For seventy years the solar corona has been observed at practically all eclipse expeditions. It consists of an approximately spherical halo, with faint streamers roughly radial in direction, extending a solar diameter or two from the photosphere, mainly from the spot zones. Curved rays also extend from the polar regions of the sun. The extent and shape of the corona vary appreciably from eclipse to eclipse, being closely related to sunspot activity. Total solar eclipses have provided a wealth of observational material on the corona despite the limited time of totality, seven and a half minutes at most. Since 1930, the corona has also been studied extensively without an eclipse, using coronagraphs, especially at Pic du Midi, France, by B. Lyot,1 who successfully built the first coronagraph ; at Arosa, Switzerland, by M. Waldmeier who has obtained very important observational and theoretical results ; and, more recently, at the Harvard station at Climax, Colorado, by W. O. Roberts. Promising experiments by A. M. Skellett and by H. W. Babcock have also been made with a coronaviser designed to observe the corona in full sunlight by photoelectric scanning of an optical image. Eclipse observations of the corona may still be needed, especially for the detection and study of very faint emission lines in the ultraviolet. A number of expeditions are being planned to observe the total solar eclipse of July 9, 1945. Twenty-two missions of approximately 170 astronomers are being organized in Russia. The longest duration of totality in Russia will be only slightly more than one minute. Other scientific expeditions are being organized by the Swedish and Finnish astronomers for the same eclipse. Observations with the coronagraph have the obvious advantage of not being strictly limited in time ; thus they add much valuable information to the observational data gathered by eclipse