32 Prof. J. H. Poynting on 



now took ten seconds less for its ascent. After a day's repose 

 it rose again in the oil with its original velocity. 



13. The fundamental formulas given in this paper admit of 

 easy extension to other cases. I have, for example, found in 

 the ascent of a hollow glass bulb a good indication of the 

 maximum density of water, and of the amount of dissolved 

 matter in saline solutions; and it could obviously be applied to 

 determine the specific gravity of solid bodies. 



The experiments which I have recorded show clearly that 

 the rate of ascent of a solid through a liquid depends, not only 

 upon the density of the liquid, but upon some other property 

 — probably the viscosity. When the ascending solid approxi- 

 mates in diameter to the diameter of the reservoir, special retar- 

 dation occurs, and some new law controls the motion. The com- 

 mencement of this retardation is earlier with oils than with 

 water, earlier with water than with alcohol and water — earlier, 

 in short, when viscosity is greater. In reservoirs of sufficient 

 diameter, the rate is inversely proportional to the square of 

 the altered condition. 



My thanks are due to Messrs. Ellis and Smith for their aid 

 in this investigation. 



Glasgow, March 1881. 



V. Change of State : Solid-Liquid. By J. H. Poynting, 

 Late Fellow of Trinity College, Cambridge, Professor of 

 Physics, Mason College, Birmingham* '. 

 [Plate I.] 



TWO distinct types of change of state from solid to liquid 

 have usually been recognized. The most familiar of 

 these is the ice-water type, in which, as the temperature rises, 

 the solid remains quite solid up to the melting-point ; when 

 this is reached it begins to melt at the surface, and the tem- 

 perature remains constant till the whole is liquid, when the 

 temperature again rises. Corresponding to this change of 

 state there is a definite latent heat. In the second class of 

 bodies, of which sealing-wax and phosphorus are examples, 

 there is a gradual softening as the temperature rises ; and this 

 softening takes place throughout the mass. There is no definite 

 arrest of the rise of temperature, and no definite latent heat. 



It has sometimes been supposed that the ice- water type is 

 merely a limiting case of the sealing-wax type, where the 

 softening takes place, but through a very small range of tem- 

 perature. Prof. Forbes held this view, and by it attempted 

 to explain regelation ; but subsequent experiments have not 

 * Communicated by the Physical Society. 



