Celestial Chemistry. 401 



stages of the process of condensation, a few points of light 

 being visible in some where a little solid or liquid had probably 

 formed. This supposition, with the addition that, at some 

 distance from the surface the matter was sufficiently condensed 

 to prevent light from behind penetrating it, was then complete 

 as a cause of the uniform planetary light giving surface. 



Lord Rosse discovered an analogy between the annular 

 nebulae and the planetary forms, increasing the number of the 

 former from two to seven, and showing that a nebula with a 

 hollow centre, imperfectly seen in telescopes of less aperture, 

 might present the appearance of a planetary nebula ; but the 

 question remained one of extreme difficulty when Mr. Huggins, 

 in the autumn of this year, took up the subject, and attempted 

 to bring analysis by the prism to bear upon these remarkable 

 bodies, which seemed a class, sui generis, and of an order "en- 

 tirely different to the sun and fixed stars. 



He was now working alone, and in September, 1864, com- 

 municated to the Royal Society results of so important a cha- 

 racter that they have been at once printed as a supplement to 

 the paper of May last. The apparatus was the same that we 

 have described, Mr. Browning's excellent prisms being used; 

 the cylindrical lens was generally removed, since the nebulae 

 present a visible disc instead of a point when in focus. The 

 first object attacked was 37 H. IV. Draconis, which is described 

 by Sir J. Herschel in his latest catalogue as "very bright; 

 pretty small ; with a very small nucleus." Its colour is greenish 

 blue. Looked at with the telescope and spectrum apparatus, 

 Mr. Huggins was astonished to find no spectrum visible, but 

 only one short line of light in the direction which the dark 

 lines always occupy in the spectrum. At first he suspected 

 derangement of the apparatus, but this being found in good 

 order, it became apparent that this celestial body differed from 

 all others that had been examined, not in degree, but abso- 

 lutely in kind ; for its light was not composed of rays of dif- 

 ferent refrangibilities, but mostly of one monochromatic light 

 only. Careful examination with a narrower slit detected a 

 more refrangible, but much fainter bright line, and at about 

 three times the distance another still fainter was seen. The 

 direct comparison of these lines with those of the air spectrum 

 was then made, and it was found that the brightest line cor- 

 responded with the strongest line due to nitrogen, and the 

 faintest with one of the hydrogen lines, while the intermediate 

 one was nearly, but not quite, coincident with a barium line. 

 The appearance of this most remarkable result is given in the 

 woodcut, at least as far as such an illustration will represent 

 the different intensities. An excessively faint spectrum was 

 also detected on both sides of the bright lines, which is, doubt- 



