THE FIRST AND THE LAST CATASTROPHE. 273 



to red light. The waves of red light are not made to travel so slowly 

 as the waves of blue light, but, as in the case of waves travelino- over 

 the sea, when light moves in the interior of a transparent body the 

 large waves travel quickest. Well, then, by using such a body as 

 will separate out the different colors a prism we are able to affirm 

 what are the constituents of the light which strikes upon it. The 

 light that comes from the sun is made up of waves of various lengths 

 but making it pass through a prism we can separate it out into a 

 spectrum, and in that way we find a band of light instead of a spot 

 coming from the sun, and to every band in the spectrum corresponds 

 a wave of a certain definite length and definite time in vibration. 

 Now we come to a very singular phenomenon. If you take a gas 

 such as chlorine and interpose it in the wave of that light, you w^ill 

 find that certain particular rays of the spectrum are absorbed, while 

 others are not. Now, how is it that certain particular rates of vibra- 

 tion can be absorbed by this chlorine gas while others are not ? That 

 happens in this way, that the chlorine gas consists of a great number 

 of very small structures, each of which is capable of vibrating inter- 

 nally. Each of these structures is complicated, and is capable of a 

 change of relative position among its parts of a vibratory character. 

 We know that molecules are capable of such vibrations, such internal 

 vibrations, for this reason, that if we heat any solid body sufficiently 

 it will in time give out light; that is to say, the molecules are got 

 into such a state of vibration that they start the ether vibrating, and 

 they start the ether vibrating at the same rate at which they vibrate 

 themselves. So that what we learn from the absorption of certain 

 particular rays of light by chlorine gas, is that the molecules of that 

 gas are structnres which have certain natural rates of vibration, pre- 

 cisely those rates of vibration which belong to the molecules natu- 

 rally. If you sing a certain note to a string of a piano, that string 

 if in tune will vibrate. If, therefore, a screen of such strings were 

 put across a room, and you sang a note on one side, a person on the 

 other side would hear the note very weakly or not at all, because it 

 would be absorbed by the strings ; but if you sang another note, not 

 one to which the strings naturally vibrated, then it would pass through, 

 and would not be eaten up by setting the strings vibrating. Now this 

 question arises. Let us put the molecules aside for a moment. Sup- 

 pose we do not know of their existence, and say, " Is this rate of vi- 

 bration, which naturally belongs to the gas, a thing which belongs to 

 it as a whole, or does it belong to separate parts of it ? " You might 

 suppose that it belongs to the gas as a whole. A jar of water, if you 

 shake it, has a perfectly definite time in which it oscillates, and that 

 is very easily measured. That time of oscillation belongs to the jar 

 of water as a whole. It depends upon the weight of the water, and 

 the shape of the jar. But now, by a very certain method, we know 

 that the time of vibration which corresponds to a certain definite gas 

 VOL. vn. 18 



