image on the screen, and this made the diameter of the Moon's disc about one foot. This image was then copied by hand, in crayon, with great care, by Mr. FREDERICK W. LIPPELT, a most conscientious draughtsman, whose skill in crayon work can hardly be excelled. Each negative was then in turn studied in the same way and under a magnifying glass, and every detail of the corona platted in, taking care to orient each plate properly. All of this work was subjected repeatedly to the criticism of the different members of the eclipse party, and, when completed, was artotyped by Mr. EDWARD BIERSTADT, of New York. This artotype production forms the frontispiece of the present number, and I beg to bring it to the attention of the Society as the definitive result obtained at this station for the form of the corona,. a result possessing substantially the accuracy of the original negatives.

In studying this reproduction it is to be noted that the lower edge of the picture corresponds accurately to a horizontal line which is inclined at an angle of 20° 12'.3 to a parallel of declination. This method of orienting the picture was chosen to avoid the necessity of placing reference lines upon the page.

STRUCTURE OF THE CORONA.

It has been stated that the negative of twenty-eight seconds exposure showed the coronal streamers to a distance of more than a degree from the Sun. Señor VALLE, at Norman, using a disc to cover the brighter inner corona, was able to trace the streamers with the naked eye to a distance of more than three degrees. The testimony, therefore, which these negatives are able to supply, does not apply to the fainter coronal streamers outside the smaller limit mentioned above.

The marked structural features of the corona, as presented by the negatives, are [a] the so-called filaments, and [b] the streamers extending approximately in the direction of the ecliptic.

The filaments extend over a region of twenty degrees or more on each side of the poles. They are straight lines of light arranged somewhat like the spines of a fan, and are not radial. The dark spaces between them are not entirely free of coronal matter, but can be traced in some cases to within a short distance of the Sun's limb. Comparing our negatives with a copy of a negative taken by the Lick Observatory party I find the number and arrangement of these filaments to coincide accurately. There

are slight differences in the lengths of some of the filaments, but no greater than might be accounted for by differences of exposure and atmospheric conditions.

The broad and strongly marked equatorial belt stretches directly across this mass of filaments, apparently cutting off the filaments at the somewhat irregular line of separation. The impression conveyed to the eye is that the equatorial stream of denser coronal matter extends across and through the filaments, simply obscuring them by its greater brightness. The effect to the eye is just as if the equatorial belt were superposed upon, or passed through, the filamentary structure. There is nothing in the photographs to prove that the filaments do not exist all round the sun.

The testimony from negatives of different lengths of exposure goes to show that the equatorial streamers are made up of numerous interlacing parts inclined at varying angles to the Sun's equator, but all trending, in a general way, along it, or, roughly speaking, along the ecliptic. The direction and character of these component streamers can be best studied at the edges of the photographs, where, on account of the smaller number shown, their direction and force can be made out. It seems probable that, could we have a faithful reproduction of the extreme outer corona, where individual streamers could be traced out for a considerable distance, our knowledge of the coronal structure would be materially increased.

VALUE OF THE PHOTOGRAPHIC TESTIMONY.

The corona, as projected on the photographic plates, may be described briefly as made up of bright streamers extending across the Sun's equator, surmounted at each pole by fainter coronal matter, traversed by non-radial rectilinear rays, or filaments. A physical and mathematical theory, which will account for this complexity of structure, becomes at once a matter of greatest inThe resemblance of the polar filaments to the lines of force about a magnet would naturally suggest a discussion upon the basis of a magnetically polarized Sun, did it seem possible to account physically for the existence of magnetic properties under the conditions which are known to exist on the Sun.

terest.

The problem of discussing the filaments as lines of force under the assumption of an electrically polarized Sun, is also open to the objection that it is difficult to see how a difference of electrical potential can be maintained upon the solar hemispheres. Such a

theory involves the assumption of some law of the electrification over the Sun's surface.

The force acting upon any point due to such electrification could then be expressed as a function of the co-ordinates of the point. Assuming such an equation of condition and comparing with the observed polar filaments considered as projected upon a plane perpendicular to the line of sight, it is possible to represent the observations equally well by numerous assumptions as to the distribution of the electrification. In fact, the polar filaments on the photographs are too short and uncertain of measurement to serve as the basis of verification of such a theory.

It does not seem easy to explain such an arrangement of matter as that shown in the photographs-assuming the coronal matter about the pole to be actually arranged in convergent lineson the theory of gravitation alone. The assumption that the socalled polar filaments shown on the photographs are proofs of the actual arrangement of the coronal matter in lines of force, does not seem to me altogether safe. Much of this apparent arrangement may be due to perspective and irregularities in the irradiation and reflection of light.

