May 27, 1922] 



NATURE 



683 



The Blue Flame produced by Common Salt 

 on a Coal Fire. 



; It is sometimes stated that the blue flame which is 



seen when common salt is thrown on to a coal lire is 

 due to traces of copper in the coal. It is much more 

 likely that this flame is that of carbon monoxide 

 produced by the cooling of the hot coal by the salt, 



; and certain observations lend support to this view, 



^ such as the following : 



\ (i) The blue flame is visible under proper condi- 



I tions without the addition of salt (and is commonly 

 held to portend frosty weather) . 



(2) The addition of salt to a fire consisting of white- 

 hot embers — that is, one from which most of the 

 carbon has been burnt — gives no blue flame, which it 

 should do if the flame is due to copper chloride. 



(3) The addition of other substances than sodium 

 chloride produces the same effect, a spent filter paper 

 for example. 



(4) The colour of the flame seen is apparently 

 identical with that of the carbon monoxide flame but 

 not with the green copper colour. 



(5) The sodium flame is never observed in this case 

 because the temperature is too low to volatiUse the 

 sodium chloride. For a similar reason it is unlikely 

 that the copper flame can be observed. 



Possibly I am wrong, but the matter is interesting, 

 and deserves to be made clear. W. Hughes. 



63 Goldington Avenue, Bedford. 



Mr. Hughes's letter raises some interesting 

 questions regarding the conditions under which the 

 " blue flame " of the coal fire appears. There is no 

 reason to suppose that under appropriate conditions 

 the flame of burning carbon monoxide cannot be seen 

 in a coal fire, but it would be difficult to identify 

 since the spectrum is, in the main, continuous. On 

 the other hand, the blue flame of copper chloride, 

 which is distinct from the green flame of the oxide, 

 has a very characteristic spectrum, and there is no 

 difficulty in its identification. There is no doubt, 

 however, that for the appearance of the spectrum of a 

 compound certain accessory conditions have to be 

 fulfilled, and in many cases their effect is by no means 

 obvious. Perhaps one of the most striking instances 

 of this is to be found in the appearance of the spectrum 

 associated with burning sulphur in the flame of an 

 ordinary bunsen burner when the gases of the flame 

 are cooled, e.g. by holding a thick plate of cold metal 

 in the flame. In this case the sulphur occurs as 

 an impurity in the coal gas but is not seen in the 

 spectrum of the burning gases unless they are cooled. 



T. R. Merton. 



Winforton House, Hereford. 



Pilot Lamps in Laboratories. 



Types of neon vacuum tubes recently placed on 

 the market as low candle-power glow lamps for 

 household electric lighting circuits, apart from 

 other uses, have several convenient applications in 

 laboratories as indicators to show when the supply 

 current is flowing in any given circuit. These lamps, 

 which are said by the makers to give only \ c.p., 

 have a very high resistance and small current con- 

 sumption : one type tested on a 200-volt circuit 

 took either 3 or 10 milliamperes, according to the 

 polarity of the connections, while another type took 

 12 and 30 milliamperes under the same conditions, 

 though individual lamps of the same type vary 

 considerably. In each case the lamp behaved well 



^ 



w\ 



with a resistance of more than 20,000 ohms in series, 

 and a current consumption of less than one milli- 

 ampere. 



In the case of electric furnaces, muffles, etc., one 

 lamp in parallel with the heating winding serves to 

 show when the current is " on," obviating the chance 

 of the apparatus being left under load when the 

 laboratory is closed at night, and effecting obvious 

 economies by indicating the consumption of 

 current at other times. With electrically heated 

 thermostats, incubators, constant 

 temperature ovens, etc., where the 

 means of heating are not directly 

 visible, a neon lamp serves usefully 

 to indicate contact when making 

 adjustments, while the use of a 

 second lamp shunted across the 

 contact breaker reduces arcing 

 and removes any doubt as to the 

 supply of current when the other 

 lamp is out. Fig. i shows a 

 simple method of placing the two lamps in such a 

 circuit. H is the heater winding, B the break, 

 L^ and L* the two neon lamps, while R is a 

 high resistance of 20,000 ohms or more, made by 

 drawing pencil lines on a piece of ebonite between 

 two terminals until a satisfactory glow is given by 

 the lamps. 



The working of such an arrangement is self- 

 explanatory, one of the two lamps always being 

 alight while the current is on. The current con- 

 sumption on 200 volts is only ith watt — or 5000 

 hours' service for the cost of one unit. 



Where these lamps are required as " pilots " for 

 a large number of circuits, advantage may be taken 

 of the special types designed for advertising purposes, 

 where the electrodes are given the form of letters 

 and other symbols. The makers are prepared to 

 manufacture these in any form if necessary, and 

 such simple words as " on " and " off " could be 

 provided if required. For details of other uses of 

 these interesting lamps reference may be made to a 

 communication in Nature, March 16, p. 343. 



H. J. Denham. 



Botanical Laboratory, Shirley Institute, 

 Didsbury, May 8. 



The Speckled Wave Front o£ Light. 



In view of Sir J. J. Thomson's suggestion (quoted 

 by Reiche, " Quantentheorie," p. 25) that the wave 

 front of light may have a speckled structure, it may 

 be asked whether anything peculiar happens when 

 two specks, belonging to different waves, collide. 

 They might, for example, be deviated from their 

 courses. If so, one light wave would cause some 

 scattering of another wave with which it collided, 

 and the direct light in the second wave would be 

 dimmed. This effect, if it exists at all, must be small 

 or it would have been noticed. Evidence as to its 

 existence might be obtained in the following way. 

 The supposed dimming of the second ray would be 

 likely to depend on the angle between it and the first 

 ray. On this supposition the brightness of a star 

 lying on the ecUptic would vary slightly with the 

 angle between it and the sun, as seen from the earth. 

 Russell {Astrophvs. Journ., vol. xliii., 1916) has 

 shown that the brightness of the moon does so vary, 

 but that has been explained otherwise. 



On the Maxwellian view of light the reduction of 

 observations which is here suggested appears so mean- 

 ingless that it has possibly never been tried. 



Lewis F. Richardson. 



Westminster Training College, S.W.i. 



NO. 2743, VOL. 109] 



