January 26, 1893] 



NATURE 



".o: 



I 



doubtless the reason that on some occasions, as in 1870, the 

 green line is seen beyond the corona — even upon the lunar disc. 



Prof. Hastings, in 1883, examined simultaneously with a 

 special arrangement the spectra of east and west portions of the 

 corona, and proved, conformably to the theory that he pro- 

 pounds, that the green line varied in length during the duration 

 of the eclipse, and that it always extended furthest on the most 

 illuminated side of the edge of the moon. Mr. Keeler repeated 

 the experiment in 1889, and also noted that the length of the 

 green line depends upon the position of the sun with respect 

 to the moon. The question would be worth studying further. 



The green line is not the only bright line in the spectrum of 

 the corona, the hydrogen lines have also been discovered in it, 

 but these never extend further than about 10' from the sun's 

 limb. Other bright lines in the red and in the violet were 

 observed by M, Tacchini and by Thallon in 1882. It was 

 in 1882 also that Prof. Schuster obtained the first photograph of 

 the coronal spectrum upon which some thirty bright lines may 

 be counted. 



In addition to the incandescent solid or liquid matter pro- 

 ducing the spectrum of the corona, and the incandescent gases, 

 which give rise to bright lines, there must also be in the circum- 

 solar regions matter reflecting the light of the photosphere, as 

 our own atmosphere does. This is proved by the polarisation 

 of the light of the corona, and by the presence in its spectrum 

 of the dark lines of the Fraunhofer spectrum. We owe the dis- 

 covery of these dark lines to M. Jaussen. In 187 1 he observed 

 only the lines D and b, but, since, in 1883, he has recognised 

 some hundred dark lines 'bus showing that the complete 

 Fraunhofer spectrum is found in the coronal spectrum. These 

 dark lines are necessarily very faint, for they are drowned in the 

 continuous spectrum. As a rule the line D is most conspicuous, 

 although, according to Prof. Hastings, if a faint solar spectrum 

 is projected on to the continuous spectrum of a gas flame, it is 

 not the line D, but rather the group b, which is by far the most 

 apparent. Prof. Hastings concludes from this experiment that 

 the continuous spectrum of the corona is richer in green than in 

 orange radiation, since it causes the group b to disappear before 

 the line D. 



In conclusion I must quote a remarkable observation made 

 by Prof. Tacchini in 1883, which, should it be confirmed, would 

 suggest a very fascinating theory of the corona. Upon examining 

 the spectrum of one of the sheaves (panaches) of the corona 

 with a considerable dispersion and a wide slit. Prof. Tacchini 

 thought he recognised two or three bright bands characteristic 

 of the hydrocarbons, which are always present in the spectra of 

 comets. Father Perry in 1886 proposed to verify the observa- 

 tion of Tacchini, but unfortunately could not re-observe 

 the bands in question. Certainly he used a spectroscope 

 with slightly illuminated cross wires, and when the period of 

 great solar activity had already passed. It would be well in 

 future eclipses to devote some seconds to the search for these 

 bands, for, if the presence of carbon were recognised in the 

 coronal atmosphere, it would be a new proof of the analogy 

 which exists between the corona and cometary masses. Like 

 comets the corona seems formed of matter subject to a repulsive 

 force on the part of the sun, indeed it is probable that solar 

 gravity does not act upon the corona, for unless this were so, the 

 lower parts, having to support the weight of the upper, would be 

 much more dense than the latter. It would thus result that the 

 lines of the coronal spectrum, the line 1474 for instance, would 

 be wider at their bases than at their upper extremities ; but 

 nothing of the kind has hitherto been observed. Moreover, so 

 that the corona may be visible at 30' or 40' from the sun, the 

 coronal matter must necessarily not be too rare in these extreme 

 regions ; but even in ascribing an extremely low density to this, 

 we should find upon allowing for solar gravity that the pressure 

 near the sun would have a considerable value, although it is 

 proved that the pressure at the base of the corona does not ex- 

 ceed some millimetres of mercury. 



It is also sought to prove the slight density of the middle 

 corona by the fact that it has never offered any resistance to 

 comets, which, on several occasions, have passed through it ; 

 but as comets themselves experience no appreciable resistance 

 when they encounter a body it is impossible to tell whether 

 the absence of resistance is due to the comets or to the 

 corona. 



