4o6 



NA TURE 



\AligUSt 2 2, 1889 



the middle of Nj (" 1 the Motions of sonna of the Nebulae 

 tovard. or from the Eirth," Roy. Soc. Proc, vol. xKii. (1874) p. 

 2^2). In Dicem'ier 1872, I comoarei this line directly with 

 TSTi, and found it su1fi:iently near in position to serve as a fidu:ial 

 line of comparison. 



Six other gx?eous n ibuloe were also eKamined, each on several 

 nights, w'.th the result that "in no instance was any change of 

 relative position of the nebular line and the lead line detected" 

 {ibid., p. 253). 



In the simultaneous observation of the nebular line and the 

 lead line, it was found if the lead line was made rather less 

 bright than the line of the nebula, the small excess of apparent 

 breadth of this latter line appeared to overlap the lead line to 

 a very small amount on jts less refrangible side, so that the more 

 refrangible sides of the two lines appeared to be in a straight 

 line across the spectrum. The clearness of position of the two 

 lines was shown by the observation that when the line of the 

 nebula passed across the field of the spectroscope, and the lead 

 line was thrown in, the lead line was not seen, but only an in- 

 crease in brightness of the nebular line. By comparing the 

 end of the nebular line near the Trapezium where it is refined to 

 a point, I estimated that the difference of position of the middle 

 of the lead line and that of the nebular line might be pos-ibly 

 from A oooo'2 to A. 0000*3 {ibid., p. 252). Some recent measures 

 of the position of the lead line with the middle of N^ show that 

 the lead line is about A oooo'i2 more refrangible. 



These direct comparisons of the nebular line with the lead 

 line confirmed, therefore, my former conclusion, that the brightest 

 line in the gaseous nebvdre is very near Nj, when seen under a 

 dispersion equal to nearly eight prisms of 60, namely 36° 25' 

 from A to H. 



This result is based on direct comparisons, on twenty-four 

 different nights, with Nj or with the line of lead. 



The wave-length of Nj has been determined by Kirchhoff, 

 Thalen, and by myself. Watts's reduction of my measure to 

 v^rave-lengths is clearly not accordant with my measures of air 

 tines immediately preceding and following this line. I have 

 therefore reduced my original measure to wave-lengths, and find 

 for Nj the value A 5004*5. 



Kirchhoff 5004-6 



Thalen S005T 



Thalen's value is clearly too high, as Thalen gives for the lead 

 line, coincident nearly with 'H-^, A 5004*6, and Nj is seen on the 

 more refrangible ^side of the solar iron line given by Angstrom 

 as A 5004*9. In Angstrom's map Nj is laid down on the more 

 refrangible side of the iron line 5004*9, at about 5od4'5. The 

 same position is givento N^^ in Kirchhoft's map. 



I have made a new determination of the position of N^, using 

 -the second spectrum of a grating 17,300 to ^the inch, relatively 

 to the solar iron line at 5004*9 according to Angstrom. 



The value came out A 5004*6, which agrees with Kirchhoff's 

 value, and with Thalen's measure ^of the lead line " hich falls 

 upon it, and also with the maps of Angstrom and of Kirchhoff. 



The wave-length of the brightest nebular line may therefore 

 be taken at from 



A 5004*6 to A 5004*8 (i) 



The micrometric measures of this line, given by D'Arrest, 

 "Vogel, and Copeland, agree closely with this value. 



D'Arrest's^ mean value 5004 



Vogel's ,, 5004 



Copeland's^ ,, 5004 



b. Second line. — In 1872 (Roy. Soc. Proc, vol. xx., 1872, 

 p. 385), I stated that I had found this line, by comparison with 

 a line of barium, and subsequently with an iron line, to have 



A wave-length of A 4957*0 , . (2) 



D' Arrest's mean value from micrometric 



measures A 4956*6 



Copeland's mean value from micrometric 



measures 4958*0 



c. Third lin:, — In my original paper (Pail. Trans., 1804, 

 p. 437), "On the Spectra of some of the Nebulae," in 1864, I 

 showed, by direct comparison with hydrogen, that this line is 

 undoubtedly the line of that gas at F of the solar spectrum 

 This observation was afterA'ards repeated (Phil. Trans., i85S, 

 p. 545), and has been con'irmedby the photographs of 18S2 and 

 1 388. 



I " Un^ersojaU'^r overde njbulose Syem^r " (Copsaha^err, 1372, p. 23). 

 " Monthly Notices R.A.S., vol. xlviii. p. 361. 



