168 ILLUMINATION IN MICRO-METALLOGRAPHY: 



but it requires cumbersome accessories and so much attention that it 

 need hardly be considered unless an electric supply is quite out of 

 reach. 



The mercury vapour lamp ought to be mentioned, but the writer 

 has no experience of it, can find no information as to its intrinsic 

 brilliancy, and has seen no form of the lamp which is convenient for 

 metallography. The nearly monochromatic quality of the light is in its 

 favour, and it is probably capable of being put into a very useful and 

 efficient shape for the purpose. 



The arc crater gives the highest light intensity. With the 

 requisite attention it is unsurpassable in rapidity and quality of nega- 

 tives. But it has some disadvantages. The positive crater is somewhat 

 obscured by the tip of the negative carbon, the crater surf ace is not always 

 of uniform brilliancy all over, and the crater may shift during an exposure 

 from one part of the carbon to another. Such unsteadiness may arise 

 from the arc length being too great from an endeavour to get the 

 negative tip out of the field of view; from impurities in the carbons, 

 or from draughts. The arc length has to be adjusted at intervals of 

 a few minutes. The arc does not steady down until it has been burning 

 for several minutes. It is therefore an item of the equipment to be 

 attended to and Waited on. The arc gives off a large amount of 

 radiant heat which has to be considered in relation to any auxiliary 

 lens system required. It therefore leaves something to be desired in 

 point of convenience and its rapidity in photographing is subject to 

 some discount for the time taken in attending to it. 



The Nernst lamp is excellent in many ways. It requires no globe, 

 no attention, and is quite steady. It gives off relatively little radiant 

 heat. It can be used on either continuous or alternating current. The 

 shape of the radiant surface, a rod of quite small diameter is some- 

 what inconvenient, as it calls for very accurate centering of any auxi- 

 liary lens system. It has been unobtainable in this country for some 

 years, as it is made only in Germany. The small diameter of the rod 

 obliges one to magnify it considerably by the auxiliary lens system 

 used, so that the effective brilliancy is not so great as might appear. 

 With a one-ampere Nernst lamp I have done a good deal of work on 

 steel at 700 to 1,000 diameters, and find the exposure required to be 

 from five to ten minutes at such powers, using fast plates and a light 

 filter. 



The Pointolite or tungsten arc is free from the inconveniences of 

 the carbon arc. The source of the light is a small ball of tungsten 

 which appears in the field of view as a disc of uniform brilliancy fixed 

 in position. Its intrinsic brilliancy from some rough tests of my own 

 seems to be about one-third that of the carbon arc crater. It lights 

 up at once, is normal in a few seconds, requires no adjustment or 

 attention, and gives off a relatively small amourit of radiant heat. It 

 is very promising and the larger sizes which are being developed may 

 prove to be as quick in work as the carbon arc, when the absence of 

 attention and unsteadiness are taken into account. Like the carbon 

 arc, it requires continuous current for its operation. 



Half-watt lamps with straight filaments as made for motor-car 

 head lights are quite useful. Their intrinsic brilliancy is little inferior 

 to that of the tungsten arc. The small diameter of the coiled filament 

 is open to the same objection as the Nernst filament, i.e., it has to be- 



