520 



NA TURE 



[Sept. 25, 1884 



CONSTITUTION AND ORIGIN OF THE GROUP 

 B OF THE SOLAR SPECTRUM "> 



VATHEN a single prism spectroscope is directed towards the 

 ** sun at the moment when it is in the neighbourhood of 

 the zenith, we perceive near C, at about a fourth of the dis- 

 tance separating it from the extreme red, a strong black line, 

 which Frauenhofer has named B. Under a more powerful in- 

 strument of five or six prisms this line becomes a very black 

 broad band, separated from the region of C by what may be 

 described as almost an empty space, the lines which do exist 

 in it being few and faint. On the other side this band is followed 

 by well-marked lines, which appear to be very regularly spaced, 

 and the first of which show some indications of being double. 

 Father Secchi had vainly attempted to resolve the band B ; on 

 this subject he writes : — 



"Certain bands which in ordinary instruments seem to be 

 stumped consist in reality of numerous perfectly distinct lines, as 

 is seen in a spectroscope possessing great dispersive power ; but 

 some of them are really massed towards the edge, and there it is 

 impossible to separate them, however powerful the instrument 

 employed. We may cite as an example the lines of the group B " 

 (Le Soleil, vol. i. ed. 1875, P- 2 35)- 



The statement of the learned physicist shows that, at the 



time he wrote the above, spectroscopic apparatus had not been 

 brought to very great perfection or power. In point of fact the 

 dispersion of eight or ten prisms suffices to show that the band 

 B is really formed of a large number of distinct lines. With my 

 highly-dispersive instrument it becomes resolved in a truly mar- 

 vellous manner. The seventeen lines composing it are distin- 

 guished with the greatest clearness, and may be very exactly 

 measured. Those following it on the red side are all broadly 

 doubled, presenting in their regularity a very remarkable ap- 

 pearance. 



In 187S Messrs. Piazzi-Smyth and Langley succeeded for the 

 first time and almost simultaneously in resolving this group. 

 Piazzi-Smyth, working with prisms, obtained only an incomplete 

 resolution, whereas, by means of Rutherford's excellent appli- 

 ances Langley not only separated all the lines, but also de- 

 termined their wave-lengths. Being unaware of the work of 

 these physicists, I fancied I had been the first to obtain these 

 results in 1879, when making the first essays with my highly- 

 dispersive apparatus. The sulphide of carbon compound 

 prisms, which M. Laurent had just made, were merely mounted 

 on a drawing-board, were badly regulated, badly sheltered from 

 variations of temperature, and could not yield the results that I 

 now obtain. The drawing which on that occasion T published 

 in the Comptes Rendus is incomplete and inaccurate. That 



S .l.ii' Spivlru 

 Position of the lines. 

 Mean aspect of region B when the ■ 



4. Non-telluric lii 



Spectrum of r 



-Region B, by M. L. Thollon. 



is So from the zenith. 



6o° ,, moist weather. 



6o° ,. dry ,, 



1 B as it would appear if observed outside the 



-phc 



which accompanies this article has been made with the greatest 

 care and to a scale on which the errors of position can scarcely 

 exceed i/iomm. It is more complete than any I have yet seen. 



The explanations given at the foot of my design enable us to 

 recognise at a glance : (1) the metallic lines, (2) the telluric lines 

 produced by the variable element of the atmosphere (beyond all 

 doubt aqueous vapour) ; (3) the telluric lines proceeding from the 

 constant elements (oxygen, nitrogen, carbonic acid). The group 

 is thus seen to consist of a pencil of seventeen rays, constituting 

 properly speaking Frauenhofer's B-band ; of a system of twelve 

 couples with intervals increasing regularly from right to left, 

 whereas the distance of the constituents diminishes with equal 

 regularity in the opposite direction ; lastly, of a somewhat im- 

 portant group belonging to the vapour of water, the whole inter- 

 spersed with a few weak metallic linos. 



When we survey the solar spectrum given by my apparatus, 

 beginning with the violet, and when we behold the thousands of 

 lines composing it distributed in all the regions without any 

 apparent kind of order, on arriving at the group B we feel, as 

 Mr. Langley remarks, the same impression as does a traveller 

 lost in a virgin forest when he suddenly finds himself in the 

 presence of a perfectly straight avenue of trees planted witli 

 mathematical regularity. It should be added that A at the 

 ' Paper by M. I. Thollon in Bulletin Astrommiquc, May 1884. 



extreme red, and a comprised between C and D are identically 

 constituted. If they attract less attention it is because A is 

 nearly at the limit of the field of vision, while the regularity of a 

 is disguised by a laige number of lines foreign to its origin. 

 Beyond these three cases chance alone seems to have presided 

 over the distribution of lines throughout the whole spectrum. 



But the interest r attaching to these groups is due not only to their 

 remarkable appearance, but also and especially to the question 

 of their origin. In what regions are produced the absorptions 

 which give them birth ? In the solar atmosphere, in the terres- 

 trial atmosphere, or in some medium comprised between the sun 

 and the earth ? Do these absorptions proceed from a common 

 element, and if so what is its nature ? The most contradictory 

 answers have been given to these various questions, which have 

 already long since been asked. 



To speak only of B, after his memorable experience of La 

 Vdlette in 1866, M. Janssen asserted that this group was at least 

 to a great extent due to aqueous vapour. The little sketch pub- 

 lished by him in the Annates de Chimie et de Physique (4th 

 series, vol. xxiv. p. 217) in fact shows, facing B, bands of 

 absorption due to the vapour of water, and corresponding 

 exactly to the spectral bands yielded by the setting sun in this 

 region. On the other hand, Angstrom, who had devoted much 

 study to the question, tells us that at the low temperature of 27 C. , 



