February 2, 191 1] 



NATURE 



439 



juine under normal conditions ; they were taken in a 

 sealed-off portion of Seaham Collierj" after an explosion 

 and an undergrour\d fire, and thus represent what took 

 place under an entirely abnormal condition of the mine. 

 Apart from this point, the Author of the Warnings con- 

 trives to give his reader the impression that Mr. Corbett's 

 Seaham Colliery records entirely favour his own views, 

 that high barometric pressure causes an increase in fire- 

 damp in mines ; so far from this being the case, how- 

 ever, Mr. Corbett's own words (Trans. North Eng. 

 Inst. Min. Eng., vol. xxxii., 1882-3, P- S^o) ^''^ • — ^" ^^ 

 is well known that gas is frequently found in colliery 

 workings before any fall of the barometer commences. 

 , . . The barometer, so far as an indication showing that 

 gas may be expected, cannot be said to be reliable." In 

 the discussion of this paper, Mr. J. Daglish {ibid., p. 311) 

 said that he had made experiments at Hetton Colliery, 

 and that " the results he arrived at were precisely such 

 as were given by Mr. Corbett, namely, that there was no 

 connection whatever between the variations of the baro- 

 meter and the prevalence of gas in the galleries of the 

 mine." The chief witness cited in his favour by the 

 Author of the Warnings is thus seen to give evidence quite 

 directly against him when he is quoted correctly. Further, 

 if the .Author of the Warnings attaches the importance 

 ihat he appears to do to these records of pressure, why 

 does he not quote the verj- well known and much more 

 applicable experiments of Sir Lindsay Wood, who deter- 

 mined the pressure of firedamp in normal coal seams by 

 boring holes into them and inserting pressure gauges? 

 His general conclusions (Trans. North Eng. Inst. Min. 

 Eng., vol. XXX., 1880-1, p. 224) are : — " There is no 

 connection between the variations of the barometrical 

 column and the temperature with the quantities of gas 

 evolved " ; only in one set of tests, namely, at Eppleion 

 Colliery, was any connection traceable, and, respecting 

 these. Sir Lindsay Wood (ibid., p. 182) states : — " With 

 the barometer steadily rising, the gas pressure (with one 

 cr two exceptions, when there was an increase) steadilv 

 decreased. " 



Personally, I attach relatively little importance to rec<M-ds 

 of pressure alone, even to such careful ones as those of Sir 

 Lindsay Wood, Nasse, Broockmann, &c. ; in the absence 

 of analyses, it is only a conjecture that the pressure was 

 caused by firedamp, and in the case of Seaham Collierv 

 it is quite likely that other gases were present in large 

 quantity. I hold that there is only one correct method of 

 attacking this question, as has already been pointed out 

 by Oberbergrat G. Kohler, and that is by svstematic 

 chemical analyses of the return mine air combined with 

 barometric obsenations, as has been done on several 

 occasions on the Continent, e.g. by Hilt at the Gemein- 

 schaft and Alt-Gourley pits at Aachen, and, above all, bv 

 AV. Kohler at the Grand Duke Frederick pits at Karwin. 

 All the observations corroborated each other, and agree 

 with the summary of W. Kohler: — "The proportion of 

 firedamp in the air of the mine decreases in general with 

 rising atmospheric pressure, and increases with falling 

 atmospheric pressure. The proportion of firedamp in- 

 creases the more rapidly the more steeplv the curve of 

 atmospheric pressure descends, and decreases the faster 

 the more steeply the curve of atmospheric pressure rises." 

 Harze in Belgium and Behrens in Westphalia have con- 

 firmed these conclusions in their elaborate works on the 

 subject. All this is the result of accuratelv observed facts, 

 into none of which "theory" enters. All workers and 

 observers in this subject have come to one of two con- 

 clusions, either that barometric variations have no decisive 

 influence on the evolution of gas, or else that a falling 

 barometric gradient increases the outflow of gas. Not a 

 single writer, so far as I know, shows that a rising 

 barometer increases the evolution of firedamp. Whilst 

 most English authorities hold the first view, the universally 

 held opinion in Germany is summed up thus bv the well- 

 known Saxon authority, E. Treptow : — " Im 'besonderen 

 ^t es als erwiesen anzusehen, dass nach einem schnellen 

 Fallen des Barometers starkere Gasentwickelungen statt- 

 finden. Es ist daher die fortlaufende Beobachtung der 

 Barometerstande von grosser Wichtigkeit ; tritt ein Baro- 

 metersturz ein, so ist besondere Vorsicht geboten. Ein 

 Barometerfall von i mm. in einer Stunde "ist schon sehr 

 NO, 2153, VOL. 85] 



bedeutend." (In particular, it may be looked upon as 

 demonstrated that, after a rapid fall of the barometer, 

 stronger evolutions of gas take place. The continual 

 observation of the height of the barometer is therefore of 

 great importance ; if a drop of the barometer takes place, 

 special caution must be observed. A fall of the barometer 

 of I mm. per hour is already very serious.) 



