G. F. Barker—Spectrum of the Aurora of Oct. 1872. 88 
Scale Auroral Other 
Lines, number. Wave-length. lines. measurements, 
2 517 
6 
(4) band 182-135 505-499 505-499 {3% Barker. 
(5) band 138-142 493-485 493-485 {38 Bane 
86 . 
1415 4. 
(6) band 150-155 474-467 474-467 fi wmnce 
G 181 431 
(7) line 181 431 431 494 A. Clark, Jr. 
In this table, column 1 gives the lines observed, both Fraun- 
hofer and auroral; column 2, the corresponding number on the 
scale of the instrument; column 8, the wave-lengths caleu- 
new, yet that no previous observer has seen all of them at once, 
Vogel having seen five, and four having been seen by myse 
appear to coincide nearly with the solar lines FandG. But 
the want of a line corresponding to the C line, shows that these 
lines cannot be due to hydrogen. Moreover, the 8d band 
Includes the E line within it. On plotting the spectrum of 
made by Professor Pickering. The existence of the 
between 2 and 4 gi 
fixi 
auroral lines. It is certainly clear that if the identity between 
these lines and those of the air-spectrum—of course under mod- 
ified conditions—is to be established (as Professor Vogel thinks 
will be the case), this can only be done by absolute verification 
of lines by measurement. If the auroras at 
of temperature and pressure; conditions already abunda " 
shown to have a marked effect on the spectrum, gece 
“lent number of accurate line-measurements, therefore, it may 
