196 Sir William Thomson [Feb. 2, 



electric action we are enabled as it were to sound the depth of the 

 ocean of molecules attracted to the metallic surface by the vapour or 

 gas entering into combination with it. 



When we come to thicknesses of considerably less than a wave- 

 length we find solid metals becoming transparent. Through the 

 kindness of Prof. Dewar I am able to show you some exceedingly 

 thin films of measured thicknesses of platinum, gold, and silver, 

 placed on glass plates. The platinum is of 1*9 x 10"^ cm. thickness, 

 and is quite opaque ; but here is a gold film of about the same 

 thickness, which is transparent to the electric light, as you see, and 

 transmits the beautiful green colour which you see on the screen. 

 The thickness of this gold (1*9, or nearly 2) is just half the wave- 

 length of violet light in air. This transjiarent gold, transmitting 

 green light to the screen as you see, at the same time reflects yellow 

 light to the ceiling. Now I will show you the silver. It is thinner, 

 being only 1 • 5 x 10~^ of a centimetre thick, or |ths of the air-wave- 

 length of violet light. It is quite opaque to the electric light so far 

 as our eyes allow us to judge, and reflects all the light up to the 

 ceiling. It is not wonderful that it should be opaque ; we might 

 wonder if it were otherwise ; but there is an invisible ultra-violet 

 light of a small range of wave-lengths, including a zinc line of air 

 wave-length 3-4 x 10"^ which this silver film transmits. For that 

 particular light the silver film of 1 • 5 x 10~^ thickness is transparent. 

 The image which you now see on the screen is a magic lantern repre- 

 sentation of the self-photographed spectrum of light that actually 

 came through that silver. You see the zinc line very clear across 

 it near its middle. Here then we have gold and silver transparent. 

 The silver is opaque for all except that very definite light of wave- 

 lengths from about 3 • 07 to 3 • 32. 



The different refrangibility of different colours is a result of 

 observation of vital importance in the question of the size of atoms. 

 You now see on the screen before you a prismatic spectrum, a well- 

 known phenomenon produced by the differences of the refractions of 

 the different colours in traversing the prism. The explanation of it 

 in the undulatory theory of light has taxed the powers of mathe- 

 maticians to the utmost. Look first, however, to what is easy and 

 made clear by that diagram (Fig. 3) before you, and you will easily 

 understand that refraction depends on difference of velocity of propa- 

 gation of light in the two transparent mediums concerned. The 

 angles in the diagram are approximately correct, for refraction at an 

 interface between air or vacuum and flint glass ; and you see that in 

 this case the velocity of propagation is less in the denser medium. 

 The more refractive medium (not always the denser) of the two has 

 the less velocity for light transmitted through it. The " refractive 

 index " of any transparent medium is the ratio of the velocity of pro- 

 pagation in the ether to the velocity of j)ropagation in the transparent 

 substance. 



Now that the velocity of the propagation of light should be dif- 



