circular spot.

This is a frequently observed phenomenon, but the cause of its blackness, even in direct sunshine, is a profound mystery. Orbit of the Fifth Satellite of Jupiter.M. Tisserand reports the results of his researches on the orbit of the minute satellite of Jupiter, discovered Sept. 9, 1892, by Dr. Barnard, of the Lick Observatory. In his investigations he has used a circular orbit, a fixed elliptic orbit, and a variable elliptic orbit, the last method giving greatest satisfaction. The eccentricity of the orbit is very small, equaling about 0.01, making the ellipse almost a circle. Owing to the equatorial protuberance of Jupiter, the major axis makes a complete revolution in the astonishingly short period of five months.

Diameter of the Satellites of Jupiter.These satellites have been subjected to filar micrometer measures by Dr. Barnard, using the 36-inch glass with a power of 1,000. Unless the night were exquisite enough to permit the employment of this power and give distinctly defined disks, no measures were attempted. In every case the measurements were made by the method of double distances, 3 to 4 settings both before and after the reversal of the wires being employed. The great power and the large scale of the 36-inch equatorial render it very suitable for the determination of very small quantities, as, for instance, the diameters of the larger asteroids, the satellites of Jupiter, etc., always using the full aperture of the telescope:

The "Red Spot" of Jupiter.-Between November, 1894, and March, 1895, as observed at Greenwich, this object was carefully examined whenever visible. It was never easily or distinctly seen. No color was noticed in it, it being simply an elliptical outline. Best views of it were usually had before and after it had passed the central meridian.

Saturn. In 1894, during the opposition of Saturn, Dr. Barnard made a long series of observations, extending from February to July, with the 36-inch refractor of the Lick Observatory, of this planet, its rings, and its satellites. They confirm in a remarkable degree the prior measures of Prof. Asaph Hall with the 26-inch refractor of the Naval Observatory, Washing ton, D. C., from 1884 to 1887. These harmonizing observations of the two astronomers would indicate that no change has occurred in the Saturnian system since the first systematic measures were taken, and negative the assertion once made by a distinguished astronomer, that the rings were approaching the planet and would, in a few years, coalesce with it. One conclusion arrived at was that, contrary to some former assertion, the center of the planet was in the optical center of the ring.

The black and white spots alleged to have been seen on the planet by small telescopes were never found either with the 12-inch or the 36inch telescope, though carefully sought for, and he is confident they do not exist. The outer edge of the inner crape ring appeared to unite with the inner edge of the middle bright ring. No spots or markings of any sort were seen on

the crape ring, nor was its inner edge serrated, as some users of small instruments have claimed.

The following are the micrometrical equatorial measurements of the planet, the result of twelve nights of observation, from March 25 to June 25, 1894, reduced to mean distance, equaling 9.538861, Earth's distance =1.00. Corrected for phase:

66

66

Equatorial diameter in arc = 17-744" 76,150 miles. Polar =16-807" 69,980 66 Polar compression = 6 170 miles, or, in simplicity, 6,170 miles shorter in diameter from pole to pole than through his equator, while the Earth's polar axis is but 26 miles shorter than its equatorial diameter.

Saturn's Rings. The announcement by Prof. J. M. Keeler, of Allegheny Observatory, Pennsylvania, that by the spectroscope he had obtained proof positive that the rings of Saturn are composed of countless millions of minute satellites (a fact long suspected) has awakened renewed and widespread interest in the unique system of this wonderful world. When it became known that its rings were multiple, then arose the question of what are they constituted, and are they solid or liquid, or formed of discrete particles, analogous to the tails of comets or the rings of meteors which surround the Sun? The theory held by many astronomers was that, if solid, they could not remain intact, as the great centrifugal force experienced would tear them asunder, even if of the strength of steel. This spectroscopic evidence, which is indisputable, that they are neither gaseous, liquid, nor solid, must put to rest all speculation, Prof. Keeler having demonstrated the correctness of the conclusions previously reached by many astron

omers.

If the spectroscope be pointed to the Sun's center, which has no motion to or from us in the line of sight, the lines will occupy the same positions as those produced in the chemist's laboratory by raising to incandescence the substances producing the solar lines. When, however, it is aimed, say, at the upper limb of the Sun, which by his rotation is approaching us, the lines are moved toward the violet end of the spectrum, but if, on the other hand, it be directed to the Sun's lower limb, which is moving from us, the same lines are deflected toward the red end of the spectrum. Now, if the ring of Saturn be a solid, the outer edge, being larger than the inner, should move the faster, but if it be composed of minute satellites the inner ones, being nearer the planet, will move more swiftly, for the same reason that Mercury's orbital motion is faster than that of Venus, and that of Venus faster than that of the Earth, etc. This simple reasoning explains the nature of Prof. Keeler's discovery, which is, that the inner edge of the ring moves faster than its outer. By well-known spectroscopic processes Prof. Keeler has ascertained that the inner edge must move at the rate of 13.06 miles per second, and its outer edge 10.65 miles.

