July 29 18S0] 



NATURE 



291 



perfect of these patches. The soil is a sandy loam upon the 

 Teensand, and the crop is vigorous, with strong stems, and I 

 could not trace locally any circumstances accounting for the 

 peculiar forms of the patches in the field, nor indicating whether 

 it was wind or rain, or both combined, wbicli had caused them, 

 beyond the general evidence everyw here of heavy rainfall. They 

 were to me suggestive of some cyclonic wind action, and may 

 perhaps have been noticed elsewhere by some of your readers. 

 Guildown, Guildford, July 23 J. Kand Capron 



The Inevitable Test for Aurora 



I HAVE not long returned from abroad, and have only re- 

 cently had the opportunity of perusing in Nature (vol. xxii. 

 PP- 76. 96, 145) the correspondence of Messrs. De La Rue and 

 MiiUer, Prof. Piazzi Smyth, and Mr. Backhouse on this subject. 

 I do not understand Messrs. De La Rue and Midler as claim- 

 ing their electric discharges to be in the nature of an actual 

 auroral discharge, but rather that their experiments inform us 

 inductively at what heights aurorx are to be found. This, how- 

 ever, doubtless assumes that the discharges in question and 

 anrorje must have something very much in common ; and Prof. 

 Piazzi Smyth is quite to the point in rem.arking that unless the 

 citron line (and, I would add, the red line) are present in the 

 spectrum, the identity of the discharges with the aurora has not 

 even a foundation. 



The fact t, that many of the electric discharges in air and the 

 air gases, and the circumstances attending them — we may in- 

 stance the ordinary tube glow, its change from rose-tint to violet 

 under magnetic .influence, the aura-arc accompanying the 

 spark discharge under similar conditions, the dark space between 

 the terminal and the glow, the change of colours in a hydrogen 

 tube, and other appearances which I have not time to capitulate 

 — so closely resemble auroral incidents, that one is quite disap- 

 pointed to find on examination no concordance in the spectra. 

 At the most, in a vague and unsatisfactory way one or two of 

 the blue anil violet lines in the aurora spectrum have been 

 assigned to one or other of the atmospheric gas-S ; but as Prof. 

 Smyth points out, the red and green giant lines of the spectrum 

 have up to the present time found no terrestrial analogues. I 

 have examined the air spectrum and the spectra of the com- 

 ponent gases of air under many various conditions, but always 

 without success so far as these lines are concerned. 



I have not, however, had the opportunity of doing this in the 

 case of direct discharges from large secondary batteries ; and it 

 would undoubtedly be a valuable addition to our knowledge of 

 facts relating to aurora; if Messrs. De La Rue and Midler would 

 undertake this examination, and clear up matters in that respect. 

 With regard to the heights at which aurora obtain, the evidence 

 is very conflicting. Certainly they have been seen very near the 

 earth ("Aurora;, their Characters and Spectra," pp. 37 to 40. 

 Height of the Aurora). It is unfortunate that simultaneous 

 observations of the auroral corona are almost entirely wanting. 

 I think I once saw one in print, but missed it subsequently, and 

 would be glad if any particulars could now be furnished me. 

 Prof. Newton, by calculations based on observations of auroral 

 arches in 2S aurorre, lias assigned a height of from 33 to 281 

 miles, with a mean of 130 miles. 



Messrs. De la Kue and MuUer, I notice, deduce experimentally 

 that at 124 miles no discharge could occur. As to whether the 

 red or the white aurora is the nearest to the earth, my impression 

 certainly is that the apparently low-lying aurora have generally 

 been the \\hite. I may instance the aurora seen by Mr. Ladd a 

 Margate, "a white ray," and that seen by me in the Isle of Skye 

 in September, 1S74. In Lapland, too, the aurora; seem almost 

 universally yellow, but it can hardly be assumed that they are all 

 thirty-seven miles high. The apparently lower position of the 

 red tint is by no means universal, and can hardly be relied upon 

 as evidence on the point, especially when so many aurora; are 

 seen in which it is wanting. I have great hopes, with a spectro- 

 cope specially prepared for the purpose, of getting the photo- 

 graphed spectrum of an aurora. 



