172 



NATURE 



[December 21, 189; 



carbon dioxide. These flames are therefore described by him as 

 being "simply a hollow conical sheath of pretty uniform char- 

 acter." This is undoubtedly a true description ; neither of 

 these flames presents the appearance of double coned structure 

 which is seen in such flames as cyanogen, carbon disulphide, 

 ammonia, and others ; and it is hardly possible that in a hydro- 

 gen or carbon monoxide flame there can be two distinct areas 

 or cones in which different chemical processes are going on. 

 It occurred to me that it might throw some light upon the real 

 value of this cone-separating apparatus as an appliance for 

 dissecting flames, to try its effect upon the single-coned flames 

 of carbon monoxide and of hydrogen. When air was cautiously 

 admitted into these gases, as they burned at the top of the tube, 

 I found that the flame travelled quickly right down the tube, 

 and did not stop at the narrower tube when the upward rate of 

 movement was greater, and did not appear to leave any remnant 

 at the top of the wider tube. I have no doubt but that Prof. 

 Smithells has made this experiment, and with a similar result. 



I have found, however, that by a slight modification of the 

 apparatus, it is quite easy to drag down an inner flame from 

 either the flame of carbon monoxide or of hydrogen. In order 

 to do this, all that is necessary is to provide the top of the inner 

 and narrower tube with a cap made of fine wire gauze, either 

 copper or platinum. When this small addition to the original 

 apparatus is made, and the experiment with carbon monoxide is 

 repeated, it will be seen that as air is gradually introduced a 

 portion of the flame descends the tube and sits quietly upon the 

 wire gauze, and, in spite of the flame-extinguishing power of 

 the carbon dioxide^t there generates, a remnant of the original 

 flame remains feebly burning at the top. In the case of hydrogen 

 a similar result is obtained, a portion of the flame descending to 

 the gauze, where it burns with a pale blueish flame, while 

 the remnant burns freely at the top. These experiments show 

 that whatever is the structure of the flame, a part of it can be 

 torn away from the rest by the regulated introduction of air : 

 that in order to divide a flame by this method it is not a neces- 

 sary condition that the flame should consist of more than one 

 "cone," or, in other words, that there should be two distinct 

 areas of combustion. If, therefore, a "simple" flame like 

 that of hydrogen, consisting of a single cone of uniform character, 

 can be divided, the fact that other and more complex flames can 

 also be so divided, does not seem to throw much light upon their 

 structure. As soon as sufficient air has been admitted into a 

 flame, of whatever burning gas, to produce a certain volume of 

 an explosive mixture whose rate of explosion exceeds the rate of 

 efflux of the gases, that exploding mixture will become detached 

 from the remainder of the burning gas, and travel back down the 

 tube. In the case of hydrogen, where a very wide margin exists 

 within which mixtures of this gas and air are rapidly explosive, 

 the admission of a very small quantity of air is sufficent to form 

 such a mixture, and so drag down a comparatively small por- 

 tion of the entire flame. In the space between these two flames 

 there can only be water vapour as the product of combustion, 

 atmospheric nitrogen, and the excess of hydrogen. The lower 

 flame is a burning mixture of air and hydrogen in which an ex- 

 cess of air is taking part in the combustion, and represents a 

 condition of things certainly not far removed from, if not iden- 

 tical with, the old phenomenon of air burning in hydrogen. It 

 is difficult to see in what way the separation of other flames 

 differs from this. 



I have no doubt that everyone who has read the account of 

 Prof. Smithell's lecture will have been struck, as Dr. Armstrong 

 was, with the manner in which the classical researches of Dr. 

 Frankland are brushed aside, and the difficult question as to 

 the true causes of the luminosity of flame is settled by an 

 appeal to the " opinion of the majority." 



Without touching the question as to whether or not solid 

 carbon is actually precipitated during the decompositions that 

 are going on in a coal-gas flame, the recent experiments of Prof. 

