T)40 C PIAZZI SMYTH ON THE 



ascertained to be in the very middle of a blank region of M. Fievez's 

 Solar Spectrum map, — why this is, in so far, just the kind of result that would 

 be given either by the Krakatao volcanic explosion having caused a transmis- 

 sion of some new and strange gas to the upper regions of the atmosphere, — 

 or by something still more extraordinary having happened in the Sun. And 

 yet I do hope M. Fievez will forgive me, if I am more inclined at present to 

 attribute the ominous-looking blank in his work to the imjDerfections necessarily 

 inherent in any single-drawn spectrum map (which is by its nature positive on 

 everything, but may be mistaken on anything likewise), than to any new gas 

 having appeared within the last two years in either the Earth's atmosphere or 

 the Sun's surrounding. 



But when Ave come to the question of a possible recent increase in opaque 

 dust effects, or a general dulling of the whole Solar and Telluric spectrum, — 

 such indications in the Winchester work, and equally in all three of its 

 spectra, are vast and undeniable. These dust effects, however, are more easily 

 recognised towards each end of the Spectrum ; for there, the " continuous " 

 light, elsewhere in blinding excess, fades away into utter darkness, and 

 increases thereby the sensitiveness of the photometric scale. 



Compared then with what I was enabled to observe of the Solar Spectrum 

 in Portugal in the years 1877-78, — each Winchester spectrum is deficient at the 

 Red-end by the whole of what precedes great A ; and deficient at the Violet-end 

 by all that follows little " h," including therefore those notable spectrum mile- 

 stones — so grand when the air is pure and clear, viz., great H, and great K. 



But the Winchester Spectrum was observed on a grating, while prisms were 

 used at Lisbon ; and some gratings are very limited in the length of spectrum 

 they are capable of reflecting at any time. I proceeded therefore, on returning 

 to Edinburgh last July, to arrange a prismatic Solar spectroscope very like that 

 employed in Portugal, though furnished with a stronger preliminary condenser 

 for the Solar rays, and armed with very transparent simple glass prisms, in 

 place of rather dark compound ones, — so as to make up, in Edinburgh, for the 

 want of the clearer air, and brighter Sun, of Lisbon. 



With the Edinburgh arrangement then in 1884, and during many trials there 

 in August of that year, — I could not only see the middle of the spectrum, but 

 also as far towards the Red as great B, rather better than I used to do at 

 Lisbon in 1877-78. But great A, further towards the Red-end, not so well; 

 Brewster's Z still further that way and always faint in the middle of the day, 

 not at all ; his great Y line and accompanying bands, by no means so well ; and 

 his great X line, witli its distant companions, not in the least degree, — though 

 they were abundantly clear at the southern station in 1877. 



Hence an extra-dulling of the Red end of the spectrum in 1884 is established 

 with much certainty. 



