620 



NATURE 



[Oct. 28, iSSoi 



Hence if a small propoition of the whole fluid is irrotatioiial, 

 it is clear that there may be a minimum energy, and therefore a 

 stable configuration of motion wilh the whole of this in one 

 of the wide parts of the canister ; or the whole in the other ; or 

 any proportion in one and the rest in the other; or a small 

 portion in the elliptic whirl in the connecting canal, and the rest 

 divided in any proportion between the two wide parts of the 

 canister. 



ON THE SPECTRA OF THE COMPOUNDS 

 OF CAP BON WITH HYDROGEN AND 

 NITROGEN 



TV/TESSRS. LIVEING AND DEWAR have made a long 

 series of observations on this subject, of which the follow- 

 ing is a brief abstract ' by the authors : — 



The first experiments were made with a De Meritens dynamo- 

 electric machine, arranged for high tension, giving an alternating 

 current capable of producing an arc between carbon poles in air 

 of from 8 to 10 millims. in length. The carbon poles used had 

 been previously purified by prolonged heating in a current of 

 chlorine. 



The arc was taken in different gases inside a small glass globe 

 about 60 miUims. in diameter, blown in the middle of a tube. 

 The two ends of the tube were closed with dry corks, through 

 which were passed (i) the carbons inserted through two pieces 

 of narrow glass tubing ; (2) two other glass tubes through which 

 currents of the gases experimented with were sent. 

 _ The arc taken in a globe of air gave a tolerably bright con- 

 tmuous spectrum, above which the green and blue hydrocarbon 

 bands were seen, also the seven bands in the indigo (wave-len<Tt]is 

 4,600 to 4,502, Watts) as in the flame of cyanogen, and m°uch 

 more brightly the six bands in the violet (wave lengths 4,220 to 

 4>IS8> Watts) and five ultra-violet. 



Carbonic add gas, hydrogen, nitrogen, chlorhte, carbonic oxide, 

 nitric oxide, and anniionia were then successively substituted for 

 air in the globe, with the result that in carbonic acid, hydrogen, 

 chlorine, and carbonic oxide, the above-mentioned bands in°the 

 indigo, violet, and ultra-violet died away, while in nitrogen, 

 nitric oxide, and ammonia, they were always well seen. 



These different gases were u?ed in order to eliminate to a large 

 extent the influence of electric conductivity on the character of 

 the spectrum ; and the green and blue hydrocarbon bands were 

 seen, more or les5, in all of them. 



Next observations were made of the spectra of flames of 

 sundry compounds of carbon. 



In the flame of cyanogen, prepared from well-diied mercury 

 cyanide, passed over phosphoric anhydride inserted in the same 

 tube, and burnt from a platinum jet fused into the end of the 

 tube, the hydrocarbon bands were almost entirely absent, as 

 Pliicker and Hittorf had found ; only the brightest green band 

 was seen, and that faintly. The indigo, violet, and ultra-violet 

 bands, on the other hand, were well developed. 



These three sets of bands in the indigo, violet, and ultra-violet 

 are in the sequel referred to as the "cyanogen bands," though 

 it is possible that they may be producible by other compounds of 

 carbon with nitrogen. 



The flame of hydrocyanic acid burning in air showed very 

 mnch the same as that of cyanogen. 



In the flame of a mixture of purified hydrogen and carbon 

 disulphide no hydrocarbon bands at all could be detected. 



Nor could they be detected in the flame of a mixture of car- 

 bonic oxide and hydrogen burnt in air. 



When a mixture of hydrogen or of carbonic oxide with carbon 

 tetrachloride vapour was burnt, hydrocarbon bands made their 

 appearance, but were weak. 



On the other hand, chloroform, when mixed with hydro'^en 

 save, when burnt in air, the hydrocarbon bands very strongly! 

 On a review of the whole series of observations, certain points 

 stand out plainly. In the first place, the indigo, violet, and 

 ultra-violet Irand^, characteri.stic of the flame of cvanogen, are 

 conspicuous in the arc taken in an atmosphere of nitrogen, air, 

 nitric oxide, or ammonia, and they disappear almost, if not 

 quite, when the arc is taken in a non-nitrogenuus atmosphere of 

 hydrogen, carbonic oxide, carbonic acid, or chlorine. These 

 same bands are seen brightly in the flames of cyanogen and 

 hydrocyanic acid, but are not seen in those of hydrocarbons, 

 carbonic oxide, or carbon disulphide. The conclusion seems 

 irresistible that they belong to cyanogen ; and this conclusion 

 ' For fuller delails see Proc. R.S., xxx. pp. 153, 494, 



does not seem to be at all invalidated by the fact that they are' 

 seen weakly, or by flashes, in the arc or spark taken in gases' 

 supposed free from nitrogen by reason of the extreme difficulty' 

 of removing the last traces of air. They are never, in such a 

 case, the principal or prominent part of the spectrum, and in a' 

 continuous experiment they are seen to fade out in propoition as 

 the nitrogen is removed. This conclusion is strengthened by the 

 oL'servations that cyanogen (or hydrocyanic acid) is generated in' 

 the arc in atmospheric air in large quantity. 



