70 



NATURE 



{May 1 6, 1889 



that " le spectre de I'aurore boreale resulte de la superposition 

 de plusieurs spectres differents," and that "la raie principale 

 forme un de ces spectres elementaires ; elle apparait tres souvent 

 seule." A similar view was taken many years ago by Angstrom 

 (Nature, vol. x. p. 210) and by Vogel {Leipzig Math. Phys. 

 Berichte, vol. xxiii. p. 298). 



[After consideration, I think that I ought to point out that 

 Mr. Lockyer's recent statement (Roy. Soc. Proc, vol. xlv., 1889, 

 p. 234), that "the characteristic line of the aurora is the remnant 

 of the brightest manganese fluting at 558," is clearly inad- 

 missible, considering the evidence we have of the position of 

 this line. 



• In support of this statement Mr. Lockyer says : — "Angstrom 

 gave tlie wave-length of the line as 5567, and since then many 

 observers have given the same wave- length for it, but probably 

 without making independent determinations. Piazzi Smyth, 

 however, gives it as 558, which agrees exactly with the bright 

 edge of the manganese, fluting. R. \l. Proctor also gives the 

 line as a little less refrangible than Angstrom's determination. 

 He says : ' My own measures give me a wave-lengih very 

 slightly greater than those of Winlock and Angstrom ' (Nature, 

 vol. iii. p. 468)." 



By reference to Gyllenskiold's table it will be seen that the 

 probable errors of the determinations by Piazzi Smyth and 

 Proctor, 5579 ±9'5 and 5595 ± 25-0 respectively,^ are too 

 large to entitle these measures to special weight. 



Mr. Lockyer says, further : — " Gyllenskiold's measures with 

 the Wrede spectroscope also give 5580 as the v\ ave-length of 

 the characteristic line. I feel justified, therefore, in disregard- 

 ing the difference between the wave-length of ihe edge of the 

 manganese fluting and the generally accepted wave-length of the 

 aurora line." 



GyllensUold's single measure of 5580, on which Mr. Lockyer 

 relies, differs widely from the values which Gyllenskiold himself 

 assigns to this line — namely, from observations at Cape Thord- 

 sen in 1882, A 5568 ± i'6, and from observations at Upsala in 

 1884, with the Wrede spectroscope, A. 5569 ± 6 '2. 



Speaking of Krafft's observations, Mr. Lockyer says (Roy. 

 Soc. Proc, vol. xlv., 1889, p. 241): — "The wave-lengths ob- 

 tained for the aurora line were 5595, 5586, and 5587. Unlike 

 most observations, these place the aurora line on the Itss re- 

 frangible side of the manganese fluting. Hence, we have an 

 additional reason for neglecting the difference bet\^ een the wave- 

 length of the brightest edge of the manganese fluting, and the 

 commonly accepted wave-length of the aurora lin^, as given by 

 Angstrom. . . . These observations are the latest which have 

 been published, and were obviously made with a full knowledge 

 of all previous work, so that their importance must be strongly 

 insisted upon." 



I have already pointed out that Krafft's measures were not 

 made under circumstances which sssurtd to them a high degree 

 of accuracy ; and Krafft's own words, which I have quoted, 

 disclaim expressly any special attempt on his part to redeter- 

 mine the position of the principal line with a hijiher degree of 

 accuracy than the observers who preceded him.^March 4.] 



May 2. — " The Accurate Determination of Caibonic Acid 

 and Moisture in Air." By J. S. Haldane, M.A., M.B., and M. 

 S. Pembrey (Physiological Laboratory, Oxford). Communicated 

 by Prof. J. Burdon Sanderson, F. R.S. 



The authors show that, in spite of the efforts which have been 

 made in recent years to improve the method of Pettenkofer for 

 determining COo in free air, the results obtained by different 

 observers still seriously disagree. They also point out the 

 serious defect in the ordinary " chemical " method of determining 

 moisture in air, that in spite of its superior accuracy it only gives 

 accurate results over a long period, while the proportion of 

 moisture in the air is constantly changing. 



A method is then described for determining simultaneously 

 the CO2 and moisture in air. The method is gravimetric. The 

 CO2 is estimated by the increase in weight of an apparatus of 

 simple construction containing soda lime, through which a known 

 volume of the air has been passed. The moisture is similarly 

 estimated by means of an apparatus containing pumice soaked 

 in sulphuric acid. The increased accuracy and convenience of 

 the method depend on the facts : (i) that a very rapid current 

 of air may be passed through the apparatus without fear of non- 

 absorption of either COo or moisture ; (2) that by the method 

 of counterpoising with a dummy apparatus during weighing the 



' Gyllenskiold's sta'.emer.t of Proctoi's valve is based on Natur", vol. iii. 

 p 347 and p. 68. 



