said to have been seen on the 13th of November, 1784; the second upon the 14th of November, 1789; the third in 1833 (Nov.?); the fourth on the 17th of November, 1848; and finally the fifth and sixth now recorded.

First aurora on the night of Aug. 28th-29th, 1859.-The first appearance of a reddish gleam was seen at 5 minutes past 9 in the evening, which rapidly rose exactly in the north and extended over the space embraced between the N.E. and N.W., reaching the height of Polaris about 23°. Some persons, it is said, saw it as early as 8h 45m. Its color grew brighter until 9h 30m, but from this time it faded to its total disappearance at 10h. A slightly luminous and whitish tint afterwards covered this part of the sky. But at 1 o'clock it reappeared, reaching again to Polaris. It attained its maximum brilliancy at 4h to 4h 10m its base being of a beautiful carmine red, from which rose divergent rays of a variable diameter, some fire-colored, others whitish, and rising to the zenith, the reddish tint covering a space of 180° from N.E. to N.W. At 4h 20m the aurora disappeared entirely.

A

Second Aurora on the night of the 1st-2d of September.-This second Aurora having been incomparably more brilliant, more extended, and more permanent than the first, it seems best to notice the details of its development with care, as points of comparison with observations in higher latitudes. This aurora was not visible before 12h 30m, and from that moment to 5h A. M. I followed all its changes. From 12h 30m to 12h 45m it spread towards the cast, and afterwards towards the west, then turning yet more towards the east with white rays which grew pale at the extreme west. From 12h to 1h after the white rays became extinct a portion of the east appeared of a beautiful fire-red. part of the west became also more flaming, and the summit of the arch, poorly defined, attained the height of Polaris, with a movement of translation toward the east. At 1 a brightness streamed from the north moving towards the N.N.E., defining by its light the outlines of cumulus clouds, of the horizon, of the sea, and the entrance of the fort. As this brilliancy increased and rose above the horizon its tints passed into light blue, involving the red portion at the northeast, and presently it began to fade out. The upper red segment rose considerably above Polaris. The illumination faded towards the northwest and embraced the whole of the auroral base; afterwards it rose again to the height of 12°. White rays with red and blue were then seen towards the west, which dilated longitudinally, oscillated laterally, were extinguished and resumed their brilliancy again by turns. The intensity of the illumination increased towards the east, and the red segment towards the west became more brilliant and more extended, until at the E.N.E. it

reached its maximum of brilliancy. At 1h 15m these rays were spread over the whole Aurora. The illumination attained to the E.N.E. in the space of three minutes, then it extended to the N.N.W. The east and still more the west then became very red. The illumination reappeared next at the east. The whole Aurora now became very red with rays to the north and west. This shade spread almost to the zenith. The fire-red of the west remained constant. The general depth of the Aurora faded while the whitish and reddish rays became more brilliant. But it was from 1h 30m to 3h 15m that the half hemisphere of the north from east to west was completely covered by a rich red tint, more orange than carmine, the gently arched summit of which passed the zenith towards the northeast, attaining the height of 100 degrees, accompanied with whitish rays and also with the red rays, more vivid then the general tones of the segment rising to the zenith, yet without passing it. At 2h the Aurora had attained its highest magnificence. The heavens then appeared stained with blood and in a state of complete conflagration. At a vast distance above the upper red segment appeared a second whitish segment which rose 23° above the horizon, while the upper red segment spread for 100° to the northeast and towards the constellation of Orion. The illumination whose different phases I have followed then constituted a white arch, the central and visible base of the Aurora above a bed of cumulus clouds which reached 8° above the horizon. At 2h 45m the two segments or arches of the Aurora declining toward the horizon, the lower white one first disappeared at 3h 15m A. M. From 3h 30m to 4h the general reddish tint disappeared and reappeared many times, but remained more intense towards the northeast. From 4h to 5h it gradually declined as the dawn commenced. At last the Aurora disappeared at 5h A. M. in the prolongation of the magnetic meridian where it made its first appearance. From 1h the west was constantly more flame-colored than the east.

These two Auroras have manifested the following peculiarities worthy of remark. 1st. The reappearance on the third night. 2d. Their magnificence: in height considerably more than 100°, in extent over 180°, their long continuance to day-dawn here under this latitude of 23°. 3d. The absence of an obscure lower segment although it might readily have been covered with the cumulous clouds which rose 8° above the horizon: above all, the expanse of the Aurora, a segment the extent of which has not been well established. 4th. The great height of 23° of the luminous segment or lower white arch visible only in the second Aurora. 5th. The rays or jets of light, some of which rose diverging from a point very far below the horizon, while others springing from the centre of the Aurora appeared to converge

slightly toward the zenith. Again they vanished for an instant to reappear over other points, some having a brilliant red, others a dense white mass with a feeble lateral pulsation and an alternate elongation and shortening. Sometimes the base of the rays was most brilliant and most deeply red colored, soon again the deepest and most brilliant color was on the upper extremities. 6th. The reiterated movement of translation of the whole aurora from east to west, followed by retrocession in an opposite direction, movements noted as being very rarely observed."

