PHYSICAL BASIS OF SOLAR CHEMISTRY. 339 



clocks leaning against the same rail; one of them, 

 which we may call A, was set going; the other, B, not. 

 Some time afterwards he found, to his surprise, that B 

 was ticking also. The pendulums being of the same 

 length, the shocks imparted by the ticking of A to the 

 rail against which both clocks rested were propagated 

 to B, and were so timed as to set B going. Other curi- 

 ous effects were at the same time observed. When 

 the pendulums differed from each other a certain 

 amount, A set B going, but the reaction of B stopped 



A. Then B set A going, and the reaction of A stopped 



B. When the periods of oscillation were close to each 

 other, but still not quite alike, the clocks mutually con- 

 trolled each other, and by a kind of compromise they 

 ticked in perfect unison. 



But what has all this to do with our present sub- 

 ject? The varied actions of the universe are all modes 

 of motion; and the vibration of a ray claims strict 

 brotherhood with the vibrations of our pendulum. 

 Suppose ethereal waves striking upon atoms which os- 

 cillate in the same periods as the waves, the motion of 

 the waves will be absorbed by the atoms; suppose we 

 send our beam of white light through a sodium flame, 

 the atoms of that flame will be chiefly affected by those 

 undulations which are synchronous with their own 

 periods of vibration. There will be on the part of 

 those particular rays a transference of motion from the 

 agitated ether to the atoms of the volatilised metal, 

 which, as already denned, is absorption. 



The experiment justifying this conclusion is now 

 for the first time to be made before a public audience. 

 I pass a beam through our two prisms, and the spec- 

 trum spreads its colours upon the screen. Between the 

 lamp and the prism I interpose a snapdragon light. 

 Alcohol and water are here mixed with common salt, 



