Wave-lengths, in hundred-thousandths of a millimetre.

It is unfortunate that in this illustration and in Professor Herschel’s paper the wave-lengths of the Aurora-lines are not given in figures, but must be roughly calculated from the scale. Professor Herschel speaks of Ångström’s drawing as representing a normal spectrum, and as derived from authentic sources, such as Vogel, Barker, and others; but beyond this we are not certain as to its origin.

In illustration of the difficulty of constructing any thing like a general typical Aurora-spectrum I append a Table of eight Auroral spectra taken at hazard:—

Table of compared Aurora.

Auroral lines and bands.

Observers.Red.Yellow.Green.Blue.Indigo.Violet.
Vogel, April 9, 18716297556953905233518950044694 to 4629
Barker, Nov. 9, 187162305620517050204820
Barker, Oct. 14, 1873630055505330 to 52005050 to 49904930 to 48504740 to 46704310
A. Clarke, junr., Oct. 24, 18705690532048504350
Backhouse, 1873606056605165501546254305
Backhouse, Feb. 4, 1874*557051804980483046404320
H. R. Procter, 1870*****
Lord Lindsay, 1870*****

* Mr. Procter’s and Lord Lindsay’s lines had no wave-lengths.

Ångström’s drawing discussed.

On examining Ångström’s diagram it certainly seems to me that, upon the showing of the drawing itself, the coincidences are not very exact. All three of the violet-pole bands appear to be less refrangible than the Aurora-lines with which they are compared-the middle one (at 47) considerably so, the one near E (at about 52·30) appreciably so, and the third (at 43) slightly so.