A phenomenon strikingly suggestive of the polar filaments may be seen by looking, from a distance of a few hundred feet, at an electric light shining through an atmosphere containing large quantities of dust and smoke. The effect of alternating bright and faint streamers is produced by small dust clouds of varying density. Of course, the effect is increased in this case by the fact that the light emanates from a point.

In the absence of a theory of the filaments as lines of force which would commend itself for strong physical reasons, it seems to me that all the testimony of the negatives may be reconciled with the idea that the corona, consisting of almost infinitesimal particles, is made up of matter circulating around the Sun, or gravitating toward it, not ejected from it. Such a view would harmonize with what is known of the zodiacal light, which, it seems to me, must be included in any satisfactory explanation of the corona.

The physical condition of matter in the corona must, in all probability, be similar to that of the matter composing the tails of comets, especially in cases of near approach to the Sun. A study of the behavior of comets in their approach to the ultimately throw light on the structure of the corona.

Sun will The tele

scopic observation of a comet very near the Sun, during a total eclipse, would form an exceptional opportunity.

It must, after all, be conceded that the value of the photographic testimony obtained during this eclipse belongs rather to the future than to the present, and that the negatives then obtained will be of increasing value as they come to be compared with negatives obtained during future eclipses.

SUGGESTIONS FOR THE FUTURE.

It would seem to be a matter of importance that original negatives, obtained during various eclipses should, as years go by, be placed in secure hands where they may be accessible to all astromomers. The National Observatory, which is soon to have a new building, provided with proper vaults, would seem to be the proper repository for such photographic records.

As a result of the study of the photographs of the January eclipse of 1889, the two following points seem to me to demand at present special attention from eclipse observers:

1. The desirability of obtaining photographs of the outer coronal streamers, whose delineation would doubtless go far toward a real knowledge of the structure of the corona. This involves the problem of designing the photographic outfit specially for this work.

2. The desirability of photographing the corona from points as widely separated as possible.

In 1871 the corona was photographed from points in India. and Java, separated by 2000 miles. The next total eclipse available for observation, occurring April 15-16, 1893, will present a most excellent opportunity for such observations. This eclipse begins in the South Pacific Ocean; is total on the west coast of South America at o" Gr. M. T.; is total on the east coast of South America at 1 Gr. M. T., and crossing the Atlantic the shadow cone reaches the west coast of Africa at 2h Gr. M. T. The shadow cone first touches the South American continent at Sarco, in Chile, latitude 29° south, and, crossing Chile, traverses the Argentine Republic, passing nearly 400 miles north of Cordoba, and touching the corner of Bolivia, passes directly through Brazil and reaches the Atlantic at Forteleza, in the Province of Ceara, in latitude 4° south. Then, crossing the Atlantic, the shadow cone will arrive one hour later at the west coast of Africa, entering it near Albreda, in Senegambia. It finally leaves the earth at a

point some 2500 miles in the interior, situated in the Great Sahara Desert.

At least one good observing station can be reached in Chile; a number are available in Brazil, and at least one in Africa. Between observations in Chile and those in Africa there would be an interval of over two hours. When it is remembered that the duration of totality will be over 41⁄2 minutes, it will be seen how favorable are the circumstances of this eclipse. Photographs of the corona taken by parties in Chile, Ceara and Senegambia would go far toward solving some of those problems of the corona which seem ripe for solution at the present time. To make the results strictly comparable the photographic equipments should be similar, and the negatives developed by the same operator, using the same method of development.

In the matter of eclipse observation, as in other departments of astronomical work, an intelligent co-operation among astronomers of different nationalities would go far toward increasing the value of the results. The elaboration of some feasible plan for securing co-operation in the observations of this important eclipse might, it seems to me, be very properly undertaken by this Society, which had its practical beginning in the co-operation of observers of the eclipse of January, 1889.

WASHINGTON UNIVERSITY, January, 1891.

LUNAR WORK FOR AMATEURS.

BY THOMAS GWYN ELGER, F. R. A. S.

As it may undoubtedly be assumed that a majority of those who join a Society like this are desirous of undertaking active work of some description, and that all who possess suitable telescopes are anxious so to employ them as to add something, however little, to the general sum of astronomical knowledge whatever branch of observation they wish to pursue, the following short paper on Lunar Observation may, perhaps, be acceptable to those who are thinking of turning their attention to this promising and attractive subject.

To the observer possessed of a moderate-sized telescope,— i. e., an achromatic of from 4-in. to 8-in. aperture, or a flector of from 6-in. to 12-in., the study of the Moon's surface

re