1889.] chief Fluting seen in the Specti^um of Manganese. - 39 



have been sufficiently impressed by the modesty of Krafffc's statement 

 with respect to the accuracy of his observations. 



Although his attention was directed mainly to other observations, 

 and spectroscopic observations were of secondary importance, the 

 instruments were specially designed for the work, and it is fair to 

 assume that the measurements were made as carefully as possible. 

 Krafft's statement — " Leider gestatteten die obligatorischen Beobacht- 

 ungen nicht, den spectroscopischen Untersuchungen die gehorige 



Aufmerksamkeit angedeihen zu lassen Ich glaubte 



ausserdem diese Messungen um so mehr auslassen zu konnen, als der 

 Platz der gewohnlichen Nordlichtlinie oft und sehr genau bestimmt 

 ist " — does not, I think, necessarily imply that the observations 

 actually made were made carelessly. 



Further inquiry is, therefore, I think, necessary before we can 

 finally conclude that the suggestion that the chief line of the aurora 

 is the remnant of the manganese fluting at 558 is inadmissible, as 

 Dr. Huggins states it to be. 



That the spectrum of the aurora may be due to the integration of 

 two or more elementary spectra, as first suggested by Yogel in 1871, is 

 quite in accordance with the meteoric dust theory. As I stated in my 

 former paper,* " the aurora spectrum can be built up from the lowest 

 temperature spectra of manganese, magnesium, lead, and thallium, 

 and the brightest flutings of carbon." And not only this, but when a 

 line of any substance appears alone, it is generally the line which 

 first appears in its spectrum in the laboratory ; when two appear, the 

 second line is also generally the one which is added to the first by the 

 first increase of temperature. Any departure from this may be 

 ascribed to the incompleteness of the observations. The differences 

 between the spectra of auroras which are apparently at nearly equal 

 temperatures are in all probability due to slight variations in the 

 composition of the meteorites concerned. 



A low-temperature aurora will give a spectrum generally consisting 

 of the Mn fluting only, while at gradually increasing temperatures 

 various other lines will be added according to the volatilities of the 

 substances which they represent. 



[Note, April 30th. — The existence of a fluting near wave-length 558 

 in the spectra of all bodies which there is evidence to show are 

 swarms of meteorites, lends an additional interest to the appearance 

 of a line at about that wave-length in the spectrum of the aurora. 

 It is probable that the spectrum of the aurora has a meteoritic origin, 

 and the 558 fluting therefore might be expected. It is seen in the 

 aurora, in bright line stars, in new stars, in comets, and in swarms 

 of Group II. 



* ' Koy. Soc. Proc.,' vol. 45, p. 236. 