It is facts like the above-quoted analyses that alone can 

 decide this question ; it is quite useless to inquire whether 

 the barometer was high or low at the time of any particular 

 collier}- explosion, because a serious collier}- explosion can 

 only be brought about by the fortuitous coincidence of a 

 number of contributory conditions, only one of which (and 

 in all probability a relatively unimportant one) can be 

 ascribed to the state of the barometer. The Author of 

 the Warnings implies that my views have been influenced 

 by newspaper statements as to the height of the baro- 

 meter at the time of the great Courrieres disaster ; but 

 not <K»ly do I, as I have said, regard such evidence as 

 useless, but, above all, I would not commit the crowning 

 absurdit}- of quoting in a discussion on firedamp the 

 Courrieres explosion, which is perfectly well known to 

 have been a coal-dust explosion in a non-fiery pit. 



Perhaps the most interesting point in the letter of the 

 .Author of the Warnings is his explanation of the reason 

 why high barometric pressure must increase the per- 

 centage of gas in a pit ; he believes that the increased 

 pressure of the air squeezes down the earth's crust, and 

 squeezes the gas out of it. I presume that he wishes this 

 explanation to be taken seriously ; but surely he has over- 

 looked the very obvious fact that any increase of pressure 

 on the surface of the earth, tending to squeeze gas out, 

 is counterbalanced by an exactly equal increase of pressure 

 upon the face of the coal in the mine, tending to keep the 

 gas in, and that no variation of atmospheric pressure can 

 thus disturb the previously existing regime. Even if this 

 were not so, and if the crust of the earth could respond 

 to such pressures, they are too insignificant to have any 

 practical effect. An enormous fluctuation of barometric 

 pressure, such as a rise of i inch, would correspond to a 

 pressure on the earth's crust of only 70 lb. per square 

 foot, or a good deal less than that of an ordinary crowd 

 of people standing on the ground ; the very suggestion 

 that such a trifling weight can have any effect through 

 thousands, or even hundreds, of feet of strata is so absurd 

 as to require no refutation, and least of all to the mining 

 engineer who has had to timber underground workings, 

 and who knows that the roof pressure in a mine must be 

 gauged, not in pounds, but in tons on the square foot, 

 and that 70 lb. more or less will make no practical differ- 

 ence whatever. That such a theory should be relied on 

 in defence of the Colliery Warnings surely justifies their 

 opposition by mining engineers, and forms an emphatic 

 endorsement of the verdict of the last Royal Commission 

 — which, by the way, was not composed of professors of 

 mining or theorists — upon these Warnings as misleading 

 and serving no useful purpose. H. Louis. 



The Afterglow of Electric Discharge in Nitrogen. 



Ix a paper published in the current number of the 

 Physical Society's Proceedings, I showed that the vellow 

 afterglow produced by the electric discharge in rarefied 

 air is due to the oxidation of nitric oxide by ozone, both 

 substances being formed in the discharge. In a second 

 paper, in course of publication, it is shown that several 

 other oxidisable gases or vapours inflame spontaneouslv 

 when mixed with ozone at a low pressure, and bum with 

 phosphorescent flames of low temperature. 



.\n afterglow in nitrogen has been recorded by Mr. 

 Perceval Lewis (Phys. Zeit., v.. p. 546, 1904) which is 

 obtained only with condenser discharges. This glow is 

 orange in colour, and possesses a visual spectrum of three 

 bright bands in the green, yellow, and red regions, in 

 contrast to the continuous spectrum of the glow which I 

 have traced to nitric oxide and ozone. 



I have recently experimented with Lewis's nitrogen 

 glow, using the method, introduced by Dewar in 1888, of 

 drawing a continuous current of the gas through the 

 vacuum tube into another vessel on its way to the pump> 



