into view. When this photograph is well illuminated, traces of the spiral structure, so apparent in the rest, are seen; but it would be difficult to recognise it without the other pictures. Passing on, the Moon's limb is on the edge of the faintly luminous stratum I have described in No. 1, which is, of course, now far more marked; and as we approach the south we see the prominences mentioned under No. 1 as being in the faint stratum, and then reach the upper part of the former terminal prominence, where the long exposure has now blotted out all traces of structure.'

Photograph 3. Exposed one minute six seconds later, and for ten seconds. It is evidently less affected by light than No. 1, and was probably taken through a light cloud. Here the great horn again claims attention; its spiral structure is now clearly made out. The outline of the Moon is wanting till we reach the south terminal protuberance of No. 1, which is now nearly hidden; and in its neighbourhood we again see the flare of which I spoke in describing No. 2.' Photograph 4. Exposed fifty-two seconds later, and for five seconds. 6 Nothing is seen but the great horn, but that picture is of singular beauty. The continually varying intensity of the light and its markedly spiral structure are the material points.'

Photograph 5. Exposed one minute three seconds

* The neighbourhood of both these prominences is marked by light, flaring into the corona in irregularly-curved lines,' adds Major Tennant, ' and this corresponds to the part of the corona which has been described by Captain Branfill and others as most bright.' This remark will be found important when we are upon the subject of the corona.

T

later, and for a short interval only.

The great horn is still conspicuous on the east side of the disc; and the outline of the Moon is marked round three-quarters of the circumference. The top of a prominence on the north-west side of the limb is seen,'

Photograph 6. This plate was exposed but for a short time, like the first and fifth, and one and a half minutes after the last. It is a picture as late as it was thought prudent to take it during the absence of the Sun. On the south-east side, where the corona was brightest, are traces of it; and the whole limb from north to south-west shows a trace of corona, and a series of beautiful protuberances, only one of which, however, is large. In the original it appears inflated. To me it resembles exactly a toy balloon in the shape of an animal. There is a mark as of an open mouth; the eye and the neck are marked, and the tail stands out stiffly behind. I mention these details because they complete the appearance of inflation by the resemblance they cause to what one has seen in that state. I would especially draw attention to the appearance as of a strong current of air (so to speak) blowing from north to south, and bending over and even detaching and carrying away the tops of prominences,'

It is well worthy of notice, also, that the horn still continues partly visible. It cannot, indeed, have been completely hidden even at the moment when the Sun was about to reappear.*

* I think this account would be incomplete were I not to quote the generous remarks with which Major Tennant closes this section of his

At other places along the track of the eclipse similar appearances were perceived; nor have we any

report. Throughout,' he says, 'I have spoken of myse.f as sharing in the photography. I certainly exposed and developed a few plates, and generally superintended all that went on, and so far I was the superintendent of this department; but the trouble of endeavouring to break in new and unskilled hands, and of carrying out the details, fell on Sergeant Phillips, to whose steadiness and determination to succeed, the result in this matter is mainly due. He was, moreover, most useful

distinct evidence pointing to a change of figure during the progress of the Moon's shadow from Aden, in the west, to Mantawaloc Kekee, the most easterly station where the phenomena of the eclipse were witnessed by competent observers. The accompanying picture (fig. 70) represents the eclipse as photographed at Aden about forty minutes before the shadow reached Guntoor. There is little in this picture to establish the occurrence of a change of figure in the great horn prominence. In comparing it with Major Tennant's views it must be remembered (in justice to the German photographers) that the atmospheric conditions were much less favourable at Aden, where the eclipse occurred very shortly after sunrise. In the next chapter a picture (fig. 84) of the eclipsed Sun as seen at the extreme easterly station, Mantawaloc Kekee, exhibits a tolerably close general accordance with Major Tennant's views. Other pictures and descriptions need not here be referred to, because they differ in no essential respects from those relating to previous eclipses.

Had the work done during this eclipse been limited to what I have hitherto described, the results, however interesting, would not have advanced in any important respects our knowledge of solar physics. We have now, however, to consider far more important results.

At nearly all the stations preparations had been

in other ways, for I could always rely on his care, and was thus enabled to avoid constant exposure to the Sun. Indirectly he has been the means therefore of facilitating my other work.'

made to observe the coloured prominences and the corona with the spectroscope. All the spectroscopic observers were successful in determining the general character of the prominence spectrum. I select the graphic and lively narrative of Lieut. (now Captain) Herschel, who commanded the expedition sent out by the Royal Society, and observed the eclipse at Jamkhandi. The week preceding the event,' he writes, had quite prepared me for disappointment. There seems to be an annual cloudy and rainy season at Jamkhandi, which lasts about a fortnight, and was said to be somewhat more marked than usual this year. The morning broke, however, as usual-clear, but the driving monsoon-clouds soon showed the kind of sky we were to expect. About a quarter of a minute before totality a third cloud obscured the Sun.' He had directed the spectroscope towards that part of the Sun which would be the last to disappear, and was watching the solar spectrum as it grew gradually narrower and narrower. It will be understood from what has been said towards the close of Chapter III. that the moment this spectrum disappeared the spectrum of the prominence matter (or failing the presence of such matter, the spectrum of the corona) would start into view. You may conceive,' he says, addressing Dr. Huggins, 'my state of nervous tension at this moment. Whatever the corona was competent to show must in a few seconds have been revealed; unless, indeed, it should happen that a prominence or sierra should be situated at that particular spot, in which case the double