March 27, 1879] 



NATURE 



495 



possessed sufficient explosive energy. No one supposes that the 

 volcanoes on the earth at present eject the fragments which will 

 constitute future meteorites, but it seems probable that the earth 

 may be now slowly gathering back in these quiet times the frag- 

 ments she ejected in an early stage of her history. 



Assuming, .therefore, that the meteorites have had a quasi- 

 volcanic origin on seme considerable celestial body, I am led to 

 agree with those who believe that most probably that body is the 

 earth. Robert S. Ball 



RECENT RESEARCHES ON ABSORPTION 

 SPECTRA 



TTHE numerous absorption spectra of soluble substances which 

 •*■ have been described hitherto, have referred as a rule 

 to the solutions of the substances, and but rarely to the solid 

 substances themselves. It is true that certain differences were 

 remarked between the spectra of certain solutions, those of ura- 

 nium and didymiam salts, for instance, and the spectra of the 

 solid salts ; yet, on the whole, these ditferences were so slight 

 that it was generally believed that the spectra were essentially 

 the same. On the other hand experiments had shown that the 

 spectra of solutions differed according to the dissolving medium ; 

 indeed Herr Kundt established the fact that the absorption 

 band of a substance in solution lies the nearer to the red end 

 of the spectrum the stronger the dispersion of the dissolving 

 medium. In these experiments the fact seeais to have been 

 overlooked that when changing the dissolving medium often the 

 whole character of the spectrum is changed, so that comparison 

 with the former one becomes extremely difficult. Close investi- 

 gation of these differences was therefore an important deside- 

 ratum, both for the theory of absorption spectra as well as for 

 practical absorption spectrum analysis. 



In the Monthly Report of the Berlin Academy of Sciences, 

 Herr Vogel has recently published the results of such investiga- 

 tions, to which he was led by the remarkable differences between 

 the spectra of solid and those of dissolved substances which he 

 had observed in the case of certain pigments. 



For the examination of these absorption spectra Herr Vogel 

 used instruments of but moderate dispersion, which allow of an 

 easier survey of the whole spectrum, and consequent judgment 

 of its general character, than is the case with strongly dispersing 

 spectroscopes. The absorption spectra of solid salts and pig- 

 ments were obtained fi-om thin layers of these substances, 

 prepared upon glass plates, through evaporation of a few drops 

 of solution. Herr Vogel reproduces the spectra he observed 

 on two plates, which at once show not only tht differences in the 

 spectra of one and the same solid substance and its solution, but 

 frequently an extraordinary coincidetue in the position of the 

 absorption bands belonging to totally different substances (for in- 

 stance, in nitrate of uranium and permanganate of potash). Of 

 several substances, such as iodine, hyponitric acid, and indigo, 

 the spectrum of the vapour is also given for comparison, and in 

 most cases the aqueous, alcoholic, and some other solution of 

 each substance has been examined. 



Without entering into the highly interesting details for which 

 we must refer our readers to the original paper, we confine 

 ourselves to stating the results of Herr Vogel's researches, which 

 are the following : I. Considerable differences exist between 

 tlie spectra which a substance gives in the solid, liquid, or dis- 

 solved and gaseous state. Characteristic bands which are shown 

 in the spectrum of one state are either not reproduced in that of 

 the other (this is the case with chrome alum, chloride of cobalt, 

 iodine, bromine, naphtaline red, fuchsine, indigo, cyanine, 

 aniline blue, methyl violet, eosine, carmine, purpurine, alizarine, 

 santaline), or they reappear in a different position, or different 

 intensity (exampl?~ : nitrate of uranium, permanganate of 

 potash, hyponitric .icid, alcanna red). Sulphate of copper and 

 chlorophyll show the same absorption both in the dissolved and 

 in the solid state. 



2. The spectra given by the same substance when dissolved in 

 different media are the same in some cases (purpurine in alcohol 

 or sulphide of carbon, aldehyde green in water or alcohol, 

 methyl violet and indigo-sulphuric acid in water or aniylic 

 alcohol) ; in other cases they differ only in the position of bands 

 (chloride of cobalt, fuchsine, coralline, eosine and iodine green 

 in aqueous or alcoholic solutions) ; and again in others their 

 character is totally different, so that no point of coincidence 

 remains (iodine in -sulphide of carbon or alcohol, naphtaline. 



aniline blue, purpurine, hsematoxyline, brasiline in water or 

 alcohol). 