"errors of weighing" are reduced to about a tenth of what they 

 would otherwise be. It is shown that with these two improve- 

 ments the method for moisture gives in a period of experiment of 

 one minute a result equal in accuracy to that obtained with the 

 ordinary method in a period of two hours. 



Using their own method for CO.i as a standard, the authors 

 have also tested the Pettenkofer method. They find that the 

 latter method usually gives resuUs for free air about a fifth too 

 high, but that the error is less in proportion with air containing 

 larger amounts of CO2. 



As a number of sets of absorption apparatus can easily be 

 carried about, the new method is well suited for experiments in 

 hygiene, and especially for cases in which a series of experiments 

 require to be made in rapid succession. Both kinds of absorption 

 apparatus last over a large number of experiments without 

 refilling. 



Physical Society, A] ril 13. — Prof. Reinold, President, in 

 the chair.- — Mr. Sbelford Bid well, F.R. S., showed a lecture ex- 

 periment illustrating the effect of heat en the magnetic suscepti- 

 bility of nickel, and an experiment showing an effect of light on 

 magnetism. In the first experiment a piece of nickel was 

 attached to one side of a copper pendulum bob, which was held 

 out of the vertical by bringing the nickel in contact with a fixed 

 magnet. On placing a spirit-lamp flame below the nickel, the 

 bob was, after a short time, released, and oscillated until the 

 nickel had cooled, when it was again attracted and the operation 

 repeated itself. The second experiment had been recently 

 shown before the Royal Society. One end of an iron bar, which 

 had been ^magnetized and then demagnetized, was placed near a 

 magnetometer needle. On directing a beam of light on the bar 

 an immediate deflection of the needle resultedj and on cutting 

 off the light the needle promptly returned to near its initial 

 position. The direction of magnetization induced by the light 

 is the same as the previous magnetization, and the bar 

 seems to be in an unstable magnetic state. That the effect is 

 due to light and not heat, the author thinks is rendered probable 

 by the suddenness of the action. The President said he had 

 tried the experiment himself and failed to get any effect, but 

 after seeing the arrangement of apparatus used, he believed his non- 

 success due to the comparatively great distance between Ms bar 

 and needle.- Mr.- C. Richardson asked if the results were different 

 for different coloured rays, and Prof. S. P. Thompson inquired 

 whether the magnitude of the effect varied with the intensity of 

 illumination as in selenium, and also if any change was produced 

 by altering the direction of vibration of the incident light. Mr. 

 G. M. Whipple wished to know whether any difference was 

 produced by blackening the bars, and as bearing somewhat on the 

 same subject mentioned an induction magnetometer in which an 

 iron bar used was demagnetized by plunging in hot water. The 

 results obtained were very irregular after the first magnetization, 

 and this may have been due to the instability shown to exist by 

 Mr. Bidwell's experiment. In reply, Mr. Bidwell said red light 

 produces most effect, and blackening the bar makes the action 

 much slower. As regards . selenium, the character of the effect 

 is similar, but he believes the causes to be different. Polarized 

 light produces no change. . In answer to Prof. Herschel, he said 

 that any part of the, bar. is, sensitive to light, and showed that 

 illuminating both sides of the bar increased the effect. — Mr. 

 G. M. Whipple read a note on the dark flash seen in some light- 

 ning photographs. After expressing his dissent from the 

 explanations offered in the report of the Lightning Flash Com- 

 mittee oftbe Meteorolegicail Society and Prof. Stokes respecting 

 ribbon lightning and dark flashes-, the author described some 

 experiments he had made on the subject. Ribbon lightning he 

 conceived to be an effect produced by taking the photographs 

 through windows, and to test this, lines on a blackboard were 

 photographed, (i) direct; (2) through good plate-glass placed 

 obliquely ; and (3) through window-glass, the result being that 

 the double, triple, and ribbon flashes -were closely imitated. As 

 regards " dark flashes," the author believes the appearance due 

 to the prints being taken in oblique light, and to be produced by 

 successive reflection from the reduced silver forming the dark 

 line on the negative and the upper surface of the glass of the 

 negative. Prof. Perry suggested that this might be easily proved 

 by examining a negative, the prints from which .'how the dark 

 flash. Mr. Baily pointed out that, if the explanation given were 

 correct, ih^ d,ark line should be parallel to the bright one, and 

 this he understood was not always the case. Mr. Boys remarked 

 that one dark flash exhibited minute wriggles not seen in the 

 bright one, and Mr. C. V. Burton thought these might be due 



