250 FRAGMENTS OF SCIENCE. 



point. Those waves are absorbed whose vibrations syn 

 chronize with those of the molecules or atoms on which 

 they impinge ; a principle which is sometimes expressed by 

 saying that bodies radiate and absorb the same rays. This 

 great law, as you know, is the foundation of spectrum- 

 analysis ; it enabled Kirchhoff to explain the lines of Frauen- 

 hofer, and to determine the chemical composition of the 

 atmosphere of the sun. If, then, after such a change as 

 that involved in the passage of a vapor to the liquid state, 

 the same waves are absorbed as were absorbed prior to the 

 passage, it is a proof that the molecules, which must have 

 utterly changed their periods, cannot be the seat of the ab 

 sorption ; and we are driven to conclude that it is to the 

 atoms j whose rates of vibration are unchanged by the 

 change of aggregation, that the wave-motion is transferred. 

 If experiment should prove this identity of action on the 

 part of a vapor and its liquid, it would establish in a new 

 and striking manner the conclusion to which we have pre 

 viously leaned. 



We w r ill now resort to the experimental test. In front 

 of this experimental tube, w r hich contains a quantity of the 

 nitrite-of-amyl vapor, is placed a glass cell a quarter of an 

 inch in thickness, filled with the liquid nitrite of amyl. I 

 send the electric beam first through the liquid and then 

 through its vapor. The luminous power of this beam is 

 very great, but it can make no impression upon the vapor. 

 The liquid has robbed it completely of its effective waves. 

 When the liquid is removed chemical action immediately 

 commences, and in a moment we have the apparently 

 empty tube filled with this bright cloud, precipitated by 

 one portion of the beam, and illuminated by another. Thus 

 we uncover to some extent the secrets of this world of 

 molecules and atoms. 



Instead of employing air as the vehicle by which the 

 vapor is carried into the experimental tube, we may em- 



