- 

 vember 1886, II, V, 486.) 



Ayrton and Perry have communicated to the Physical Society of Lon- 

 <lon a paper upon the expansion produced by amalgamation. They 

 find for example that the amalgamation of brass is accompanied by a 

 great expansive force. If one edge of a straight thick brass bar be 

 amalgamated it will be found that in a short time the bar is curved, the 

 amalgamated edge being always convex and the opposite edge concave. 

 The authors suggest that a similar action may be the primary cause of 

 the phenomena presented by the Japanese " magic mirrors." •• Japanese 

 mirrors are made of bronze and have a pattern cast upon the back ; and 

 although to the eye no trace of it can be discovered upon the polished re- 

 flecting surface, yet when light is reflected from certain of these mir- 

 rors on to a screen the pattern is distinctly visible in the luminous 

 patch formed. This is due to the polished side opposite the thinner 

 I)arts of the casting being more convex than the others, a conclusion 

 verified by the fact that the pattern is reversed when formed by a con- 

 vergent beam of light. Such a condition of things would evidently re- 

 sult from a uniform expansive stress taking place over the reflecting 

 surface, the thinner — and consequently the weaker — parts becoming 

 more convex or less concave than the others. Hitherto this inequality of 

 curvature has been attributed to a mechanical distortion to which the 

 mirrors are intentionally submitted during manufacture to produce the 

 general convexity of the polished surface; but the authors now think 

 it possible that the use of a mercury amalgam in the process of polish- 

 ing may have an effect in the jiroil action of this inequality of curvature. 

 (Xature, April, 1886, xxxrii, 575; Phil. Mag., October, 1886, V. xxii, 

 327.) 



AVarburg and Ihmore have experimented to determine the cause of 

 the layer of water which forms on glass and other bodies. They find : 

 (1) That above the dew-point no weighable deposit of water could be 

 detected on bodies with smooth surfoces, insoluble in water, such as pla- 

 tinum, glass'with a coating of silica, glass free from alkali. Had the 

 thickness of such a layer exceeded one or two millionths of a milli- 

 meter, the balance was sensitive enough to have detected it. (2) That 

 the film of water which forms on alkaline glass above the dew-point 

 arises from a small quantity of free or loosely combined alkali on the 