The red line is of course out of the question, but judging from 

 experiments on gas tubes I think the green might be got, and the 

 blue and violet I make in anticipation pretty sure of in the event 

 of an aurora lasting some hours. The principle of the instrument 

 is a long collimator, a single fluid prism, and a short focus- 

 projecting lens, used with rapid dry plates. 



Guildown, Guildford, July 23 J. Rand Capro.n 



Experiment with Glass Tubes 



I HAVE just been repeating a very beautiful experiment of 

 Prof. Quincke's which he showed me some weeks ago in his 

 laboratory at Heidelberg. The experiment was, I believe, 

 described in Poggcndorff about two years ago, but I have not 

 seen it noticed in English papers, and a few words about it may 

 interest your readers. 



Prof. Quincke, with a view to test the porosity of glass for 

 gases, sealed up tubes in which hydrogen and carbonic acid were 

 generated in great quantity, and weighed them from time to time. 

 Up to the present time, as I learned from him, t\.-> loss of weight 

 has been detected. He obtained, however, a very curious result. 

 As I do not know precisely how Prof. Quincke filled his tubes, 

 let me describe what I did myself three weeks ago, remarking 

 that I have done nothing but attempt 

 to repeat what he showed me in 

 Heidelberg. 



I took a glass tube, A B, about S 

 inches long and f inch in external 

 diameter, with good stout walls. I 

 closed the end A, and let the glass 

 fall in at c, keeping it still very 

 strong, and annealing very carefully 

 at A and c. I introduced some sul- 

 phuric acid into the part c A, care- 

 fully keeping the neck C dry, and 

 dropped into the part B c some frag- 

 ments of marble, previously washed, 

 in order that no little particles should 

 tumble down through the neck, c, and 

 commence efferveseirg before I was 



ready. I then drew out the tube at B, making a small hook, by 

 which the tube can be suspended if necessary, closed it very 

 strongly, and annealed the extremity carefully, wrapped the tube 

 in cotton wool, and inverted it. The sulphuric acid attacked the 

 marble, and carbonic acid was given oft" no doubt in great 

 quantity. 



For the first few days there was nothing particular to be 

 noticed. The tube vvas filled with a bubbling mass of liquid and 

 white mud. Latterly, however, it has begun to show the 

 phenomena which Prof. Quinclie observed. The liquid now no 

 longer wets the glass as it did at first, but creeps away from it, 

 giving very much the appearance of the "tears of strong wine." 

 Day by day this is getting more marked, and I expect that 

 soon, as was the case in the Heidelberg tubes, the acid -will roll 

 about in the tube like so much quicksilver. Meantime it is most 

 interesting to watch. 



I believe Prof. Quincke considers that a thick layer of gas is 

 condensed over the surface of the glass, and that it is this which 

 gives rise to the very peculiar capillary phenomena that present 

 themselves. 



I feel bound to remark that the experiment is one that ought 

 not to be attempted without great care and caution. 



J. T. BOTTOMLEY 



Physical Laboratory, University of Glasgow, July 15 



On the Colours of Double Stars 



If any light whatever has its intensity increased the effect on 

 the eye is to add to the sensation a certain yellow element which 

 I have accurately defined by experiment (Am. your. Sci., ApriJ, 

 1877, vol. xiii. p. 247). A red light brightened becomes 

 yellower, a green light yellower, a yellowish white less white, a 

 blue or violet light whiter. The phenomena are described at 

 length in Prof. Rood's "Modern Chromatics." The fact that 

 an incandescent body becomes less red and more yellow when it 

 is heated is probably due to this physiological principle. That 

 the incandescent body ultimately becomes white is probably 

 owing to some not understood modification of the principle for 

 excessively bright lights. 



It follows that if two stars are of unequal brightness they will 

 appear of different coloiu-s unless the qualities of the two lights 

 have a peculiar relation to one another ; and the brighter star 

 will usually be the yellower. Accordingly, if we refer to Mr. 

 Burnham's lists of binaries recently published by Prof. Holden 

 (Am. Jour. &;'., June, iSSo, vol. xix. p. 467) we find that although 

 differences of colour are so little distinguished that three-quarters 

 of all the pau-s are considered to be of the same colour, yet of the 

 twenty-four pairs which differ in brightness by two magnitudes 



