246 Intelligence and Miscellaneous Articles. 



easy to decide how to take the means. Mr. Dewar is now working 

 with substances chemically pure, and obtains much more constant 

 results. 



The unit employed is the electromotive force of an ordinary Da- 

 rnell's cell. The Grove's cells used in the electrolysis had (very 

 constantly) an electromotive force about 1*74 as great. 



I. Freshly -burned Platinum Plates. 

 Number of Grove's cells in 



decomposing battery 

 Resulting polarization .. 1*64 1-98 201 2*12 2-30 



II. Platinum +, Palladium — . 



Cells 1 2 4] 



Polarization 1*50 1'82 1*85 



III. Palladium-}-, Platinum—. 



Cells 1 2 4 



Polarization 1*60 1*92 191 (?) 



IV. With Three Cells. 



\ 1 



Platinum +, Iron 

 Polarization.. 2*16 



Platinum — , Iron -f. Iron plates. 



0-0 00 



V. Aluminium Plates. 



Cells 1 2 3 4 6 



Polarization . 1 -09 21 7 2*44 (?) 4-01 5*20 

 The last results are very remarkable, showing, as they do, from 

 aluminium electrodes a reverse electromotive force of more than five 

 Darnell's when six Grove's are in circuit. The polarization alters 

 so rapidly during the electrolysis (in the case of aluminium) that I 

 cannot be certain that the numbers above given represent fully the 

 maximum effect. Various other combinations have been tried, but 

 are being repeated by Mr. Dewar. — From the Proceedings of the 

 Royal Society of Edinburgh, Session 1868-69. 



SPECTRUM OF THE AURORA BOREALIS. 



o 



BY J. A. ANGSTROM. 



From the time when Franklin made his remarkable experiments 

 on lightning, to the present time, a complete parallelism has been 

 shown to exist between the actions of the forces of nature and those 

 of frictional electricity ; and hence it might have been expected 

 that the spectrum of lightning would be like the spectrum produced 

 by the ordinary electrical discharge. This has also been fully proved 

 by M. Kundt's observations. As, moreover, the aurora borealis and 

 terrestrial magnetism are so intimately connected that the occurrence 

 of the former phenomenon is always connected with disturbing ac- 

 tions on the magnetic needle, it might be assumed that the northern 

 light is nothing more than an electrical luminosity, such as is pro- 

 duced in the electrical egg in rarefied air. 



This, however, is not the case. In the winter of 1868-69 I was 

 several times able to observe the spectrum of the luminous arc which 

 surrounds the dark segment, and is never wanting in faint aurora?. 



