Constants of Solid Bodies. 273 



surfaces are not surfaces of equilibrium, as in the case of liquids, 

 that therefore, as in these too, we are ignorant of the nature of 

 the molecular forces and the distance at which they act. 



For this reason the author has directed his attention for the 

 present to actual liquids, and has endeavoured to determine the 

 capillarity-constants of fused bodies, especially metals, of which 

 hitherto little was known. His expectation, that the capillarity- 

 constants of metals near their solidifying- point must have com- 

 parable values, he has found confirmed. From the nature of the 

 case these determinations must be considered not more than 

 approximations to the true value, and indicate greater discrepan- 

 cies than is the case with the determinations of liquids at ordi- 

 nary temperature, like water, mercury, alcohol, &c. 



The noble metals, in the form of vertical wires, were melted in 

 a flame at as low a temperature as possible, until the suspended 

 drop fell. Neglecting the small quantity of fused metal adhering 

 to the thin wire, the weight P of the drop is the greatest weight 

 which the fused vertical mass can support, or 



V = a2r7r; (3) 



where 2r is the diameter of the wire in millimetres, a the capil- 

 larity of the fused metal. 



In the same manner were heated glass threads which had been 

 drawn out of the same glass rod at the blowpipe-table. 



The flame used was generally a small coal-gas one, of 10 mil- 

 lims. height and 3 millims. diameter. Only in the case of pla- 

 tinum was a current of oxygen passed into this flame from an 

 ordinary blowpipe. Special experiments with gold wires showed 

 that flames at high temperature give drops which are compara- 

 tively smaller only because the highest layer of the surface of the 

 drop touches solid metal, and must therefore be near the fusing- 

 point of the metal. In thick wires the influence of the high 

 temperature of the flame must have least play, and hence these 

 give in general the highest values of the constants a. In this 

 arrangement of the experiment the author thought the highest 

 values of « to be the most trustworthy. 



The following are measurements on a series of platinum wires 

 in confirmation of the relation expressed by equation (3) : — 



iameter of the 



Weight of the 



Capillarity- 



wire. 



drop. 



constant. 



2r. 



P. 



cc. 



millim. 



gr. 



mgr. 



0-5675 



0-2912 



163 



0-3689 



0-2055 



177-4 



0-1921 



00996 



165-1 



0-0993 



00530 



169*8 



0-0767 



0-0410 



169-9 



169-04 

 Phil. Mag. S. 4. Vol. 36. No. 243. Oct. 1868. T 



