This base may be either a flat of an eroded elevated reef or of a similar substructure of volcanic rocks, the nature of that base depending absolutely upon its character when elevated in a former period to a greater height than it now occupies. . . . The Fijis are not situated, as was supposed, in an area of subsidence, but on the contrary, they are in an area of elevation, so that the theory of Darwin and Dana is not applicable to the islands and atolls of the Fiji group." As it was a study of these and similar groups which led Darwin and Dana to adopt their theory of subsidence, Mr. Agassiz seems to dispose of the theory as applied to any islands, although he says in a later paragraph, "The great variety of causes which have been active in shaping the present physiognomy of the reefs and atolls of Fiji shows the impossibility of assigning any one factor . . . as the single cause for the formation of the many different kinds of atolls and barrier-reef islands to be found in the Fiji group." Mr. Agassiz promises a fully illustrated report some time during the coming summer.
New Theories regarding the Rainbow.—The old theory of the rainbow, which is still found in optical text-books, only imperfectly accounts for the true bow and fails entirely to explain the "spurious bows" which often accompany the former. It is usually possible by close examination to distinguish certain colors on the inside edge of the primary which are not consistent with the simple series of spectrum colors, accounted for by the Descartes theory. In a paper on The Intensity of Light in the Neighborhood of a Caustic, Airy seems to have laid the foundation for an adequate theory of the rainbow. Au article in Nature on this subject says that Mr. J. M. Pernter is working out the new theory, and his general conclusions are given as follows: "The greater the drops the more spurious bows." A chief bow of intense pink and green (hardly any blue) indicates drops of diameters ranging from one to two millimetres; intense red always speaks for big drops. Secondaries (spurious bows) of green and violet (the blue is marked by contrast) without yellow, immediately forming the chief bow, correspond to drops of 0.5 millimetre, while five and more secondaries without white and without breaks mark drops of 0.1 millimetre. A partly white bow is produced by drops of 0.06 millimetre, and when the drops are still smaller, a real white bow with orange-yellow and blue margins is the result. The net result of these elaborate investigations will be to add a new interest to a natural phenomenon already endowed with many associations of magic and beauty."
The Recent Total Eclipse of the Sun.—While there have been no complete reports from any of the observation parties of the total eclipse of the sun of Saturday, January 22, 1898, enough information has been received to indicate that the majority of them were successful and that much data of extreme scientific value has been obtained. The only district in which the weather was not propitious seems to have been southern Russia, only a few good photographs being obtained. A telegram received from the Rev. J. M. Bacon, in charge of the British Astronomical Association at Buxar, announces perfect weather and "observations satisfactory all round." This party is reported to have taken a successful series of cinematograph pictures. The party located at Jeur, from the Lick Observatory, under the direction of Prof. W. D. Campbell, reports very successful observations. The unusually favorable weather, the number of trained observers on the ground, and the delicate and powerful instruments at hand, have evidently combined to make this one of the most important scientific events in the closing years of the century. In calling attention to this fact Nature says: "With such a list of successes we may safely say that this eclipse, as befitting the last one of the century, has surpassed all previous records; but, unlike many eclipses at the beginning of the century, it can not be truly said that the event of Saturday was over at the end of two minutes of totality. To many the eclipse has yet to begin and will last for many months, during which time each line in the spectrum, each streamer of the corona, each prominence on the sun, will be analyzed little by little to discover if we have similar streamers in other coronas or identical lines in our laboratories. It may be confidently expected that the results obtained
