444 EEPORT— 1882. 



In January 1865/ I found that the light of the Great Nebula of Orion is re- 

 solved into the same set of bright lines. Further observations on the spectrum of 

 the nebula in Orion, and the spectra of other nebulae, were made subsequently with 

 more powerful apparatus.^ 



The result of these observations was to show that in the visible region, the 

 spectrum of the Great Nebula in Orion consists of four bright lines, the positions 

 of which are X 5005, 4957, 4861 , and 4340. The least refi-angible line appeared, 

 in the instrument used, to be coincident with the less refrangible component of 

 the brightest double line of the spectrum of nitrogen. The third and fourth lines 

 agreed in position with the lines /3 and y of the hydrogen spectrum, with which 

 they were compared directly in the spectroscope. 



During the last few years I have made several attempts to extend our know- 

 ledge of the spectra of the nebulae into the more refrangible region beyond the 

 reach of the eye by means of photography. On March 7 last (1882), I succeeded 

 in obtaining a photograph of the spectrum of the Great Nebula in Orion. 



The apparatus was the same with which I had photographed the spectra of 

 stars, aud the spectrum of comet h, 1881. It consists essentially of a spectroscope 

 furnished with a prism of Iceland spar, and lenses of quartz, placed so that the 

 slit shall be in the principal focus of a metallic speculum 18 inches in diameter, 

 and driven by an electrically controlled clock. 



A. gelatine plate was used very sensitive from about Fraunhofer's h of the 

 •spectrum, to a long distance in the ultra-violet. 



The exposure was limited by the coming up of clouds to about 45 minutes. 

 The opening of the slit was made wider than in my work on the stars. The slit 

 was kept upon a bright part of the nebula near the trapezium in the ' Fish's 

 moulli ' of the nebula. 



The photographic plate shows a spectrum of bright lines sufficiently distinct to 

 admit of measurement ; there is also present an exceedingly faint continuous spec- 

 trum. 



In my eye-observations of the visible region of the spectrum, I was nearly always 

 conscious of the presence of a faint continuous spectrum, but in no part of the 

 Debula did this continuous spectrum become strong relatively to the bright lines, 

 which were present throughout the whole extent of the nebula. 



The great range of the s]iectrum to which the plate was sensitive enabled me 

 to see upon it all the lines which had been previouslj^ observed with the eye. In 

 ■addition to these known lines the plate shows a relatively strong line in the ultra- 

 Tiolet region, which is new to us. 



As the slit was wide, this line appears broad on the plate, but for the same 

 reason so do also the known lines. Eye-observations have shown these lines to be 

 extremely narrow and defined at the edges. There is reason to believe that this 

 new line, though relatively intense, is similar in character to the other lines, 

 namely, narrow and defined. The great advantage of this control from lines 

 ■common to the photographic plate and to observations by eye, is seen further in 

 the following circumstance. The spaces occupied on the plate by the broad lines 

 ■do not appear to be quite imiform, but somewhat less strong in the middle ; but 

 the results of the eye-observations show conclusively, as might be suspected for 

 other reasons, that this appearance does not indicate duplicity of the lines, but 

 simply a want of absolute precision of focal adjustment. 



The broad character cin the plate of this new line does not permit of quite the 

 same accuracy of determination of position as would have been possible with a 

 narrower slit. There is little doubt, however, that this new line agrees in position 

 with the line ^ of the typical spectrum of the white stars.^ For this reason, this 

 typical spectrum has been placed in the diagram by the side of the photographic 

 spectrum of the nebula. The wave-length of the new line is therefore X 3730. 



There can be little doubt that this typical spectrum is due to hydrogen, and 



^ Proc. Roy. Sue. vol. xiv. p. 39. 



' Pltil. Tr'ajis. 1868, p. 540, ard Proc. Boy. Soc. vol. xx. p. 380. 



5 Phil. Trans. 1880, p. 672. 



