SCIENCE AND ELECTRIC LIGHTING 483 



Certain other additives — hydrogen for example — change the efficiency 

 without much alteration of the colour. We have here a lamp with neon gas, 

 fitted with a palladium tube for the purpose of admitting hydrogen. You 

 will see when this is done that the light returns to normal as soon as the 

 supply of hydrogen is stopped. Gases are not good mixers. 



Of all the gases and vapours you have seen three stand out as the most 

 promising ones to use if we want light to be abundant and cheap. 



Cheap you say — yes, and what about the colour rendering of the human 

 face ? We have to go a step at a time. We have aimed first at high 

 efficiency, and if this could be attained we have been confident that we could 

 in time improve the colour. We will try to see how science and research 

 have brought us, so far, in the direction of combining high efficiency with 

 tolerable colour. I am going first to floodlight you successively with the 

 light from sodium, neon, and mercury. Don't look at the lamps themselves 

 but at your surroundings. 



The sodium lamp is pretty bad and we don't yet know a good way of 

 correcting its colour — which is a pity, for at present it is the most efficient 

 of all. It gives out, as we saw, just yellow light, and all objects, including 

 our faces, are a monotone in yellow-brown. When it is used at the relatively 

 low illuminations which prevail in street lighting it is tolerable, and is 

 being employed for this purpose, for which it has certain advantages. 



Here is a colour chart which we will illuminate first by a tungsten lamp 

 and then with a sodium lamp. It is obvious how such a lamp, used for 

 general illumination, would take all the interest and colour out of our 

 surroundings. 



The neon lamp has been used for floodlighting red-brick buildings, for 

 which it is particularly suitable, and for obtaining other colour effects. 

 Its luminous efficiency is about the same as that of the familiar gas-filled 

 tungsten filament lamp. The light is, however, almost completely deficient 

 in green and blue rays and considered as a lamp for general illumination 

 purposes, neon has little to commend it. It has, however, been used 

 together with special green tubes for efficient interior lighting— -lighting, 

 that is to say, which must give reasonably faithful colour rendering — and 

 which, by the way, must be just sufficiently unfaithful to the complexion 

 to be flattering. 



Owing to the high efficiency of the green tube used in these combined 

 lighting units, the overall efficiency of the combination is approximately 

 twice that of the tungsten filament lamps. But I would like to show you 

 the beginning of something better — the result of some recent investigations. 

 We find that if we coat the inside of a neon tube with a special luminescent 

 powder, the activity of the gas in the tube excites the characteristic 

 fluorescence of this powder. It therefore gives out light, light of the 

 colour we want which, as it mixes with the red light from the neon gas, gives 

 us a series of agreeable colours very suitable for interior lighting. 



Here are three such tubes. The left-hand ends have been left uncoated 

 so as to show what the light is like without the powders. In these coated 

 tubes the presence of the powder not only improves the colour but doubles 

 the light from them. The lowest of these gives a very fair white light. 



Hitherto it was only found possible to excite fluorescence to an appreciable 

 extent by direct excitation of the powder by the electrons of which the 

 current consists, or else by means of the mercury discharge, which is very 

 rich in ultra-violet lines. The great advance lies in the discovery that neon, 

 which is comparatively poor in ultra-violet light, can in certain circumstances 

 and with suitable powders be made to excite luminescence. 



I want to return to these luminescent materials shortly. But before 