During the same opposition of Saturn in which Prof. Keeler achieved his wonderful discovery Dr. Barnard made a prolonged series of micrometrical measurements, of which the annexed

table is the result:

In comparing these elements, which agree quite closely, the greatest difference will be found in longitude of node and perihelion distance. Notwithstanding the slight apparent difference of orbit, the following diagram will show their agreement, and that the comet is identical with that of De Vico. is almost without doubt. If

this be so, then the latter comet during the fifty years it was lost must have been subjected to some greatly perturbing influence besides the Sun, which could not have been other than that of the giant planet Jupiter, near which it passed in 1885, when the approach was so close (see diagram) that both objects, for two years not only moved in the same direction, but also in nearly parallel paths. This long-continued attractive power of Jupiter was sufficient to change the orbit of De Vico's comet into that of Comet Swift.

90°

Parabolic.

Orbit

1885.0

1885.0

1897.0

Comet 1894 II (Barnard).-This comet, whose period-five and a half years-is just expiring, has not of a surety been detected, though sought for with great assiduity, particularly at the Lick and Lowe Observatories. At its return in 1890, as was expected, it eluded observation also. On the morning of June 30, Dr. Swift, of the Lowe Observatory, observed not very far from the comet's ephemeris place a very faint cometary-looking object which he took for a nebula, as Sir William Herschel had several near. Returning to it on July 4, he found the body missing, and came to the conclusion that it was the much desired Barnard comet of which he had had a view, but which long-continued effort failed to refind. Owing

Jan. 20

An inspection of the drawing will show that the long-lost comet's orbit runs also close to that of the planet Mars, and that in the distant future, when the comet itself shall travel thither, it must encounter another disturbing force and be again diverted into a new orbit. Its next near approach to Jupiter will occur in 1897, when it must again be greatly perturbed or, perchance, again lost.

1885.6

1886.0

of

Mars

Brorsen's Comet. This periodic comet, which was discovered in 1846 and has a period of 5.8 years, is now due, but, because of its proximity to the Sun, it has not yet been detected, and very probably it may not be seen at all. As it was not found at its last apparition, though it was favorably situated for visibility, it may have shared the fate of the disintegrated comet of Biela.

270°

1887.0

to the position of its orbit the comet can not be again seen until 1906, its detection in 1901 being improbable. The approximate place of the suspected comet on June 30 was right ascension 11 20m 45; declination north 2° 55'.

Comet 1895 a (Swift).-On the morning of Aug. 20, Dr. Lewis Swift, Director of the Lowe Observatory, discovered a very faint but rather large comet in right ascension 0h 27m 40°; declination north 5° 30'. It had a very slow motion in a northeasterly direction. Following are its elements computed by Prof. Lewis Boss:

Dr. Berberich's determination of its period is 7.06 years. This comet proves to be another of the Jupiter family of comets, which now numbers 20. They are so called because of the change of original orbit by the attraction of that_great_planet's mass as they journeyed to the Sun. Save Halley's, all the known periodic comets move direct, and all but that one belong to the Jupiter group.

Fay's Periodic Comet.-This comet was detected on Sept. 26, 1895, at the observatory of Nice, France, in right ascension 21h 8m 11.5; declination south I° 54' 11". Though it has passed the point nearest to the Earth, it will not arrive at perihelion until March, 1896, and therefore it ought, in strictness, to be included in the list of comets of that year.

Total Eclipses of the Moon.-On March 10, 1895, there occurred a total eclipse of the Moon visible from both continents. During the various stages of its progress it exhibited phenomena of great interest. In coloration, the density of the shadow, and the semiobscuration of the Moon during totality it bore a great resemblance to its last return, also total, in 1877. Several attempts to photograph-with both short and long exposures the Moon during the total phase, proved abortive. One exposure of an entire minute photographed only one faint, neighboring star. One conspicuous feature of the eclipse was the extraordinary brightness of Aristarchus, for which its general high albedo seemed hardly to

account.

