LOCKYER'S Conclusion that at sun-spot maximum most of the known lines in spots are replaced by lines of unknown origin.

(4) All of the silicon lines in the region studied are much weakened, while other substances have only a small proportion of weakened lines.

(5) Table III gives the measured wave-lengths of the faint ("band") lines photographed in the region A 5030-A 5215. Comparison with ROWLAND's table shows that these lines, which are considerably strengthened in spots, correspond with the extremely faint lines of the solar spectrum.

(6) A review of the literature indicates that these lines are the ones seen visually by YOUNG, DUNÉR, and others who have resolved the general absorption in spots, and that they account for most, if not all, of the so-called "bands."

(7) Our results confirm YOUNG'S conclusions that the absorption in sun-spots is due to gaseous matter, and not to solid or liquid particles.

(8) After discussing the views of EVERSHED and WILSON on the cause of the darkness of sun-spots, we conclude that it may be sufficiently well accounted for by absorption alone. GEORGE E. HALE AND WALTER S. ADAMS.

SOLAR OBSERVATORY.

COLOR OF THE SHADOWS OF JUPITER'S SATELLITES PROJECTED ON THE DISK OF THE PLANET.

On the evening of December 23, 1905, I was showing the visitors Jupiter through the large refractor. The seeing was good and the sky clear. The shadows of satellites I and III were on the disk of the planet. The shadow of I fell upon the dark-red equatorial belt of the planet near the meridian, just northwest of the great red spot; the shadow of III had just entered upon the disk very near the south pole of the planet.

It was at once seen that the two shadows were not equally dark. The shadow of III was a dense black, while that of I was not completely devoid of color. The character of the backgrounds was quite different, that upon which the shadow of I was projected being several shades darker than the white, cloudy region upon which the shadow of III fell. The effect of contrast would be, of course, to make the shadow which fell upon the brightest region (III) appear the darkest. The

shadows were, however, of appreciable size, so that it was possible to study them to some extent as surfaces and not as mere points. Then, too, the shadow of III was much larger than that of I, which should tend to reduce somewhat the effect of contrast. The difference of brightness in the backgrounds was carefully considered at the time of observation, and, after making what was considered an ample allowance for any such effect, the shadow of I appeared to have a decidedly more brownish color than the shadow of III.

Powers of 270 and 520 were used. The latter power, especially, gave the shadows sufficiently large disks for me to feel very certain of the color in that of I, and to feel equally certain that the blackness of III was real. The difficulty of properly interpreting such an observation is fully recognized.

It is seldom that the shadows of two of the satellites fall together upon favorable portions of the planet's surface for such an observation as the above. My reason for publishing this single observation is to call it to the attention of observers having the use of large telescopes, in the hope that they may take advantage of any opportunity to compare the density of the shadows of any of the satellites.

There can be little doubt of the absence of sensible atmosphere upon the Jovian satellites, and in that event any light in their shadows would have an important bearing on the physical condition of Jupiter. Such a condition, if established, would go far toward proving the high internal temperature of that planet and explaining the rosy color of the equatorial belts. C. D. PERRINE.

MT. HAMILTON, January 25, 1906.

THE SIXTH AND SEVENTH SATELLITES OF JUPITER AT THE OPPOSITION OF 1905-1906.

The sixth satellite was first observed, at the present opposition, on July 24th by Mr. ALBRECHT, Fellow in Astronomy at the Lick Observatory, with the Crossley reflector. He had the assistance of Mr. ELLIOT SMITH, also Fellow at the Observatory.

The satellite was then in position-angle 56°.6 and at a distance of 26'.0 from its primary. A comparison with Dr. Ross's ephemeris, printed in L. O. Bulletin No. 78, indicated a lengthening of his period to 251 days and small corrections.

to the other elements. This satellite reached west elongation about October 1, 1905, and at present is passing eastern elonga. tion. This is the fourth elongation reached since its discovery.

It is now known that this satellite, as well as the seventh, is revolving about Jupiter in the same direction as the other satellites of the Jovian system,-i. e. the motion is direct.

An approximate reduction of the observation of January 4th gives the following residuals from Dr. Ross's revised ephemeris in A. N. No. 4042:—

Position-angle (Obs. — Eph.)

Distance (Obs. - Eph.)

+0°.4 -o'.4

This is a close agreement, when the approximate nature of the ephemeris and observation is considered.

