M. Kronig on the Theory of the Davy Lamp. 227 



When hydrogen was passed through a neutral solution of 

 bichloride of platinum it soon became turbid, and partly a black 

 pulverulent and partly a metallic lustrous precipitate was formed. 

 At the same time the solution became clearer, and was finally 

 free from platinum. On this deportment Brunner bases a method 

 for separating platinum from its solutions, the details of which he 

 describes ; and he indicates the possibility of a technical appli- 

 cation of this method for separating platinum from its ores. 



Palladium is as readily reduced from its solutions as platinum ; 

 while* the reduction of chloride of iridium is quite insignificant, 

 and chloride of gold is unchanged. The same is the case with 

 solutions of chloride of mercury if the hydrogen is under the 

 ordinary pressure. But Brunner found, as Beketoff had also 

 done, that under a pressure of 100 atmospheres this substance 

 is reduced and mercury deposited in distinct globules. 



Davy referred the action of his safety-lamp to the cooling 

 action of the wire gauze. Dissatisfied with the inadequacy of 

 this, Kronig proposes * the following : — 



" Although experiment shows that a wire gauze can cool the 

 gaseous products of combustion present in a flame to a point 

 below the temperature at which they ignite, the question arises, 

 on what does this action depend. Several things are possible. 

 A cold wire gauze introduced into the flame can take away heat. 

 But the cooling thus produced is less the higher the tempera- 

 ture of the gauze rises ; and a continuous cooling of the flame 

 by the wire gauze is only possible when the wire gauze loses on 

 the outside the heat it receives from the flame. Such a loss can 

 occur either by conduction or by radiation. If the flame is small, 

 heat may be conducted from the middle parts of the heated wire 

 gauze; but this conduction must be less the greater the flame. 

 Hence it is probable that the wire gauze loses heat more by 

 radiation than by conduction. 



" The assumption that metal gauze radiates more heat than the 

 gaseous flame is a matter of course, for we know that ignited 

 solid bodies radiate more light than gaseous bodies at the same 

 temperature." 



This opinion, says Kronig, has become a certainty since the 

 publication of Magnus's interesting experiments in his paper 

 " On the Constitution of the Sun"-}\ For not ° ni y does he show 

 that the introduction of a disk of platinum into a non-luminous 

 gas-flame causes it to radiate more heat, but also that this radia- 

 tion experiences a further increase when the platinum is soaked in 

 carbonate of soda. This observation appears completely to ex- 



* Poggendorff's Annalen, May 1864. 

 t Phil. Mag. S. 4. vol. xxvii.p.376. 



