to determine the sun's true direction in space by the apparent proper motions of the stars, for we may suppose a general movement of the stars in the sun's neighbourhood as drifting in a line parallel to the sun's equator (the most reasonable direction by analogy), then the direction of the Zodiacal Light would be brought nearer still to its observed direction. From modern researches in solar chemistry we are certain of the existence of the vapours of many metals, and also hydrogen, in the sun's atmosphere. Substances, therefore, of extreme tenuity exist in the vast laboratory of the solar orb. Portions of these substances, under the influence of heat repulsion, must exist at a considerable elevation above the surface, and when subjected to such commotions as have been actually observed (120 miles per second) would be transported to such a distance from the sun as to preclude their revolving around the sun in the same time; moreover, on account of their sudden translation from near the sun's surface to such an increased distance from the centre of diurnal rotation of the sun, some time must elapse before they acquire the additional velocity required. Such masses therefore lag somewhat behind in their daily rotation, and in consequence of the sun's proper motion accumulate in rear thereof. Other portions doubtless might become detached from time to time, forming comets with greatly elongated orbits, having their perihelion passages very close to, and in advance of, the sun's motion until perturbed by the planets. It becomes interesting to enquire whether the earth ever comes in contact with any portion of this matter, and if so in what part of its orbit? The illuminated medium known as Zodiacal Light has sometimes been observed reaching our zenith, proving that it extends at times to a distance from the sun fully equal to that of the earth; therefore, if its direction from the sun were truly on the plane of the ecliptic, then the earth must pass very near, if not actually through, its cone, and this at a certain fixed time aunually. In the diagram the cone is drawn on the ecliptic in Right Ascension 261° 29′. If the general direction of this cone extending from the sun were stationary, then the earth would pass very near, if not actually through, it early in December; but it must be borne in mind that the constant attraction of the earth for months too before it reaches this part of its orbit must hasten the time of contact. The November meteors appear to furnish convincing proof of such collision. They were observed in the year 472 (the sky appeared to be on fire over the city of Constantinople, with coruscations of flying meteors); next by the Moravian missionaries in Greenland, and by Humboldt in South America, in which the whole sky was filled with fiery particles, thick as hail, for four hours. Mr. Ellicot also observed these near the West India islands, when the whole heavens appeared as if illuminated