on Sunday, 16th, the shadow reached land again on the African coast near the River Gambia, about north latitude 15°. Here the eclipse was destined to receive a cordial welcome from the bands of astronomers who were ready to receive it. Sweeping onwards with a pace which had now begun again to accelerate, the shadow advanced into the interior of Africa, keeping below the parallel of 20°, and gradually curving southwards. At four o'clock on Sunday afternoon, the position from which totality was to be observed had gone to the east of the meridian of Greenwich. The end of the phenomenon was now rapidly approaching; the last glimpse that could be had of it from this earth would have been from the desert of Sahara, where, just at the moment of sunset, the phase of totality was reached. At a quarter past four, Greenwich time, the eclipse ceased to be total anywhere, but an hour longer had yet to elapse before the partial eclipse had vanished from the globe. It will, of course, be understood that at any particular locality the total eclipse only lasted for the time that the shadow occupied in passing over that locality. Thus the duration at any particular station was only about as many minutes as the hours during which the total eclipse required for its terrestrial journey.
It is plain that the best sites, so far as astronomical conditions are concerned, must be those where the duration of totality is as long as circumstances permit. To secure this, on the occasion now before us we had to occupy sites which lay as nearly as possible along the middle of a strip eighty miles wide, extending from the South Pacific to the middle of the Sahara. It fortunately happens that on this occasion those localities where the
