VARIATIONS OF SHPXUFIC GRAVITY 



2()5 



been first hammered to a somewhat more compact form. The maxi- 

 mum difference between-samples has diminished to 0.0097. 



Of greatest interest is Table III, which exhibits the variations 

 of density between cokl drawn and annealed platinum wires. 



Table III. — Platinitni irire of 1, 0.7, and O.J/. mm. diameter cold draicii {HI and 

 V. VII) and afterwards heated to incandescence {IV, VI, and VIII). 



Such are the results, which confirm the eai'lier discovery that 

 excessive compression diminishes the sjjccific gravity. I have already 

 drawn attention to the analogy here presented to the behavior of 

 gases under high pressures, which are less compressible than Boyle's 

 law requires. 



The explanation of this curious behavior of wires may, however, 

 dej)end on such factors as crystalline configuration, and, at all events, 

 requires further experiments to determine. 



A second new and very important result ajipears. that strongly 

 compressed wires increase in s})ecific gravity when heated to incan- 

 descence. This indicates that an increase of the kinetic energy of 

 the molecules suffices to restore them to their normal relations, but I 

 do not venture here upon any exi)lanation of the })henomenon and 

 content myself with pointing out this new and independent proof of 

 the variability of specific gravity." 



1 Here follows in the original articlo nn p.xplanation ol* tlio antlior's relations 

 with Hr. W. Spring in connection with tlic iniltlioation of tlioir joint researches 

 bearing on this subject. 