Space does not allow me to notice the concomitant phenomena which were produced, which from their importance will be the object of the next communication which I shall have the honor to address to the Academy. I enumerate the principal points observed. 1st. There was no noise in the aurora. 2d. The freely suspended needle of Marianini's Ré-Electrometer manifested not the slightest oscillation. 3d. The gold leaf electroscope of Bohnenberger gave no sign of atmospheric electricity. This neutrality of the magneto-electric force in the presence of so magnificent an Aurora Borealis is worthy of remark, for these two pieces of apparatus constructed by M. Ruhmkorff have great sensibility. 4th. There was no trace of polarization in the auroral light but very sensibly in its reflection upon the surface of the sea and upon the opposite clouds. 5th. It was perfectly calm. 6th. The temperature and barometric pressure were as usual. 7th. Two days after the Aurora the barometer rose from a half millimeter to one millimeter, following the height of the diurnal tide, and a northeast breeze set up.

9. Observation at San Francisco, California; by Dr. JOHN B. TRASK. (In a letter to the Editors, dated Sept. 1st, 1859.)

On the night of the 28th of August, at the hour of 10 o'clock, and continuing from that hour until near daylight we had for the first time in ten years in California a fine display of the Aurora. The sky was illuminated from the northwest to the northeast, with a flood of crimson light extending to the zenith, through which the whiter and yellow columns would start at varied intervals. It was a magnificent display and will compare favorably with the best varieties of your wintry months.

10. Height of the base of the Auroral curtain, Aug. 28.

The minimum altitude above the southern horizon of the lower margin of the meridional part of the auroral curtain, seen during the display of Aug. 28th, previous to 9 P. M., was determined independently by Prof. C. S. Lyman and by Mr. E. C. Herrick at New Haven, and by Prof. A. C. Twining at West Point, N. Y. These three determinations were made at about

the same absolute time (about 8h 40m New Haven time) and range from 10° 40' to 12°. Fortunately, à like observation was made at Philadelphia, Pa., (N. lat. 39° 57'), by Mr. Chas. J. Allen, where and at Burlington, N. J. the display was observed by Mr. Allen and by Messrs. Benj. V. Marsh and Samuel J. Gummere. Mr. Allen found this minimum altitude at Philadelphia to be about 221°. Assuming that the curtain was for a moderate distance parallel to the earth's surface, and that the observers saw the same curtain, it follows that the lower visible margin thereof was about forty miles above the earth. The probable error of this result seems to be quite small, yet it is highly desirable that the conclusion should be tested by observations taken at places between New Haven or West Point and Philadelphia and beyond, as far as Annapolis or Washington. The elevation of auroral belts observed in New England has been found to exceed one hundred miles, but the relation between auroral belts and streamers is little understood.

11. Appeal to Observers.

It is conceded that there is much connected with the auroral light which has not yet been fully explained, but it is unquestionably one of the most important of all meteorological phenomena, and its full explanation would probably bring with it the explanation of a large number of other phenomena, such as the origin and laws of atmospheric electricity, as well as of terrestrial magnetism. It is then of the highest importance to science that we should ascertain what the aurora is. The Aurora of Aug. 28th and following days affords a peculiarly favorable opportunity for deciding this question, and it is therefore important that this Aurora be thoroughly investigated. A thorough investigation of a single Aurora promises to do more for the promotion of science than an imperfect investigation of an indefinite number. It has been decided therefore to make a strenuous effort to investigate the laws of this auroral exhibition. For this purpose we need a careful collection of all the observed facts; and it is earnestly requested that every person who made accurate observations of the Aurora of Aug. 28th would communicate them to us for publication. This appeal is addressed to men of science in every part of North America where an Aurora was seen on the night of Aug. 28th. It is also addressed to observers on the ocean, and indeed throughout every portion of the globe, with the sole exception of Europe; for we assume that the appearances in Europe will be fully reported through the European journals. It is not improbable that this auroral exhibition may have been witnessed throughout the principal part of the northern hemisphere; and it is of great importance to know how far it did extend.

In order to render the communications of observers more definite and precise, we will briefly indicate the kind of information we desire.

We desire an accurate but concise description of all the phenomena with the exact time of their occurrence.

1. If a dark segment was seen resting either on the northern or southern horizon, or both of them, its altitude and position should be accurately stated.

2. If the streamers were seen to converge to a single point of the heavens, this point should be accurately located and the time of observation given.

3. If any single phenomenon (such as a detached luminous arch extending from the east to the west horizon) was so conspicuous as to be easily identified, it is important to have an accurate statement of its position and the altitude of its vertex, with the time of its formation and disappearance.

4. Was the Aurora seen in the southern half of the heavens, and how near the southern horizon did it extend?

5. Describe the color of the light, as well as its intensity.

6. If the Aurora exhibited any great variations of brilliancy it is important to know the times of least as well as the times of greatest brilliancy.

7. Did the Aurora exhibit any sudden flashes? Were there any pulsations like waves of light rushing up from the horizon?

8. If any observations were made showing the influence of the Aurora upon the magnetic needle, it is desirable that they should be communicated in detail.

9. The kind and degree of influence exerted upon telegraph wires.

10. Was any motion of translation observed in the Aurora, and if so, in what apparent direction and with what velocity?

It is proposed to publish in future numbers of this Journal, the most important part of whatever information may be ob tained as the result of this appeal; and it is intended to present the facts in such a form that each one will have all the materials which are necessary to conduct the investigation for himself. After all the facts have been communicated, it is proposed to present an analysis of the whole, with some speculations on the general subject of Auroras. Observers may forward their communications either to the "Editors of the Journal of Science, New Haven, Ct.," or to "Prof. Elias Loomis, New York City," who has consented to undertake the discussion of the phenomena.

Postscript.-Any exact data, relating to the remarkable auroral arch of April 29, 1859-mentioned by Mr. Herrick on p. 154 of this volume, will be very acceptable.