In the next place, the green and blue bands, characteristic of ' 

 the hydrocarbon flame, seem to be always present in the arc, ' 

 whatever the atmosphere. This is what we should expect if 

 they be due, as Angstrom and Thalen suppose, to acetylene ; 

 for the carbon electrodes always contain, even when they have 

 been long heated in chlorine, a notable quantity of hydrogen. 

 In the flames of carbon compounds they by no means always 

 appear ; indeed it is only in those of hydrocarbons or their 

 derivatives that they are well seen. Carbonic oxide and carbon 

 disulphide, even when mixed with hydrogen, do not show them ; 

 and if seen in the flames of cyanogen, hydrocyanic acid, and 

 carbon tetrachloride mixed with hydrogen, they are faint, and 

 do not form a princiial or prominent part of the spectrum. 

 This is all consistent with the supposition of Angstrom and 

 Thalen. The fact that the bands are not produced even in the 

 presence of hydrogen, when it is not present in the flame in the 

 form of a compound with carbon, is very significant ; for we know 

 that acetylene is present, and can eadly be extracted from the flame 

 of any hydrocarbon, and that it is formed as a proximate pro- 

 duct of decomposition of hydrocarbon by the electric discharge, 

 but we have no evidence that it is producible as a product of 

 direct combination of carbon with hydrogen at the comparatively 

 loiv temperature of the flames described. 



The hydrocarbon bauds are best developed in the blowpipe 

 flame, that is under conditions which appear, at first sight, 

 unfavourable to the existence of acetylene in the flame. How- 

 ever, by the use of a Deville's tube, acetylene may readily be 

 withdrawn from Ihe interior of such a flame, and from that part 

 of it which shows the hydrocarbon bands mo,t brightly. 



The question as to whether these bands .are due to carbon 

 itself or to a compound of carbon with hydrogen, has been 

 somewhat simplified by the observations of Watts, Salet, and 

 others on the spectrum of carbonic oxide. It can Jiardly be 

 doubted now that that compound has its own spectrum quite 

 di-tinct from the hydrocarbon flame spectrum. The mere 

 presence of the latter spectrum feebly developed in the electric 

 discharge in compounds of carbon supposed to contain no 

 hydrogen, weighs vei-y little against the series of observations 

 which connect this spectrum directly with hydrocarbons. 



In the next place, it appears conclusively from the experi- 

 ments, that the development of violet bands of cyanogen, or 

 the less refrangible hydrocarbon bands, is not a matter of 

 temperature only. For the appearance of the hydrogen lines C 

 and F, obiCrved by the authors in the arc taken in hydrogen, 

 indicates a temperature far higher than that of any flame. Yet 

 the violet bands are not seen in hydrogen at that temperature, 

 while the green bands are well developed. The violet bands 

 are, nevertheless, seen equally well at the different temperatures 

 of the flame, arc, and spark, provided cyanogen be the compound 

 under observation in the flame, and nitrogen and carbon are 

 present together at the higher temperatures of the arc and 

 spark. 



The accompanying diagram (Fig. l) shows approximately the 

 relative position of the bands in that part of the spectrum of the 

 flame of cyanogen fed with a jet of oxygen which is more 

 refrangible than the Fraunhofer line F. Only those bands 

 which are less refrangible than the solar line L have been before 

 described, but photographs show two shaded bands slightly less 

 refrangible than the solar line N accompanied by a very broad 

 diffuse band of less intensity on the more refrangible side of N ; 

 also a strong shaded band, which appears to be absolutely 

 coincident with the remarkable shaded band in the solar spec- 

 trum, which has been designated by the letter P ; and near 

 this, on the less refrangible side, a much fainter diffuse band, 

 which also seems to coincide with a part of the solar spectrum 

 sensibly less luminous than the parts on either side of it. This 

 spectrum is remarkably persistent at all temperatures of the 

 flame. Watts found that it did not disappear when the flame 

 was cooled down as much as possible by diluting the cyanogen 

 with carbonic acid ; it retains its characters when the cyanogen 

 is burnt in nitric oxide. The flame in the last case must be one 