The wave-length of this line is therefore A 4860*7 . . (3) 



D' Arrest's value from measures 4860*6 



Copeland's ,, ,, 4861*0 



d. Fourth line. — In 1872 (Roy. Soc. Proc, vol. xx., 1872, 

 p. 385) I stated that I had satisfied myself of the coincidence of 

 this line with H7, but, on account of its faintness, it is very 

 satisfactory to find this observation of coincidence confirmed by 

 the photographs taken in 1888 and 1889. There can be no 

 doubt that this is a line of hydrogen, and that 



The wave-length therefore is ... 4340*1. .(4) 



Copeland's mean value 4342*0 



Dr. Copeland gives the measures of two still fainter lines which 

 he has seen in this nebula— namely, one at A 5874, possibly co- 

 incident with D3, and a line at A 4476 (see also Mr. Taylor, 

 Monthly Notices R.A.S., vol. xlix. p. 125). I defer the 

 consideration of these and other faint lines which I have often 

 suspected in the faint continuous spectrum of the nebula, as in 

 consequence of the great strain upon the eyes from my recent 

 direct comparison of the spectrum of the nebula with the 

 spectrum of burning magnesium, I was not able, during the 

 very few fine nights when Orion was favourably situated, to 

 undertake an examination for these very faint lines. 



Comparisons %vith the Magnesitim-flame Spectrum. 



In 1882, Dr. Copeland, in his paper on Schmidt's Nova 

 Cygni {Copertticus, vol. i. p. 109), remarked in a foot-note, 

 "that it is worthy of note that this line (A 5006*5 of burning 

 magnesium) almost absolutely coincides with the brightest line 

 in the plant tary nebula?." 



This line — namely, the bright edge of the first band in the 

 magnesium- flame spectrum — is very near in position to the 

 brightest nebular line. We have seen that the wave-length of 

 this line in the nebulae (i) is 5004*6 to 5004*8 ; now the wave- 

 length of the end of the magnesium- flame band is 5006*5, con- 

 sequently it does not coincide with the nebular line, but falls on 

 the less refrangible side at a distance of A 0002 nearly from that 

 line. 



The wave-length of the termination of the magnesium-flame 

 band is, as determined by — 



Lecoq de Boisbaudran 5006*0 



Watts 5006*5 



Liveirg and Dewar-^ 5006*4 



I have recently determ ined the position of the end of the band, 



by direct comparison with the solar iron line given by Angstrom 



at A 5006*58. 



My result places the magnesium-flame band line 



at A 5006*5 (5) 



In a paper read before the Royal Society in 1887,^ Mr. 

 Lockyer says : — " Only seven lines in all have been recorded up 

 to the present in the spectra of nebulae, three of which coincide 

 with lines in the spectrum of hydrogen, and three correspond to 

 lines in magnesium. The magnesium lines represented are the 

 ultra-violet low temperature line at 373, the line at 470, and the 

 remnant of the magnesium fluting at 500. the brightest part of 

 the spectrum at the temperature of the bunsen burner." At 

 p. 137 (Joe. cit.) Mr. Lockyer says: — "In the nebulae we deal 

 chiefly with lines seen in the spectrum of magnesium at the 

 lowest temperature." 



In a later paper in 1888 (" Suggestions on the Classification 

 of the various Species of Heavenly Bodies," Roy. Soc. Proc, 

 vol. xliv. p. 21) Mr. Lockyer states: — "In a paper communi- 

 cated to the Royal Society on November 15, 1887, I showed 

 that the nebulae are composed of sparse meteorites, the collisions 

 of which bring about a rise of temperature sufficient to render 

 luminous one of their chief constituents — magnesium. This 

 conclusion was arrived at from the facts that the chief nebular 

 lines are coincident in position with the fluting and lines visible 

 in the bunsen burner when magnesium is introduced, and that 

 the fluting is far brighter at that temperature than almost any 

 other spectral line or fluting of any element whatever." 



Although the number of direct comparisons which I had made 

 of the brightest line in the nebulae with Nj and with the lead 

 line, not to speak of the accordant results of the micrometric 

 measures of other observers, left little doubt in my mind that 

 this line could not be coincident with " the remnant of the mag- 



' Roy. Soc. Proc, vol. xliv., 1888, p. 245. 



^ "Researches on the Spectra of Meteorites: a Report to the Solar 

 Physics Committee." Communicated to the Royal Society at the request of 

 the Committee. Roy. ijoc. Proc, vol. xliii. p. 118. 



