480 EVENING DISCOURSES 



This reference to the sun's light is by the way, but I mention it because its 

 colour is on the whole what we like best for illuminating the people and 

 things we ordinarily look at. The incandescent filament being a rough 

 imitation of the sun's light also on the whole gives us satisfaction for 

 illuminating purposes. 



You therefore ask, why do we want to break away from incandescent 

 solids, and employ these vapours and gases which seem to give such highly 

 coloured light ? 



The reason is easy to see if we look at this diagram. When we make 

 light by causing electricity to stimulate mercury vapour the upper part of 

 the diagram shows what we are getting from the mercury for the money 

 we spend on the electricity : a stream of green light ; somewhat smaller 

 streams of blue and yellow light ; a trickle of violet light. There are also 

 two other radiation streams which are not giving light, but which are 

 nevertheless using up some of the electricity. One small stream is beyond 

 the violet light. It is called ultra-violet, and we shall see in a moment how 

 it can be used. The other stream is beyond the visible light at the other 

 end and is called infra-red. This latter is heat without light ; very little 

 of it is wanted. The electricity which is represented by this heat radiation 

 is mostly wasted. But you see what a lot of useful light we get for the 

 electricity we pay for in proportion to the amount which is wasted as heat. 



Now look at the lower diagram representing an incandescent solid — for 

 instance, a filament. This shows what the filament gives us in return for 

 the money we spend on the electricity which makes it white hot. The 

 shaded part is the light we get. Look at the great torrent of energy wasted 

 as heat and represented by the black area. 



Some of these vapours and gases are able to give us much more light in 

 comparison with the wasted heat than the incandescent filament can ever 

 hope to do. We therefore get from them more light for the money we spend 

 on the electricity. 



That then is the lure of these electric discharge lamps. We like plenty 

 of light and we like it cheap. 



Before we examine some of the very fascinating things connected with 

 new electric discharge lamps, I want to spend five minutes thinking about 

 filament lamps. They, too, have responded to research and skill expended 

 with the object of making them more efficient. 



To make a filament lamp more efficient you can try to do either or both 

 of two things. You can make the filament give out more light by running 

 it hotter or, keeping the light the same, you can make it waste less energy 

 by preventing it from losing so much of its heat. 



Before the war, when we had this ' cage ' type lamp, with its straight 

 tungsten filament in a vacuum, we thought we had reached the end of lamp 

 evolution. For tungsten, which is a most difficult material to work, has 

 almost the highest melting point of any metal. It can therefore be run 

 very hot, and no one could see how anything hotter, and at the same time 

 more efficient, was possible. True, if the electricity were forced through 

 this filament so as to bring it closer to its melting point, it became hotter, 

 but the tungsten evaporated from the filament and the life was short. 



Why not then put gas in the bulb instead of having a vacuum, and thereby 

 prevent the molecules of the tungsten being evaporated off ? 



The lecturer proceeded to show by experiment that the introduction of 

 nitrogen into a lamp bulb containing a straight tungsten filament, diminished 

 the light emitted from the filament by abstracting heat from it ; further, 

 that forcing more current through the filament in an attempt to increase the 

 light output made the whole bulb so hot that it threatened to collapse. 