•3. The rule established by Kundt, viz., that the absorption 

 bands of a body in solution lie the nearer towards the red end 

 of the spectrum the greater the dispersion of the dissolving 

 medium is in the region of the bands, is not confirmed in many 

 cases ; on the contrary, in some instances the absorption bands 

 move towards the blue in a solution of greater dispersion (nitrate 

 of uranium and blue chloride of cobalt in water and alcohol) ; 

 in other cases their position remains unaltered for various media 

 (hyponitric acid in air and benzol, indigo-sulphuric acid and 

 methyl violet in water and amylic alcohol, aldehyde green in 

 water and alcohol, purpurine in sulphide of carbon and alcohol). 

 In some cases a great difference in the sense of Kundt's rule 

 becomes apparent, while in others for the same spectral region 

 but a very trifling one appears, according to the nature of the 

 pigment (coralline and fuchsine). Indeed it happens sometimes 

 that certain bands are in the same position with different dis- 

 solving media, while others which are simultaneously visible are 

 displaced (nitrate of uranium in water and alcohol, oxide of 

 cobalt in glass and in water, protonitrate of uranium in neutral 

 solution and in a solution of oxalic acid, chlorophyll in alcohol 

 and ether). 



4. The position of absorption bands in the spectra of solid and 

 dissolved bodies may be only exceptionally deemed characteristic 

 for any certain body. Totally different bodies show absorption 

 bands in exactly the same position (solid nitrate of uranium and 

 permanganate of potash in the blue ; naphtaline red and coralline 

 in the yellow ; indigo, aniline blue, and cyanine in the orange ; 

 aldehyde green and malachite green in the orange). Closely 

 related substances sometimes show remarkable'differences in the 

 position of their bands under prefectly equal conditions (solid 

 uranium salts). 



5. The rule set up for absorption spectra, " each body has 

 its own spectrum," can be admitted only with great restrictions. 

 The great number of polychromatic substances show different 

 colours and different spectra in the solid state, according to the 

 direction in which they are observed. Most other bodies show 

 different spectra in the solid state from those of their solutions, 

 and in the latter case again different ones according to the dis- 

 solving medium, and the question arises which of all these spectra 

 is the body's " ortvn " spectrum. 



The most important difference of the spectra of elements in 

 a state of incandescent vapour, the position of the spectral 

 lines, ceases to be characteristic in the case of absorption spectra 

 of liquid and solid bodies. In the latter spectra, however, the 

 characteristic differences shown by the spectra of incandescent 

 vapours cannot be expected. It is known that metals, which 

 give such remarkably different spectra in the state of incan- 

 descent vapours, all give qualitatively the same spectnun as 

 incandescent liquids or solids, viz., a continuous one; for this 

 reason the absorption spectra of these bodies cannot show any 

 remarkable characteristic differences, whatever quantitative 

 differences may become apparent with regard to the absorbed 

 colours. If these well-known facts show that already with 

 regard to elements the laws appljnng to the spectra of gases do 

 not apply to those of liquids and solids, then Herr Vogel's 

 investigations prove that in the case of compound bodies simple 

 relations between the spectra of their different aggregate states 

 are still less frequent and occur only exceptionally. 



The analysis of absorption sf)ectra therefore is based not so 

 much upon the recognition of the position of the absorption 

 bands of a substance, as upon the changes in the spectra of the 

 same body which take place under the influence of various 

 dissolving media and reagents. Thus cyanine and aniline blue 

 dissolved in alcohol give a very similar spectrum, dissolved in 

 water a totally different one. The absorption bands of oxyhae- 

 moglobine disappear with reducing agents ; those of carmine, 

 which are in a similar position, do not ; the band of brasiline 

 disappears when acetic acid is added to the solution, that of 

 fuch-ine does not, &c., &c. 



The position of bands becomes more characteristic for the 

 recognition of a body, if the latter shows several absorption 

 band^ But even here we should go too far if from the accidental 

 coincidence in the position of bands of two different bodies we 

 were to draw conclusions regarding any similarity or chemical 

 identity between them (this has indeed been done in certain cases, 

 particularly with blood and chlorophyll). A conclusion regarding 

 such similarity or identity is only justified if the same bands 

 show equal intensities and analogous changes under the influence 

 of the same reagents. 



