VISUAL, GRATING AND GLASS-LENS, SOLAR SPECTRUM. 541 



Again, on trying in a similar manner with the Violet end — great G, and 

 some distance beyond it, were, with the better apparatus, seen better in Edin- 

 burgh in 1884. than at Lisbon in 1878. But every trace of light died out in 

 the Edinburgh spectrum long before arriving at great H or great K, though 

 they were grandiose spectral existences on the earlier occasion in Lisbon. 



Wherefore the whole result of this prismatic appendix to the Winchester 

 grating observations evidently is, that the Violet end, joined testimony with the 

 Red, in illustrating that there actually must have been just such a dulling of the 

 Solar Spectrum in 1884, as should arise from the upper air being at that 

 time over-charged with opaque, dusty particles — whether from the Krakatao 

 explosion, or any other source. 



Leaving that matter, however, of location of the dust's origin, to geologists 

 to pursue further, — I will beg leave to terminate this paper with a few words 

 on the subject of Spectrum Scales. 



Part VIII. — Testimony of successive Gaseous Groups to the 



MOST PRACTICAL OF NATURAL SPECTRUM SCALES. 



In 1878 the British Association for the Advancement of Science, published 

 in their Dublin volume, an admirably extensive exhibition (in 52 printed pages) 

 of the numbers for a Solar Spectrum, compiled from both M. Angstrom and 

 Professor Kirchoff, with the chemical origin of the chief lines, and the places 

 of all, given throughout in terms of " Oscillation frequency." This being, 

 however, in practice, only "Inverse Wave Lengths,"' or the "Wave-number" 

 here employed, though in terms of a French, instead of a British, standard of 

 of linear measure. 



In their volume for 1881 (at York) the Association has further "strongly 

 recommended " their method of " Oscillation-frequency," as against " Wave- 

 Lengths " ; and led its Members to expect a speedy publication of lists also of 

 the lines in Chemical elemental spectra, expressed in the terms they so much 

 approved of. They had also in 1878 promised to distribute to the Members, 

 at the Sheffield meeting in 1879, a map of the Solar Spectrum in terms of the 

 same " Oscillation Frequencies." This promise, however, I have just ascer- 

 tained from the Secretary, they never fulfilled ; and now in their volume for 

 1884, where the lists of metallic elemental lines are given at last, and to the 

 noble extent of 95 pages, — the Members, and the outside world too, will be 

 much amazed to find, that without a word of explanation or apology, the 

 places are all expressed in terms of Wave-Lengths. 



Such a breaking of its previous promises, inferences, and example, on 

 the part of a great Association, supposed by many to have necessarily more 

 continuity, and less vacillation in its opinions from year to year, than any 



