LUMINESCENCE AND INCANDESCENCE 33 



rubbing or shaking the animals, it is in reality fundamen- 

 tally different since it is dependent on the presence of 

 oxygen whereas triboluminescence is not. 



Crystalloluminescence. — Crystalloluminescence is 

 observed when solutions crystallize. It was described by 

 Bandrowski (1894, 1895) in arsenious oxide, in NaP, or if 

 IICl or alcohol is added to hot saturated NaCl solution. 

 A bluish light with sparkling points appeared. All well 

 authenticated cases are exhibited by simple inorganic 

 salts and these are also all triboluminescent. The re- 

 verse is not true, however; many triboluminescent sub- 

 stances are not crvstalloluminescent. Crvstallolumines- 

 cence is much less widespread than triboluminescence. 

 Trautz (1905) has studied the matter in a number of com- 

 pounds and comes to the conclusion that the light is really 

 a special case of triboluminescence in which the growth of 

 individual crystals causes them to rub together. The 

 light becomes much brighter on stirring a mass of crystals 

 which exhibit crystalloluminescence. While in some cases 

 crystalloluminescence is unquestionably due to the tribolu- 

 minescence of crystals rubbing against each other it is not 

 in every case, as has been clearly shown by the work of 

 Weiser (1918 h). He studied luminescence of saturated 

 aqueous alkali halide solutions (NaCl, KCl, etc.,) upon 

 addition of alcohol or of HCl. The salt crystallizes out 

 under these conditions and Weiser found that the light is 

 brightest when the conditions of concentration of alcohol 

 or of HCl are such as to cause heaping up of Na and CI 

 ions. He believes that the bluish light which appears is 

 due to the combination of ions in the reaction, Na"" + 

 Cl~= NaCl. Only if this proceeds rapidly enough does 

 luminescence occur. Weiser studied also the crystallo- 



