Sept. 25, 1884] 



NATURE 



521 



when the other telluric lines had almost entirely disappeared, he 

 saw B perfectly distinct. At a like zenith distance from the 

 sun this group seemed to him even blacker and more intense 

 than usual. Hence his inference that it could not proceed from 

 the vapour of water. This eminent physicist, a thoroughly con- 

 vinced partisan of spectral unity for the simple gases, had ascer- 

 tained that in the spectrum of atmospheric emission there was 

 no trace of any ray or band corresponding to B. It followed 

 for him that the air could not absorb radiations it was incapable 

 of emitting. If therefore this group, variable like the other 

 telluric rays, could be attributed neither to aqueous vapour, to 

 nitrogen, nor to the oxygen of the air, from what element did 

 it derive its origin ? Angstrom spoke of carbonic acid, but per- 

 haps without believing in it. He seemed deeply interested in 

 Tyndall's experiments on the absorbing power of gases by heat. 

 We know that this skilful investigator had found that the co- 

 efficients of absorption of oxygen and nitrogen by no means 

 corresponded with the coefficient of absorption of the air. The 

 difference he attributed to some unknown element sufficiently 

 rarefied to escape our analyses, and endowed with immense 

 absorbing power. Angstrom had probably this unknown ele- 

 ment in his mind, but he remained uncertain to the last. 



The great authority of the Swedish physicist could not fail to 

 have its influence on the judgment of those approaching the 

 question after him. Captain Abney, who has so greatly dis- 

 tinguished himself by his remarkable scientific labours, asserts 

 in Nature (October 12. 1S82, p. 585) that the groups A and B 

 cannot be regarded as telluric, but as proceeding from a medium 

 lying between the sun and the earth. Piazzi Smyth, who had 

 at first looked on B as telluric, seems to have finally adopted 

 Captain Abney's views, and is disposed to think that B as well as 

 A may after all be the product of some interplanetary medium. 

 In his opinion the recent theories of Siemens seem to confirm 

 his view of the case. 



The attention that I have for several years paid to the portion 

 of the solar spectrum stretching from A to b has naturally led 

 me to deal with this subject. Here is the method by which I 

 succeeded in separating and classifying the spectral lines. After 

 certain preliminary measures taken with the greatest care to 

 determine their exact position, each region of the spectrum is 

 drawn on two maps. The first is intended to reproduce the 

 appearance of this region when the sun is at 60° from the zenith, 

 the second when at 80°. These distances have been chosen in 

 such a way that in the latitude of Nice the observations may be 

 continued throughout the year. When the weather seems 

 favourable at the hour when the sun is in the desired position, 

 the intensity of each line i- marked on the map itself with all 

 possible exactness, the hygrometric state of the air being indi- 

 cated each time. The process is slow, delicate, and laborious, 

 but the result is certain. After I have thus made eight or ten 

 series of observations on each drawing, they are carefully exa- 

 mined, and the indications relative to any given line enable me 

 confidently to decide: — (1) whether it is not metallic; (2) 

 whether a telluric line belongs to a constant or to a variable at- 

 mospheric element. By this method I have been able to satisfy 

 myself that A, B, and o are telluric groups due to the constant 

 elements of the air. At the same distance from the zenith they 

 have always the same intensity. I refer of course only to the 

 main groups in each of them, and to the couples following them 

 on the least refrangible side. 



It remained to determine exactly to what atmospheric ele- 

 ment the groups in question were to be attributed. M. Egoroff, 

 Professor of Physics at the University of Warsaw, has recently 

 succeeded in solving the problem. He had for several years 

 ardently devoted himself to this inquiry, and in 1882 we jointly 

 made a series of experiments on the subject in the Paris Ob- 

 servatory. A pencil of electric light directed from Mount 

 Valerien on the Observatory, distant 10 km., gave us the spec- 

 trum of the telluric rays all but complete. There was no diffi- 

 culty in distinguishing A, B, and a, which are so easily recog- 

 nised. Capt. Abney has questioned the results obtained by us. 

 Vet they are incontestable, and in any case the experiment can 

 be easily repeated. 



At last, after these preliminary studies, M. Egoroff, operating 

 directly on oxygen closely compressed in a metallic tube, and 

 traversed lengthwise by a pencil of strong light, has obtained 

 the groups A and B. The thickness of the oxygen thus tra- 

 versed was doubtless insufficient for the production of a. But 

 however this be, it may now be confidently asserted that these 

 three, which are of such remarkable appearance, and which so 



closely resemble each other, have their origin in the absorption 

 due specially to atmospheric oxygen. 



