352 Spontaneously Reversible Spectral Rays. [June, 
which is easily obtained on burning the metal in the air. 
It may be added that the so-called long rays, defined some- 
what arbitrarily by certain spedtroscopists, often enter the 
class of the spontaneously reversible rays. 
It is scarcely needful to point out that the ordinary laws 
of luminous absorption and of refrigeration suffice to explain 
all these phenomena. But if theory is able to predict the 
possibility of the reversal of a certain ray it tells us nothing 
concerning the property of certain radiations to present, 
under given conditions, spedtral reversions to the exclusion 
of others. 
What are, generally speaking, the conditions of tempera- 
ture, of pressure, and of density the most favourable to the 
production of this singular emissive power ? This, for the 
present, is difficult to say. We obtain the phenomena by the 
incandescence of metallic vapours in a space more or less 
limited, heated by the eledtric arc or by diredt combustion. 
But it is evident that these conditions have to be deter- 
mined with more precision, and that they are imperfedtly 
fulfilled by the means used in the production of emission 
spedtra. 
(To be continued). 