 Lewes leave no room for doubt that the first stage in the process 

 of decomposition and condensation that goes on, is the produc- 

 tion of acetylene, which is formed during the passage of the gas 

 through the inner dark area of the flame, where no combustion 

 is going on ; that is to say, where the hydrocarbons are being 

 simply strongly heated, tmt are not burning. This fact seems 

 to have an interesting bearing upon some of the peculiarities ex- 

 hibiteii by the well-known flame of air burning in an atmosphere 

 of coal-gas. In this flame the air is in the inside, and the hydro- 

 carbons upon the outside ; it is in effect an ordinary coal-gas 

 flame turned inside out. The formation of acetylene, instead of 



NO. 1260, VOL. 49] 



taking place within the flame, as in the usual conditions, in 

 which case it has to pass through the heated area where it is 

 further decomposed with probably the precipitation of carbon, 

 is now produced upon the outer surface or periphery of the 

 flame ; it therefore largely escapes combustion or decomposi- 

 tion, and passes into the coal-gas atmosphere with which the 

 flame is enveloped. Hence the flame is non-luminous, and 

 hence also this constitutes the ready method for obtaining large 

 quantities of acetylene first devised by Prof McLeod. I am 

 not aware that it has ever been noticed that during the combus- 

 tion of this non-luminous flame there are produced, besides 

 acetylene, other hydrocarbons of much greater density. That 

 this is so is evident from the fact that when the flame has been 

 allowed to continue burning for a length of time, the glass 

 vessel in which it is contained becomes coated with a brown 

 tarry film. This non-luminous flame of air burning in coal-gas 

 can be rendered luminous by a simple device. If the vessel 

 employed in which to burn it be an ordinary bulb-shaped 

 paraffin lamp chimney, it will be seen that when the flame is in 

 the middle and wide portion of the chimney it is non-luminous; 

 if, however, it be thrust up into the narrow part, it at once 

 shows signs of luminosity : the acetylene under these circum- 

 stances is reflected back into the flame, which, aided no doubt 

 by the radiated heat from the glass, causes the luminosity. If 

 the supply of air be regulated, the flame may be caused to curl 

 over upon itself, whereby very beautiful vortices are obtained, 

 in which Heumann's floating particles are well seen. There 

 is an old experiment in which two flames of air in coal- 

 gas are placed side by side, and so arranged that at will they 

 can be caused just to impinge upon each other. At the point 

 where they touch a small luminous area is seen to appear, the 

 luminosity being probably due to the same causes. 



G. S. Newth. 



I AM unable to understand how Prof. Smithells can in any 

 way suppose that I either have, or possibly could, cast any im- 

 putation on his honesty, "scientific" or otherwise ; and I fail 

 also to understand what has given rise to the impression, unless 

 it be that the opening sentence of my letter — which I intended 

 should convey a compliment — has been turned round and a 

 meaning given to it which I never contemplated, and which it 

 cannot fairly be made to bear. 



I have always regarded Nature as a journal which is willing, 

 to aftbrd a fair field for the consideration of scientific problems, 

 but the last place in which to raise, let alone discuss, personal 

 questions. By publishing his lecture in Nature, Prof Smithells 

 directly challenged criticism, and the only object and intention of 

 my letter was to challenge the validity of certain of his arguments. 

 That he should have taken the view he has, is to me a matter of 

 deep regret. He has now stated his position very clearly, and 

 the passage that he has been good enough to quote from my 

 letter to Sir G. G. Stokes sufficiently defines mine. I fear that 

 we must agree still to differ ; evidently we look at these matters 

 from very dissimilar standpoints. 



Henry E. Armstrong. 



The Postal Transmission of Natural History Specimens, 



At page lOo, ante, you reproduce a circular letter, sent out by 

 the Academy of Natural Sciences of Philadelphia, on this 

 subject, the object of which is the very laudable one of estab- 

 lishing an international rate of postage for natural history 

 specimens, based on that charged for bona fide trade patterns 

 and samples. It is therein stated that the United States Post 

 Office Department recently proposed to the countries comprised 

 within the Postal Union a modification of the rates in favour of 

 a charge so based, but that the Governments of very many of 

 them declined to consider the proposal, and in the list there 

 given Great Britain is included. No precise date for this refusal 

 on the part of the British postal authorities is given, but presum- 

 ably the date is not precisely recent. Early in 1891, several of 

 our Natural History Societies agreed to approach the British 

 postal authorities on this point, and a letter was addressed to 

 the Secretary of the Post Office (the late Sir S. A. Blackwood) 

 by Lord Walsingham, on March 18, 1891. A reply (which I 

 have before me) to that letter, from Sir S. A. Blackwood, is dated 

 April 13, 1891, and is published in the Proceedings of the 

 Entomological Society of London, 1891, p. 14 (and probably 

 elsewhere). An extract from the letter is to this effect : — " Your 