I need not dwell upon the importance of this result ; but how 

 is it to be reconciled with the observations of Messrs. Janssen, 

 Angstrom and Piazzi Smyth? To judge from the sketch con- 

 tained in the Annates, and above referred to, M. Janssen must 

 have seen in the spectrum of aqueous vapour bands correspond- 

 ing to those of the solar spectrum in the region of B. One of 

 them even coincides exactly with the chief member of the group. 

 According to my own observations, to produce this effect the 

 vapour of water would have to yield at this point a non-resolvable 

 band, which would simply obscure the intervals between the lines, 

 as is seen in the spectrum I of my drawing. This observation 

 should then vary according to the hygrometric state of the air, 

 and not, as it has always seemed to me, according to the height 

 of the sun. Or else this band is not represented in the exact 

 position it ought to occupy, and should be shifted more to the 

 left, where in fact are found many lines of the vapour of water 

 constituting an important group (see plate). 



If, on the other hand, Angstrom saw B more intense at a 

 temperature of 27° C, it was doubtless owing to a simple effect 

 of contrast. The other telluric lines being greatly weakened, 

 those that retain their intensity must naturally appear blacker. 

 Such an effect is frequently produced in the course of my obser- 

 vations, and against it I have to be constantly on my guard. 

 And now how can we explain why the spectrum of absorption of 

 the oxygen differs so much from its spectrum of emission ? The 

 lack of sufficient data renders all explanation impossible ; but the 

 certainty of the fact obliges us to conclude that cold has not 

 the same properties as incandescent oxygen, and allows us to 

 suspect that it may be the same with all gases. 



In asserting that A and even B do not really vary in intensity 

 when the sun approaches the horizon, such an eminent observer 

 as Piazzi Smyth would have greatly surprised professional spec- 

 troScopists, were they not aware how difficult and delicate a 

 matter is the management of an apparatus of highly dispersive 

 power. Let but the luminous pencil be badly adjusted, the 

 prisms less than faultlessly regulated, the slightest cloudiness settle 

 on the surfaces, the imager, especially in the extreme red, 

 will at once appear as if drowned in the diffused light, which 

 obscures the most evident effects and even disfigures their essen- 

 tial characters. Strange phenomena are often produced, the 

 causes of which it seems impossible to discover, and which 

 easily give rise to illusions. But when we work under favour- 

 able conditions with a well-designed and well-constructed appa- 

 ratus, it becomes superabundantly evident that A and B vary 

 considerably in intensity according to the height of the sun, and 

 are certainly telluric. 



During the total eclipse of 1882, both M. Trepied and myself 

 fancied we observed on the edge of the lunar disk a notable 

 strengthening of the rays of the B group. If Captain Abney's 

 theory could have been confirmed, it would have certainly added 

 great weight to our observations, and for my own part I should 

 have felt highly satisfied at the result. Unfortunately, the at- 

 mosphere of oxygen which should now be attributed to the 

 moon in order to produce the observed effects, seems scarcely 

 reconcilable with the absence of refraction in the luminous rays 

 striking the edge of our satellite. I greatly fear the results 

 obtained in Egypt are one of those illusions, of which nearly all 

 spectroscopists have been more or less the victims. 



It would now be important to ascertain whether the nitrogen 

 and carbonic acid of the air may not be represented by any line 

 or any group in the solar spectrum. The study I am at present 

 engaged in, according to the above described method, will not 

 fail, I trust, to yield precise results on this important point. 

 Hitherto, apart from the oxygen groups, I have discovered no 

 line that may be confidently attributed to the constant elements 

 of the atmosphere. Hence it is desirable to await the result of 

 my researches before giving effect to the project adopted by M. 

 Bischoffsheim to establish on Mont Gross metallic tubes of con- 

 siderable length, in which the spectra of absorption of gases 

 may be studied on a grand scale. 



THE MIGRATIONS OF " SALMO SALAR" (L.) 



IN THE BALTIC 

 ""FHE following statement gives further details supplementary 

 ■*• to our recent article on this subject : — 



Since the earliest times salmon have been caught in the 

 Finnish rivers which had in their mouths or entrails hooks 