The repulsive force which expels the coronal matter from the 

 sun would act in the same manner as electrical force ; indeed 

 Prof, Bigelow has noticed that the arrangement of plumes and 



sheaves round the solar disc, and the ircurvilinear forms exactly 

 recall the lines of force of an electric field. Let us complete 

 the parallel between comets and the corona by noting that the 

 tails of comets sometimes assume the curvilinear form found in 

 the sheaves of the corona. The dark parts which divide the 

 tails of comets have also their analogues in the rifts of the 

 corona. To push the comparison still further, it would be very 

 interesting to be able to prove that the corona, like cometary 

 masse.«, is transparent, and that bright stars can be seen 

 through it. Unfortunately it will be impossible to attempt this 

 experiment at the time of the next eclipse. 



An exact photometric study of the solar surface would per- 

 haps detect the transparency of the corona, indeed if we 

 suppose that the corona presents a certain opacity the parts of 

 the photosphere on which the large sheaves are projected must 

 be less luminous than the parts covered by the polar rays. 



If the corona is not subject to solar gravity it is scarcely pro- 

 bable that it shares the movement of rotation of the sun ; how- 

 ever, it would be useful to try in the coming eclipse to study 

 the question by the spectroscopic method, as M. Trouvelot 

 wished to do in 1883. It would be desirable to conduct all 

 spectroscopic observations^ of the corona by means of photo- 

 graphy. The instruments which must be used for this purpose 

 should be very luminous {i.e. give bright images), for there is 

 little light available, and the exposures are necessarily short. 

 In studying the effectiveness of a spectroscope in the case of an 

 object presenting a large apparent diameter, like the corona, it 

 is seen that the intensity of the spectrum depends entirely upon 

 the width of the slit, and the effectiveness of the object glass which 

 forms the image of the spectrum. As to the collimator and the 

 condenser their dimensions are of no importance, provided that 

 the collimator can well receive all the light of the condenser. 

 As the object glass which forms the image of the spectrum 

 must have an image long enough to givesutificient length to the 

 spectrum, one is led, in order to obtain great effectiveness, to 

 give this object glass a large aperture, and consequently to use a 

 prism of large size. 



The visibility of the bright lines depending not only on their 

 brightness, but also on their width, a wide slit must be employed 

 to obtain a good image of these lines ; on the other hand, a 

 narrow slit will give a spectrum of great purity, and will show 

 the dark lines. The employment of two different spectroscopes 

 is then plainly indicated. 



It remains for us to speak of the photometric measuring of 

 the corona by optical photometers. Bunsen's photometer has 

 already been used for this purpose, but I think that we must 

 henceforth turn to photography to obtain exact results. The 

 question should not be neglected, for it is certain that the 

 brilliancy of the corona varies considerably from one eclipse to 

 another. Thus Prof. Lockyer estimates that in 1878, at a 

 period of quiescence on the surface of the sun, the corona was 

 ten times less brilliant that in 1871. 



Let us end by pointing to the polariscope observations which 

 hitherto have been far from giving concordant results as to the 

 proportion of polarised light in the various parts of the corona. 

 Here also there are new inquiries to be made. 



Such, gentlemen, are the different problems suggested by the 

 study of the solar corona. We will hope that the next eclipse 

 will largely contribute to their solution. 



MEMORIAL OF SIR RICHARD OWEN. 



A 



MEETING was held at the rooms of the Royal Society, 

 on Saturday, to make preparations for the provision of a 

 suitable memorial of the late Sir Richard Owen. The Prince 

 of Wales took the chair, and was supported by the Duke of 

 Teck, the President, the Treasurer, and the Secretary of the 

 Royal Society, Lord Kelvin, Sir John Evans, and Professor 

 Michael Foster ; the President of the British Association, Sir 

 A. Geikie ; the President of the Royal College of Physician.<!, 

 Sir A. Clark ; the President of the Royal College of Surgeons, 

 Mr. T. Bryant ; the President of the Royal Academy, Sir F. 

 Leighton ; the Bishop of Rochester, the Dean of Westminster, 

 Lord Playfair, Prof, Huxley, Sir H. Roscoe, M.P., Sir F. 

 Abel, Sir F. Bramwell, Sir G. Stokes, Sir H. Acland, Sir 

 Joseph Lister, Mr. Ericsen, Dr. Priestley, Dr. Giinther, Dr, 

 H. Woodward, Dr. Maunde Thompson, Sir W. H. Flower, 

 Sir Erasmus Omroanney, Sir James Paget, Sir Henry Thomp- 

 son, Sir Spencer Wells, Sir Edwin Saunders, Sir John Fowler, 



NO. I 2 13, VOL. 47] 



