In the eclipse of 16 April 1893 it was shown that the lines of hydrogen and helium, seen hitherto in the corona by a number of observers, must be attributed to a scattering of the promi nence light by the earth's atmosphere and pos sibly by the corona itself, but not to the intrinsic coronal light. An attempt was also made to determine the rotation of the corona by photographing its spectrum on both sides of the sun ; the H and K lines only were employed and a speed was found almost equal to that of the solar surface and corresponding to a dis tance of 20 minutes from the limb.
prominences not visible in full sunlight; that the prominences visible only during totality are white, especially in their more elevated parts, and have a special filamentous character, the threads being thin, long and blunt at the top; that the luminous intensity of the white prominences is feeble, and for this reason they are not visible to the naked eye, unless their height surpass that of the more brilliant parts of the corona; and that the other prominences visible in full sunlight appear much wider and higher in a total eclipse, and, when of a con siderable height, have their summits white.* The total eclipse of 1 Jan. 1889 was visible in a path that lay across California and Ne vada, and was successfully observed by many American astronomers. The photographs taken are notable alike for the fine detail of the inner parts of the corona registered, and for the im mense extent of the streamers imprinted. A particularly fine negative by Barnard permitted Holden to trace rays similar in typical char acter to the polar rays all around the limb of the sun and to present a detailed study of the filamentous character of the corona. An inter esting feature of several of the photographs consisted in their showing the outline of the moon's black disc projected on the corona be fore totality had begun.
From a study of his photographs of this The eclipse of 9 Aug. 1896 was visible in its total phase in Norway, Nova Zembla and northern Asia, while the path of the following eclipse, that of 22 Jan. 1898, lay across east cen tral Africa, India and the Chinese Empire. Each of these eclipses was successfully observed by a great number of astronomers and furnished a great wealth and variety of photographic material. In the second eclipse Campbell, of the Lick Observatory, employed an objective grating spectograph which was so arranged as to give a continuous photographic account of the changes in the solar spectrum at second and third contacts. Many interesting results of differences of level and of intensity of radiation in the lines were obtained. The photographs
obtained by Campbell with a lens of 40-feet focal length showed also a great wealth of de tail. One of two seconds' exposure clearly indicates the ultimate relation of the promi nences with the synclinal structure of the corona; another, of eight seconds, gives a fine representation of the polar rays and rifts with a coronal form already clearly influenced by the approaching sun-spot minimum.
The total solar eclipse of 28 May 1900 will always be famous for its long course through civilized territory, its path crossing the United States from Texas diagonally to Virginia, and finally threading its way across Spain. It is chiefly memorable for the fine weather which prevailed along the track of totality and for the magnificent array of optical and physical ap pliances employed in its observation. The chief American parties were those of the Lick, Yerkes, Allegheny, Smithsonian, Princeton, Brown, Harvard and United States Naval observatories, conducted respectively under the direction of Campbell, Hale, Wadsworth, Langley, Young, Upton, W. H. Pickering and S. J. Brown. Long focus telescopes either directly pointed at the sun or fed by ccelostats for coronal photography, huge Rowland grating spectrographs for photo graphing the flash and the delicate bolometric appliances of Hale and of Abbot characterized the instrumental attack. A great prominence was seen in the southwest quadrant, whose white summit was in eruptive activity. Striking changes in this prominence and in others were seen on comparing the American photographs with those taken at Ovar, Portugal and in Spain about two hours later. The study of the form of the corona as presented by the photographs was of special interest as a test of a °polariza tion of its streamers developed by Bige low some 10 years previously. According to this theory the coronal lines should lie in the direction of the lines of force surrounding a spherical magnet. Such magnetization of the sun, although suggested, was then not neces sarily assumed. The progress of physical science later made it probable that matter on the solar surface was electrically ionized after the man ner of the cathode streams of the Geissler tubes. And these electrified particles, or electrons, re quire then only to be directed by a supposed magnetic field of the sun. Bigelow's °Magnetic Theory of the Corona,° supported by Eberts' experiments, seems likely both from the natural assumptions made, and from the deductions allowable, to furnish the master-key to the mysteriously definite arrangement of the corona at the sun's minimum of activity.