The seventh satellite was first reobserved on August 7th by Mr. ALBRECHT, using the Crossley reflector. Its positionangle was 289°.7 and distance 54'.6. A comparison of this position with Ross's ephemeris published in L. O. Bulletin No. 82 shows that the satellite was over a month in advance of its predicted place. On October 24th, however, it was only about a week in advance of its ephemeris place. This condition, taken in connection with the fact that western elongation was at a much greater distance than predicted, indicates that the eccentricity is much larger than 0.02, Dr. Ross's value.

Western elongation was reached about September 6th at a distance of over 61' (for distance unity of Jupiter). Observations are scattering for nearly two months past, owing to stormy weather, but those available indicate that eastern elongation was passed the latter part of December at a distance of only 43'. This would give o.18 as the minimum value of the eccentricity. As there are good reasons for believing that the major axis is considerably inclined to the normal to the line of sight, the eccentricity is probably larger than 0.18.

At the time of discovery, early in January, 1905, the seventh satellite had passed western elongation and was moving eastward. Since that time it has passed three elongations, one of which was not observed on account of the proximity of the Sun.

In the case of an orbit of such large eccentricity, it is not satisfactory to try to determine the period from the elongation times available. It is certain, however, that the period of the

seventh satellite will not differ greatly from that of the sixth, whose orbit is much better known at present.

The apparent orbit of the sixth satellite has opened out very much since discovery, while that of the seventh has closed up, so that during the present opposition the Earth is almost exactly in the plane of its orbit. These facts also prove conclusively that the motions of both satellites about their primary are direct. C. D. PERRINE.

MT. HAMILTON, January 25, 1906.

ORBIT OF THE SIXTH SATELLITE OF JUPiter.

On account of the accuracy with which this orbit is representing recent observations, it is thought advisable to reprint the elements for the benefit of A. S. P. readers. The orbit and ephemeris were derived by Dr. F. E. Ross, of the Carnegie Institution, Washington, D. C.

C. D. P.

ELEMENTS REFERRED TO THE EARTH'S EQUATOR.

Mean jovicentric right ascension...... 289°.I
Right ascension of perijove.

270

1905.0

COMING TOTAL ECLIPSES OF THE SUN.

A total eclipse of the Sun will occur on January 13, 1907. It will begin at sunrise at a point about midway between Moscow and the north end of the Caspian Sea. The shadow will pass over the northern end of the Caspian Sea, the southern end of the Aral Sea, midway between the cities of Samarkand and Taschkent, over eastern Turkestan, Mongolia, near the boundary between Transbaikal and Manchuria, and end at sunset on the mainland northwest of Saghalin. It would seem that the only location worth considering for an observingstation would be in the Samarkand-Taschkent region, longitude 68° E., latitude 40° N., where the duration of totality would be two minutes. The difficulties of transport for observers, instruments, and supplies would be practically prohibitive, especially in the middle of the winter season. The Lick Observatory will not send an expedition to observe this eclipse.

A total eclipse of the Sun will occur on January 3, 1908. Its path will cross the central Pacific Ocean, from west to east. The total phase begins at sunrise at longitude 155° E. and latitude 11° N. At local noon the shadow will be at 145° W., 12° S. The total phase will end at sunset at 85° E., 10° N., near Punta Arenas, Costa Rica, Central America. The shadow does not pass over any islands of considerable extent, but it is probable that several quite small islands lie within its borders, especially along the western half of the path. From Hydrographic Office Chart No. 1980 and other information kindly. supplied at my request by Superintendent O. H. TITTMANN, of the U. S. Coast and Geodetic Survey, it appears that Flint Island is well situated.

"Flint Island (British), discovered in 1801, lies in latitude. 11° 26' S., longitude 151° 48′ W., is 13 feet high, covered with brushwood and trees, and is visible from the masthead from a distance of 16 miles. It is about 21⁄2 miles long N. N. W. and S. S. E., half a mile wide, and is fringed by a steep coral reef, which dries at low water, and extends seaward generally about half a cable, but off the northern end of the island it extends seaward 41⁄2 cables and off the southern end E. S. S. 21⁄2 cables. In the interior are two small lagoons of brackish water. . . . There is little or no rise and fall of tide at Flint Island. The landing is very bad even for surfboats, but it is said to export nearly 200 tons of copra annually."Extract from Superintendent Tittmann's letter.