1887.] Total Solar Eclipse of August 29, 1886. 



317 



the Royal Astronomical Society. The results of his careful examina- 

 tion, and of the collation of this with other eclipse photographs, will 

 form the subject of a later communication. The spectroscopic cameras 

 mounted on the same equatorial framework failed to give any useful 

 result. 



The instrument used by myself was a 5^-inch equatorial, by Alvan 

 Clark, with a Rowland grating 14,488 lines to the inch, and the work 

 assigned by the Committee was an examination of the spectrum of 

 the inner corona immediately before and after totality, and a search for 

 the carbon bands during totality. I was assisted by Sub- Lieutenant 

 Helby, of H.M.S. " Sparrowhawk," who pointed the slit of the spec- 

 troscope, whilst my undivided attention was given to the bright lines 

 in the field of view of the small observing telescope. The object 

 glass cast a most perfect picture of the corona on the white enamelled 

 cap of the slit plate, so there could have been no difficulty in directing 

 the slit to any assigned position. 



In the search during totality for the carbon bands a. and /3, which 

 should have been well within the field of view, extending as it did 

 from wave-length 5600 to considerably beyond 6, the radial slit was 

 placed near the solar equator, at distances from the moon's limb vary- 

 ing from 0*1 to 0*5 of a solar diameter, and was then removed to one 

 of the sun's poles, and placed tangentially at successive distances as 

 before. In none of these positions could I detect the slightest trace 

 of the carbon bands. The conclusion indicated by these observations 

 seems to be that the vapour of carbon, if present in 1886, was not of 

 sufficient intensity to make an appreciable impression on the retina. 

 The dispersion was considerable, as I was using the second order of 

 spectrum with a power of about 4, but the fainter lines in the solar 

 spectrum, and the coronal lines when seen, were so distinct, that I 

 scarcely think the dispersion could have been excessive. 



The observations both before and after totality were greatly inter- 

 fered with by the clouds and heavy rain, but although rain fell within 

 a few minutes from the beginning, and also very soon after totality, 

 the sun seemed perfectly clear during the whole of the totality itself. 

 As soon as the sun broke through the clouds and became visible on 

 the slit of the spectroscope, Lieutenant Helby placed the slit at the 

 centre of the rapidly decreasing crescent, and the first line that I 

 detected was 1474 K, which extended to a distance of perhaps 8' 

 from the limb. Almost at the same instant I saw a mass of lines of 

 unequal length situated on the less refrangible side of b, but in close 

 proximity to it. Their number I estimated at about 15, but I could 

 form no idea of their relative intensities. This observation seems to 

 favour the view that the absorption producing the Fraunhofer lines 

 takes place in successive layers of the solar atmosphere, and not in 

 •any one layer exclusively. Within 20s. from the end of totality, the 



2 a 2 



