446 Prof. G. Magnus on the Change in the Radiation 



tion depended merely on a change in the density of the superficial 

 layer. In support of this view he adduces the fact that in cer- 

 tain substances, such as glass, marble, and agate, he observed no 

 change in the radiation whether the surface were rough or polished, 

 and only in the case of metals was there an increase when they 

 were roughened. As this was observed not only in the case of 

 easily oxidizable metals, but in those of gold and platinum, so 

 that the notion that the change might have been due to a layer of 

 oxide was excluded, Melloni ascribes the phenomenon to metals 

 being compressible, but agate, ivory, or marble not. Experiment 

 teaches, he says, that metal plates which are made by hammering 

 or rolling have a greater density on the surface than in the in- 

 terior. Looking at a list of the radiating-powers of bodies, it is 

 at once seen that in general this property is inversely as their 

 respective densities. Hence, if it be assumed that the same law 

 also holds for different densities of one and the same substance, 

 the greater radiation is explained ; for, by scratching, the surface 

 becomes less dense, or the interior softer places are laid bare. 



In support of this view Melloni adduces the following experi- 

 ment. He had four plates of very pure silver made — two 

 strongly hammered, and two cast and allowed to cool very slowly in 

 their moulds of sand. These he used as the sides of a quadran- 

 gular box with a metal bottom ; and in order not to alter the 

 density and hardness of the plates, he soldered them with easily 

 fusible solder. Before being joined, they were polished with pu- 

 mice-stone and charcoal without using hammer or burnisher. One 

 of the cast and one of the hammered plates were then strongly 

 rubbed in one direction with coarse emery-paper. The sides which 

 had retained their lustre reflected sharp images, the rubbed sides 

 dull and streaked ones. The silver vessel thus prepared was filled 

 with hot water. The deflections which the four sides produced 

 were : — 



o 



The hammered and polished side . . 10 



The hammered and scratched side . . 18 



The cast and polished side .... 1S'7 



The cast and scratched side . . . . 11*3 



Scratching the hammered plate had thus produced an increase 

 of the radiation, while in the cast plate it had produced a dimi- 

 nution. This unexpected fact appeared to him to prove the cor- 

 rectness of his principle. 



We are as much, and perhaps more justified in saying that the 

 surface becomes denser by scratching, than in saying, with Mel- 

 loni, that it becomes less dense; for during the scratching a 

 pressure is exerted upon the surface ; and if even it be assumed 

 that the individual scratches are not pressed in but are scooped 



