428 OPTICAL PROJECTION 



corresponding variation in the amount of current, and therefore of 

 the light. Under these circumstances it is necessary in practice to 

 keep constantly feeding the arc, and if the operator takes his hand 

 off the controlling mechanism for more than a second cr two, the 

 light will flicker or go out. 



If, however, two-thirds of the current is absorbed in a fixed 

 resistance, and only one- third in the arc, a slight variation of this 

 has only one-half the effect, and consequently the arc can be left 

 for longer periods without feeding. 



In practice a current of 80 to 100 volts is the best for lantern 

 work, and if the original current is supplied at 200 volts or so, it 

 should be transformed down to this, and then a suitable resistance 

 put in series. From a commercial point of view this is rank heresy, 

 as a resistance simply wastes the current it absorbs, but this waste 

 of current is quite immaterial for our purpose, and the light obtained 

 at this voltage is quieter, steadier, and easier to manage than at 

 any other. Such a transformer for a 10-ampere arc lamp costs 

 51. 5s. to 61. 6s., and one should always be obtained when 

 the alternating current is supplied at anything higher than about 

 120 volts. 



One or two expedients have been employed in order to keep the 

 arc in front of the carbons. Mr. Hepworth recommended the use of 

 two cored carbons with the core placed eccentrically towards the 

 front, but I have been able to trace very little benefit from this 

 method, and I believe these carbons are not now manufactured. A 

 very similar effect can, however, be obtained by filing a flat on the 

 front surface of the carbons, should anyone desire to try the experi- 

 ment. Dr. Drysdale places a pair of electro -magnets, one each side 

 of the arc, in series with it, and therefore magnetised by the same 

 alternating current. 



These magnets therefore reverse their polarity in synchronism 

 with the vibrations of the arc, and their inductive effect is to deflect 

 the arc constantly outwards towards the condenser. The same 

 effect can be obtained by means of a turn or two of copper wire, 

 which must be insulated with asbestos, placed vertically on one or 

 both sides of the arc, edgeways towards the condenser, and both 

 these methods are extremely effectual in steadying the light. 



An inexpensive method of * rectifying ' or converting the current 

 from alternating to continuous has long been sought for, but up to 

 the present is not forthcoming. 



