270 Prof. G. Quincke on the Capillary 



constants forliquids like water (0*008 gr.) or mercury (a = 0*055 gi\). 

 Yet, as Laplace has shown, these constants must increase with 

 the square of the density if the function of attraction remains the 

 same ; and it is even possible that they experience a sudden in- 

 crease near the change of condition, just as the density or the 

 electrical conductivity suddenly changes. Experiment has also 

 shown that, as the temperature of water and other liquids sinks, 

 the constant increases. The author has observed the same fact 

 in the case of mercury, in opposition to the statement of M. 

 Frankenheim*. 



Near their melting-points the capillarity-constants have com- 

 parable values, as will presently be shown. Hence metals with 

 a high melting-point, apart from other differences, must indicate 

 a high value for ordinary temperatures, as is seen from the 

 above Table. 



As a layer of liquid near the surface acts like a stretched mem- 

 brane, the surface must offer to external impressions a resistance 

 which is greater the greater the capillarity-constant. Hence the 

 metals must be ranged according to the value of a in the same 

 order as that of their hardness. Karmarscb/s experiments f ? 

 in fact, as well as those of Calvert and Johnson J, on the hard- 

 ness of metals, give values in accordance with this relation which 

 agree as well as can be expected in such determinations. 



It might be thought that wires of the same section with a 

 larger surface must exhibit greater tenacity. The author has 

 made experiments in this direction with round and flattened 

 silver and copper wires, and he has found that the tenacity with 

 a flattened wire is almost the same as in the case of a round wire 

 of the same section and the same material. It must at the same 

 time be remembered that in flattening, not merely the density 

 but the continuity of the surface is altered, as microscopic inspec- 

 tion shows, and the wire acquires cracks like flattened dough. 

 Moreover the surface of such a flattened wire is not a surface of 

 equilibrium, and other forces (the difficult displaceability of the 

 particles) prevent the formation of the surface of equilibrium 

 towards which the capillary forces tend. In the different parts 

 of the surface there will be different tension. As, moreover, the 

 section of the flattened wire is very irregular, a more accurate 

 calculation would only afford interest in case we could make 

 experiments between drawn wires having an elliptical and those 

 having a circular section. 



Gold and silver leaf have not a continuous surface, but are per- 



* Pogg. Ann. vol. lxxv. p. 26. 



t Mit. des gew. Vereinsfilr Hannover, 1858, p. 1/8. 

 % Ibid. p. 1/5, taken from the Memoirs of the Literary and Philoso- 

 phical Society of Manchester. 



