194 



NATURE 



\yuly 1, i8So 



available ; the slit, though 4 feet distant, was made 

 capable of being adjusted from the observer's chair ; a 

 variety of prisms both simple and compound, with devia- 

 tions from o" to 45", and dispersions from o°-5 to 14° 

 between A and 11 solar, but all of large size, and capable 

 of being used in quick succession at pleasure, were 

 added ; with further arrangements for bringing into 

 central view and correct measure, many other natural 

 spectroscopic milestone lines, both with blowpipe-flame 

 and induction-spark. 



Thus far the instrument had been constructed, step by 

 step to a series of orders, chiefly by M. Salleron, of 24, 

 Rue Pavde au Marais, Paris, and it was ready in the 

 beginning of 1S77 for any aurora that should display 

 itself in the north-north-western parts of the sky ; but no 

 auroras came, nor have any appeared up to the present 

 time, February, iSSo. But the instrument has not been 

 idle. Its general material, wood, allowed it to be cut into 

 and altered for any experiment, educational or otherwise ; 

 Mr. Adam Hilger, of 192, Tottenham Court Road, furnished 

 it with a train of compound prisms raising its dispersion 

 powers to 33'' A to H, with improved Huyghenian rock- 

 crystal eye-pieces and a spectrum-illuminated pointer of 

 a remarkable kind for the purity of the colours successively 

 imparted ; until, though large parts of the apparatus were 

 still rough, it had become, on the whole, an essentially 

 safe instrument for spectroscoping numerically anything 

 within its powers to spectroscope at all, and for looking 

 into any such subject in a variety of different ways, and 

 to different degrees as to definition, illumination, dis- 

 persion, and magnifying ; thereby imparting considerable 

 confidence in its final results : and this is the chief 

 reason for saying so much at starting on the mere means 

 employed. 



Colours and Absorption Spectra. — The first series of 

 observations with this new instrument was of a very 

 simple kind as to the smallness of dispersion employed, 

 and on an often discussed subject, viz., the colours both 

 of the spectrum and of various coloured media, solid as 

 well as fluid. These observations were printed by the 

 Royal Society, Edinburgh, in vol. x.\viii. of their Transac- 

 tions {1Z7S), in a paper extending through sixty-four pages 

 and illustrated with three plates ; one of them containing 

 twenty-five dift'erent colours, viewed under seven diflercnt 

 gradations. Though much of the subject matter of this 

 papercouldonly be a confirmation, perhaps strengthening, 

 of many previous workings by others in the same direc- 

 tions; yet the following points, more or less new, were 

 also clearly established ; as — 



1. Colour bands, and bounding edges of coloured 

 regions in the spectrum, are not fixed in spectral place 

 as both Fraunhofcr lines and luminous lines of gases so 

 eminently are, but have a positive power of locomotion, 

 within certain limits, according to intensity of illumination 

 and depth of colouring matter. Witness especially the 

 march of the whole red band of light, with successively 

 increased depths of solution, over, and past, the black 

 Fraunhofer line, both found on this occasion in oxalate of 

 chromium and potash dissolved in water, and proved to be 

 as fi.xed as any other Fraunhofer line in all spectroscopy. 



2. Amongst colours the same to the eye, a physical 

 difference still more important than colour was ascertained 

 to exist, accordingly as their transmitted spectra formed, 

 cither one central beam, or two widely separated beams 

 in spectral place. So that one green glass exhibited 

 only the green region of the spectrum ; while another 

 glass, of different chemical coloration, but equally green 

 to the eye, shone chiefly in setting forth the nttra-red 

 regions of the spectrum at one end, and some of the blue 

 at the other, but extinguished strangely the yellow, 

 citron, green, and all that might have been expected 

 (7 priori to have been well rendered by it. 



3. Amongst the^e double-beam colours, of which cobalt- 



blue glass is an old example, well known from the times 

 of Sir David Brewster downwards, a far more powerful 

 case was met with in Judson's green dye of the aniline 

 series ; and by merely looking through a film of that 

 (without any prismatic or spectroscopic assistance) it was 

 shown to be possible to detect copper and arsenic greens 

 among vegetable green dyes in papers and muslins ; with 

 all the facility too, of seeing the former become blue 

 or black, while the latter became red, and sometimes 

 gloriously so. 



4. W'hile the green of vegetation was in every case, 

 both abroad and at home, together with its yellows, its 

 blue as in litmus dye, and some of its browns, turned into 

 crimson or scarlet— the green of shallow sea-water, as in 

 the mouth of the Tagus, and the deep blue of the ocean, 

 as in the Bay of Biscay, were both of them totally un- 

 affected ; but brown oars dipped in the act of rowing 

 into the former, in itself unimpressible, green water, 

 came up blood-red at every stroke ; and brown sea- 

 weed floating in the blue Bay, appeared of a richer 

 scarlet than any coral. These scenes too were all the 

 more brilliant and life-like to the observer, though looking 

 through something like a black ink-bottle, from the 

 tendency ascertained of two superposed films of any of 

 these double-beam colours, when differently illuminated 

 (the one looked at having to be more strongly illuminated 

 than the one looked through), to produce light, rather 

 than double dark, in and about the F region of the spec- 

 trum ; thus recalling a remarkable feature established by 

 the late esteemed Prof. Clerk Maxwell, in his researches 

 on colour-blindness. 



Rain-hand Spectroscopy. — The next subject on which 

 the experimenter published (both in the Journal of the 

 Meteorological Society of Scotland and in the fourteenth 

 volume of the " Edinburgh Astronomical Observations ") 

 was the power of the spectroscope to foretell rain. This 

 subject had been much studied by him already in various 

 countries and climates with pocket spectroscopes, but 

 assumed a far firmer character when their indications 

 could be tested by the spectroscopic machine above 

 described. 



Every spectroscopist knows how rich in black lines and 

 grey bands is the red-end of the spectrum of the sky ; 

 especially towards sunset, and near the horizon. M. 

 Angstrom had moreover already taught that some of 

 those lines or bands were due to watery vapour, and 

 others to dry gas, in the earth's atmosphere ; while M. 

 Janssen had minutely identified the components of the 

 former as being of such an origin, by comparing them 

 with the absorption lines in a long tube of high-pressure 

 steam. The Edinburgh experimenter therefore started 

 with much prepared to his hand, when seeking to obtain 

 a practical use for meteorology out of such observations ; 

 and his further steps were these : — 



1. He ascertained by many months of continued daily 

 experience that the lines attributed to watery vapour in 

 the spectrum of the sky, though formed by that vapour 

 when in the state of a transparent, invisible gas, increased 

 in their intensity of darkness, other things being the 

 same, according to the quantity of such vapour present in 

 the atmosphere. That quantity being independently 

 ascertained for the time by reference to wet and dry bulb 

 thermometers and the usual hygrometrical calculations. 



2. To keep those " other things the same," and prevent 

 the variations they are only too capable of setting up, from 

 interfering with the one phenomenon now sought after, 

 tlie experimenter confined his spectroscopic notings of the 

 sky's light to a constant, and that a low, altitude therein; 

 as well as to an hour giving a constant, and not a very 

 low altitude to the sun, and an azimuthal direction con- 

 siderably distant therefrom. Also to blue sky itself, as 

 seen through openings between clouds, if possible, rather 

 than to any cloud surface, and much rather than to 



