white, while the colour of that part of the envelope seen in north latitude, and which is in the rear of the sun's motion pointing towards the constellation Taurus, if projected on the ecliptic, is said to be a characteristic rose-red. This indicates a difference of constitution suggesting a spectroscopic examination. The vertex of this light appears to be projected by parallax on different parts of the ecliptic by the earth's motion in its orbit, thus appearing to prove that its shape is elongated, and not circular, for if circular, then it would subtend a similar arc all the year round. It regularly decreases in altitude in the evenings by an amount corresponding to the angular motion of the earth on its axis, and when seen in the mornings it increases in the same ratio, from the first appearance of the apex above the horizon to such a time as it ceases to be seen from its delicate illumination being overpowered by the solar glare, which fact appears quite sufficient to prove its extra-terrestrial origin. From the constancy of its appearance at the above seasons it does not appear to be of an intermittent nature, such as we might expect if produced electrically, or if it was of a similar nature to our auroras. Observations on zodiacal light, comets, and meteors, are fraught with renewed interest now that such stupendous commotions are known to occur near the sun's surface, especially after the connection apparently established between the highly attenuated material of Biela's comet with the meteoric display of 27th November, 1872; and it is to the meteorological changes of matter in the sun's neighbourhood through all its possible states from gaseous to vapourous—liquid and solid, and again from solid to liquid—vapourous and gaseous, that we may probably look as furnishing a clue to one of the most important problems in modern physics, namely, explaining the action of those laws of attraction to a centre, and of heat repulsion from a centre, which appears to characterise all cosmical aggregations of matter.

Art. XVI.—On a new Thermometer for Lecture Purposes. By A. W. Bickerton, F.C.S., Professor of Chemistry in Canterbury College. (With an Illustration.) [Read before the Philosophical Institute of Canterbury, 1st October, 1874.] A thermometer of simple construction that would show variations of temperature to a large number of persons at once, would be very useful in the lectures on heat and kindred subjects. It would be of especial value in schools, where but little time can be spent in preparing experiments. A thermoelectric pile and galvanometer is a most valuable piece of apparatus,