About 7 P.M. my attention was drawn to a silvery brightness in the north-east. Above, and still more to the east, was a bright cloud of light, which looked dense and misty, and gave one the impression of an illuminated fog-cloud. The edges were so bright that the adjacent sky, but for the stars shining in it, might, by contrast, have been taken for a dark storm-cloud. The light-cloud expanded upwards until its apex became conical, and then moved rapidly from east, along the northern horizon, until it reached the due west, where it rested, and formed for some time a luminous spot in the sky. About the same time a long low arch of light formed along the northern horizon, having a brighter patch at each extremity; and these being higher in the sky, the arch and turned-up ends were in shape like a Tartar bow. This bow was permanent; and later on a cloud of rose-coloured light formed in the east, looking like the reflection of a distant fire. From the bow also shot up curved streamers of silver light towards the zenith, which at one time threatened to form a corona. This, however, did not happen, and the Aurora gradually faded away, until, when the moon rose about 8, a silver tinge in the east alone remained. I should also mention that fleecy horizontal clouds of misty light floated in the north above the bow across the streamers. Mr. H. Taylor informed me he saw a similar Aurora some three weeks before, in which the bright horizontal light and short white streamers were the main characters. I am not sure that the horizontal light-clouds were not actual mist-clouds illuminated by reflection of the Aurora; not so, however, I think, the first-mentioned cloud, which had more the appearance of the _aura_ in the large end of an illuminated Geissler tube. [Sidenote: Spectrum of the Aurora described.] I examined the Aurora with a Browning direct-vision spectroscope, and found Ångström’s line quite bright, and by the side of it three faint and misty bands towards the blue end of the spectrum upon a faintly illuminated ground. I could also see at times a bright line beyond the bands towards the violet. There was not light enough to take any measurements of position of the lines. I made a pencil sketch of this Aurora, at the time when the light-cloud had moved W. and the arc formed, and of the spectrum. These drawings are reproduced on Plate VI. figs. 1 and 1_a_. [Illustration: Plate VI.] _Mr. Herbert Ingall’s Aurora, July 18th, 1874._ [Sidenote: Mr. Herbert Ingall’s Aurora, July 18, 1874. Haze canopy formed. Bright bluish flames appeared. Beams and streamers appeared. Oscillatory motion of rays.]