The spectroscopic observations were mani fold and in the main successful. Professor Evershed, observing in Algiers, just outside of the path of totality, obtained photographs of the flash spectrum in high solar latitudes, and concluded that, in its main features, it is the same as in low latitudes. He verified the essen tial identity of the flash spectra as photographed by Shackleton in 1896 and those obtained in 1898 and 1900, and inferred that °the flash spec trum is probably as constant a feature of the solar surface as in the Fraunhofer In discussing the detailed spectroscopic results of the flash layer, Evershed infers that the ab normal intensities of the enhanced lines charac teristic of all levels and all latitudes of the flash is due to a continuous circulation of the solar gases in a radial direction, while the cooler, more diffused gases, in their subsidence, deter mine the character of the absorption spectrum. His final conclusion is °that the flash spectrum represents the emission of both ascending and descending gases, while the Fraunhofer spec trum represents the absorption of the descending gases only.° Professor Deslandres, by means of spectrographs with optical parts of Iceland spar and quartz, photographed the entire ultra-violet spectrum of the reversing layer from A 4000 to 3000, the entire ultra-violet spectrum of the upper chromosphere and the entire spectrum of the corona with two new corona! rings. Des landres, observing visually with a powerful grating, inferred from the inclination of the green coronal line on the east side of the equa tor that the corona appeared to have a more rapid rotation than the disc. Professor Turner secured photographs, in polarized light, of the corona, permitting of a quantitative investiga tion. The polarization of the corona was radial and was "not only shown in the outer streamers, but right up to the moon's limb.) An important bolometric measure of the heat of the corona was•conducted by Abbot and Mendenhall, of the Smithsonian Observatory, which led to the con clusion that the corona neither reflects much solar light nor in the main gives light of its own on account of high temperature, abut seems rather to be giving light in a manner not asso ciated with a high temperature, or at least with the preponderance of infra-,red rays usual in the spectra of hot bodies?) This result plainly points to the electrical character of the coronal light and is also in harmony with Bigelow's magnetic theory. Deslandres, using as very sensitive Mellon pile> for detecting the heat of the corona, reports positive indications which are in strange contradiction with Abbott's re sults and intensify the regret at the accident which befell Hale's bolometric appliances at the critical moment. The heat measures in future eclipses will have a great interest in relation to corona! theory.
The total eclipse of 18 May 1901 traced its path across southern Madagascar, the Indian Ocean, Sumatra, Borneo, the Celebes and New Guinea, and, on account of its long duration of over six minutes and the high altitude of the sun at the best points of observation, attracted expeditions from many countries. The general weather conditions unfortunately proved to be unfavorable; at a few of the eclipse stations, however, observations of great value were made. Professor Perrine, observing at Padang, Sumatra, through a partly clear sky, obtained excellent results. His photographs showed °conspicuous series of coronal hoods,) and also remarkable coronal disturbance resembling an inverted cone of considerable angle) and unlike anything previously recorded. This disturbance associated with a long thread-like prominence, Professor Perrine, subsequently, by means of the Greenwich solar negatives taken at Debra Dun, India, traced to a sun-spot and associated faculm, near the solar limb on the day of the eclipse, and thus plainly exhibited faculm prom inences, coronal disturbance, and sun-spots in a common origin and intimate relation. This
wonderful phenomenon was without doubt due to a solar eruption which took place near the solar limb and at, or just prior to, the time of the eclipse. The photographs of Mr. Peters, of the United States Naval Observatory, taken at Fort de Kock, Sumatra, confirm Perrine's re sults as to the formation of hoods and en velopes above some of the prominences, and are replete with structural detail. The corona, ac cording to Perrine's photographic negatives, was polarized in the portions 10 minutes beyond the solar limb in all position angles including the polar regions. In negatives of the English astronomer, Newall, some of the streamers showed greater extension in the polarized than in the unpolarized light. In the special work on the corona contemplated by Professor Bar nard's long exposure of a very large plate, and by Dr. Abbott's second attempt to measure the coronal heat bolometrically, there was uni versal interest among astronomers and univer sal regret at the compulsory postponement to a future eclipse on account of the clouds. A photographic search for an intra-Mercurial planet was conducted by Perrine using four lenses covering a stretch of 19 degrees on either side of the solar equator. The faintest stars on the plate range from 6.5 to 9.3 visual magnitude, and Perrine infers that "unless at the time of the eclipse the planetary bodies were directly in line with the sun or with the brightest portion of the corona, there is no planetary body as bright as 5.0 visual magnitude within 18 degrees of the sun, whose orbit is not inclined more than 7% degrees to the plane of the sun's equator.' In the total eclipse of 30 Aug. 1905, the shadow swept over the earth from central Brit ish America, across Labrador, France, Spain and northern Africa, finally leaving the earth at southeast Arabia. This was a very favor able eclipse, the maximum duration being nearly four minutes, and the path of totality unusually accessible. For these reasons, great numbers of parties were organized to observe it, no less than 80 stations being established. It was hoped that a comparison of photographs taken toward the west end of the path with those of Spain and Arabia, taken nearly two and one-half hours later, would throw much light on the changes of the corona during this interval and upon the alteration in appearance of it and of the chromosphere due to the rota tion of the sun. Unfortunately it was cloudy in Labrador, but hundreds of beautiful photo graphs were taken from stations in the Old World. Fabry during this eclipse made a care ful determination of the total light emitted by the corona and found that altogether we re ceive from this object almost exactly as much light as from the moon when this is full.
The shadow in the eclipse of 21 Aug. 1914 swept from the northern boundary of North America across eastern Europe to India. Sev eral parties were formed to observe this eclipse, notwithstanding the very unsettled condition of the countries through which the shadow passed, but the results were, very unfortunately, al most nothing, due to cloudy weather. It was planned to test from this eclipse for the first time the new theory of Relativity, according to which a ray of light passing through a gravi tational field should suffer a minute deflection. The measurement of this deflection seems at present to be about the only physical observa tion which can be made to test the theory, though even with a star which appears at the edge of the sun the deflection should amount to but ninety-three hundredths of a second, and to less than four-tenths of a second for a star 20 minutes away. If found, a deflection would be complicated by the refraction of the rare matter of the corona, the density of which in different portions was also to be investigated by photometric methods.