before others, but was observed upon the whole for about half an hour.
At first the phenomenon seemed inexplicable, but after a little consideration, was referred (and as it a peared from after observations correctly) to an effect of aërial) perspective. The rays which seemed to originate from a common centre on the east, were really only the intervals between long shadows caused by the occurrence of clouds far to the west, and were in fact passing to the place from whence they seemed to originate, and the circumstances of the case seem to have been as follow:—the atmosphere contained a slight haze, which allowed the sun's beams to pass forward with but little interruption, but was yet in sufficient quantity to reflect a considerable portion of light to the eye. The sun was just setting;— clouds, very far to the west and out of sight from the place where the observer stood, stopped the light wherever they interfered, and cast immense horizontal or nearly horizontal shadows along the sky, parallel to each other, and over the head of the observer. The difference between these shadows and the intervening illuminated parts, could not be observed over head or on the right or left hand, i. e. perpendicular to their direction, because of the want of sufficient depth, as it were, in the parts thus circumstanced, to make them visible; but as they receded from the observer in the direction from the sun, they became fore-shortened, and then, from the greater depth of mass, and consequently greater number of particles looked at, became visible. This is at least one reason why they were so visible towards the east; but another is the probable existence of more haze in that direction than towards the west, or to the right or left of the observer's situation: the rays could not be seen between the sun and the observer, though the sun was out of sight, and consequently the general light, it may be supposed, not too great; which seems to imply that less haze existed in that direction; and its presence was fully proved towards the east by the dull red colour which the moon assumed upon rising a short time after the appearance had ceased. The convergence of the rays to one spot, and that opposite the sun, was merely an effect of perspective, and requires no explanation here. (See Plate I. fig. 6.)
Although the appearance on this evening was exceedingly
