1st. Two Auroral spectra. I agree in this, but question whether the fainter lines may not possibly comprise more than one spectrum. 2nd. I agree also that the bright yellow-green line falls, as Professor Ångström describes, just behind the second line in the hydrocarbon yellow group (see Plate V. fig. 7). And I find, in common with the Professor, no well-marked or prominent line in the air-spectrum with which it accords. 3rd. This may be conveniently divided into two parts, viz.:— A. The proposition that “moisture in the region of the Aurora must be regarded as _nil_.” [Sidenote: Moisture probably not _nil_ in the Aurora region. Reasons for this given. Aurora in vapour or mist. Frequently near to earth’s surface.] Here I see reason to differ, since (to quote a letter of Mr. Procter’s) “the vapour-density of OH₂ is only 9 against 14 for N and 16 for O;” and again, “electrical or heat-repulsion may carry water-dust up to enormous heights.” There are, too, I think, circumstances connected with the Aurora itself which make the assumption of moisture being _nil_ in the Auroral regions untenable. The first of these is the fact that the white arc, streamers, and floating patches of light, found in some Auroræ, have frequently the peculiarly dense and solid look of vapour-clouds—a circumstance with which I have been frequently struck. Mr. Procter and others have also remarked that the Aurora is generally formed in a sort of “mist or imperfect vapour.” The second, that Auroræ, or portions of them, are frequently near to the earth’s surface. Instances of this are given in the section on the Height of the Aurora, notably the experiences of Sir W. Grove and Mr. W. Ladd. [Sidenote: Coincidence of Auroral lines with telluric solar lines.] On this point, too, note the peculiarities of the red line, which (and, as I find, the green line also) are coincident with, or very close to, telluric bands or groups of lines in the solar spectrum usually attributed to moisture. (See Plate XIII. fig. 2.) [Sidenote: Continuous spectrum.] I think we may also claim the continuous spectrum in the Aurora in further proof of water-vapour (see Plate XV. spectrum 7). The continuous spectrum of the Aurora is also, to my observation, more local and dense in the spectroscope than the glow generally seen between the lines or bands in gas-spectra. [Sidenote: Violet-pole spectrum discussed. Most spectra have a general as well as special character.]