Eclipse of Sept. 3 and 4, 1895.-This total eclipse of the Moon was a return of that of Aug. 23, 1877. In general appearance to the naked eye it was very similar to that of March 10, 1895, but with the telescope several marked differences were observable. Aristarchus, which then glowed like a diamond, attracting universal attention, was very faint and inconspicuous. At the Lowe Observatory, a phenomenon never before observed by the writer-but visible to many on this occasion-was seen; the upper portion of the Moon was of a pale but decided blue color, its upper boundary convex, agreeing exactly with the convex arc of the Moon's limb, and the chord perfectly straight-not concave like the new Moon's narrow crescent. The length of the versed sine was about of the Moon's diameter. This feature was nicely observable with field and opera glass and with the 31-inch finder of the telescope, but was less distinct, though easily seen, in the great telescope itself.

The chief value to astronomy of a total lunar eclipse is the determination of the times of occultation of stars by the lunar disk during totality. Both limbs of the Moon being then similarly illuminated, by observation of disappearances and reappearances of stars occulted the ascertainment of the Moon's diameter freed from the effects of irradiation is made possible. Also, from diminution of the Moon's light, much fainter stars may then be seen near the disk than

at other times. At Greenwich Observatory 137 observations of disappearances or reappearances were recorded by the eleven observers who watched the progress of the eclipse. Of this number of observations 124 were pronounced good. In observing an occultation, the time could be estimated to about the tenth of a second.

The ruddiness of the Moon when totally immersed in the Earth's shadow can, of course, be understood, but no satisfactory reason has yet been given for the great variation in color and brilliancy in different eclipses. Even when happening under similar circumstances, there is still wide difference in the amount of luminosity of the eclipsed Moon.

The faintness of the Moon during totality is much underrated. Prof. W. W. Pickering, in the total eclipse of 1887, estimated the eclipsed Moon to be as much fainter than the uneclipsed as the latter was fainter than the Sun. This must be an extreme view, however, as the best authorities make the Sun 700,000 times as bright as the Moon.

With regard to the Moon's spectrum very little has been accomplished. The atmospheric bands seemed so intense and broad that they practically ran one into another, and observers simply got the two ends of the spectrum cut off.

Variable Stars-Algol.-Dr. S. C. Chandler explains the periodic variations in the intervals between its minima by supposing that the bright star with its eclipsing dark companion revolves around a distant center of gravity, determined by its relation to another dark body, during a period of 130-91 years. M. Tisserand considers that they are caused by changes in the line of apsides due to a polar compression of Algol. This hypothesis requires considerable variation in the duration of the minima; Dr. Chandler's, that there should be a periodic inequality in the proper motion of Algol. In "Astronomical Journal," No. 343, Prof. Lewis Boss, of the Dudley Observatory, at Albany, N. Y., gives a list of observed positions of Algol and of 13 comparison stars for 1895 0, for the determination of the truth or falsity of Chandler's hypothesis. correct, the apparent orbital motion of Algol is now little less than at its maximum, and it will so continue for nearly twenty-five years, within which time it would be possible to truly determine the question.

If

The variability of Z Herculis was discovered by Dr. Chandler, in July, 1894, who regarded it as a variable of the Algol type, with a period of 34 23h 50m. Its variability was detected by Hartwig also, who assigned it a period of only 1d 23h 55m 40. As, however, a minimum on Sept. 20 did not occur at the time indicated by either of these periods, Prof. Dunér concludes that the star is of the V Cygni type with unequally bright components, and that the faint and very bright alternate in periods of forty-seven and forty-nine hours. The hypothesis demands that Z Herculis consists of 2 stars of equal size, one of which is twice as bright as the other; that they revolve around their center of gravity in an elliptical orbit whose semiaxis major is six times the diameter of the stars; that the plane of the orbit passes through the Sun; that the eccentricity is 0.2475; and that the line of apsides is inclined

at an angle of 4 degrees to the line of sight. In this orbit the stars revolve in 3d 23h 48m 30. Evidently Z Herculis forms the hitherto missing link between stars of the Algol and V Cygni types. The star for 1900-0 is in right ascension 17h 54m; declination +15° 8·7'.

S Corona.-When nearing minimum the nebula surrounding this star has again been seen. On Nov. 15 this variable was a stellar point in the center of a faint nebula, but on the 25th the stellar point had vanished and only the nebula remained. Unfortunately the star was then too near the Sun for its reappearance to be observed. The star for 1900-0 is in right ascension 15h 17m 19; declination + 31° 43'6'.

R Cygni.-At minimum, the nebula about this star was well seen. The star was invisible on Dec. 30, 1893. On Jan. 23, 1894, a nebulosity was suspected in its place. On Feb. 13 the very faint, diffuse, bluish nebula was certainly seen, as also on the 21st. The nebula on March 6 was very faint, with a suspicion of a condensed center, suggesting the reappearance of the star, and on March 24 the star was a minute but welldefined point, of the 12-1 magnitude, without any nebulosity. This is the first instance in which these curious changes could be watched in their entirety, and were made by Cuthbert E. Peek at the Rousdon Observatory, England. The star for 1900-0 is in right ascension 19h 34m 8; declination + 49° 58'5'.

An examination of the photographs of stellar spectra taken at Arequipa, Peru, forming part of the Henry Draper Memorial, has resulted in the discovery of 11 new variables with spectra of type III with the hydrogen lines bright.

Gould's "Astronomical Journal" for the current year has much valuable variable-star literature. In No. 347 is a revised supplement to his second catalogue of variables by S. C. Chandler. No. 339 contains an ephemeris of variables of the Algol type by P. S. Yendell. Catalogues, notes, etc., of variable stars of both short and long period, by H. M. Parkhurst, E. F. Sawyer, and others are found in Nos. 338, 346, and 350, to which the reader is referred.

Nova Auriga. This, says Dr. Barnard, is still visible as a small star without change in physical appearance since 1892. Comparison with 2 neighboring stars shows conclusively that there has been no perceptible motion in the Nova for two years. This fixity is surprising because of the enormous velocity in the line of sight assigned to the star by spectroscopic observation. Its spectrum is that of the nebula. Its place for 19000 is right ascension 5h 25m 35; declination +30° 21' 27". Its present magnitude is 9.7. It has been named T Auriga. In appearance it is a faint stellar point involved in a somewhat dense nebula.

Similar to the Auriga Nova in its spectrum characteristics is Nova Normæ, the place of which for 1900-0 is right ascension 16h 22m 12; declination - 50° 18'. This too, as also Nova Cygni, gives the nebula spectrum, and all have the same life history.

Photographic Stars.-Prof. J. M. Schaeberle, of the Lick Observatory, has been comparing the magnitudes of the faintest stars visible on Dr. Max Wolf's photographic plates with the faintest visual magnitudes in the 36-inch

telescope at Mount Hamilton. Taking the sky in the region of Algol, and using one of Dr. Wolf's enlarged silver prints in comparison, he was surprised to find that, with an exposure of five hours, the photograph revealed stars down to the 165 magnitude. As the photograph was taken at the level of the sea and with a small telescope, and as a star of the seventeenth magnitude is at the limit of vision of the 36-inch glass, Prof. Schaeberle concludes that Dr. Wolf, with his relatively small telescope, can, with an exposure of five hours, photographically chart stars down to the seventeenth magnitude, or those as small as the giant refractor of the Lick Observatory will show visually.

Faint_Star near Alpha Centauri.-Mr. Walter F. Gale, of Paddington, New South Wales, has noticed an eleventh-magnitude star, about 70" preceding our nearest stellar neighbor, Alpha Centauri, at position angle 272°. He is not aware of any previous observation of the star, which is not now visible, owing to its proximity to the bright star. If unconnected with the system, its distance twenty years hence, because of the proper motion of Alpha, will be very small. As the angle and distance do not oppose such an hypothesis, the remote possibility of the faint star being connected with Alpha, making it a triple system, renders it of special interest. He thinks that the best investigations of the orbit of Alpha Centauri have been made by Profs. Gill, Elkin, and See, though their determinations fail to satisfy the most recent measures, the companion being a few degrees in advance of the computed place.

Spectra of Stars.-The annual report of Prof. E. C. Pickering, Director of Harvard College Observatory, ending Sept. 30, 1894, contains a digest of much valuable work done with the varied instruments at his command. The Henry Draper Memorial is fruitful in good works, 1,657 spectra having been photographed with the 8-inch Draper telescope, and 1,708 with the 8-inch Bache glass in Peru. All the plates have been examined by Mrs. Fleming, resulting in the discovery of eleven variable stars whose spectra show the hydrogen lines. In addition to the above, 912 photographs have been taken with the 11-inch Draper telescope. Of these, 59 out of the 340 photo-spectrographs of Zeta Ursa Majoris and 47 out of 65 of Beta Auriga have been found periodically to have double spectral lines. These stars are termed photospectroscopic binaries, and the investigation of their duplex character is far beyond the reach of any visual telescope. Their periods of revolution are four and eight days respectively.

Nebulous Region in Orion.-Dr. Roberts's splendid photographs of the great nebula in Orion, so much talked of in late years, have not by any means exhausted the wonders of that remarkable region. Dr. Barnard also has secured not only photographs of the giant nebula itself, but of the space surrounding it, indeed, of the entire constellation, using for the purpose not a telescope, but a camera of his own construction, with the object glass of a magic lantern giving a field 30 degrees in diameter, about one half of which was flat. His negatives portray an enormous curved nebulosity encircling the belt and the great nebula covering a large portion of