the spectral lines appeared double, Professor Pickering predicted that they would be again double on or about December 9, 1889. This prediction was duly fulfilled on December 8, thus proving the reality of the discovery. Assuming that the orbit is circular, with its plane passing through the earth, or nearly so, he finds that the distance between the components is about 143 millions of miles, or about the distance of Mars from the sun, and their combined mass about 40 times the mass of the sun. Considering the brightness of the star, and its probably vast distance from the earth, this great mass is not very surprising. Mizar has long been known as a wide double star, the companion being of about the fourth magnitude, and visible with a small telescope. Its duplicity was discovered by Riccioli in 1650, and it was measured by Bradley in 1755. It was the first pair photographed by the American astronomer Bond. It must now be looked upon as a triple star. Close to it is a fifth-magnitude star, known as Alcor, which is visible to the naked eye, and was considered by the ancients as a test of keen vision. It is now, however, plain enough to good eyesight, and is sometimes spoken of as a "nakedeye double." Mizar is therefore a most interesting star; double to the naked eye, a closer double with a moderate telescope, and yet again double to the eye of the spectroscope. Between Mizar and Alcor is an eighth-magnitude star, discovered by Einmart in 1691.

Professor Pickering thinks that the greatest distance between the components of Mizar may perhaps be about 1 times its annual parallax, and is probably far too small to be ever detected by any telescope. Klinkerfues found for this star a very small parallax, indicating a distance about 5 million times the sun's distance from the earth, or a journey for light of about 76 years! The spectroscope has thus enabled us to discover the existence of an invisible body! If the orbit is slightly inclined to the line of sight, the dimensions and corresponding mass of the system would be increased. It seems improbable that the plane of motion passes exactly through the earth, for in that case there would be an occultation twice in each revolution which would probably produce some diminution in the light of the star, as in the case of variable stars of the Algol type. I am not aware that any such regular variability has been observed in the light of Mizar. We must therefore conclude that the mass of the system is really more than that computed by Professor Pickering.

A similar spectroscopic result has been found in the case of the bright star Beta Auriga, for which the observations indicate a period of about 8 days, with a diameter of the orbit of about 16 millions of miles. From these data I find that the combined mass of the com

ponents would be much less than in the case of Mizar-about 1 that of the sun. A similar variation was found to occur in the star 44 Ophiuchi. This star has been strongly suspected of fluctuations in its light, and it may possibly be a variable of the type of Algol. Professor Vogel finds a similar motion in the bright star Spica-the leading brilliant of the constellation Virgo, or the Virgin-with a period of about 4 days. Here however the lines are merely shifted, not doubled, or at least not distinctly so, as in Mizar and Beta Auriga. This indicates that one of the components is so faint that its spectrum is not seen, or only seen with difficulty, and that the observed motion is chiefly that of the brighter component. From the observed velocity-about 53 miles a second-Vogel computes that, for components of equal mass, the total mass of the system would be about 2 times the mass of the sun.1

With reference to the Algol variables, it has long been suspected that the decrease in their light at minimum might possibly be due to the interposition of a dark eclipsing satellite. This periodical variation in the light of Algol itself, seems to have been known to the ancients, as its name implies the "demon star." The true character of its variation was, however, first determined by Goodricke, in 1782, when its period from minimum to minimum of light was 2 days 20 hours 48 minutes 59 seconds. This has slowly diminished to its present value of 2 days 20 hours 48 minutes 51 seconds, according to a recent investigation by Chandler. Some few years since Professor Pickering undertook a mathematical investigation of the case, and showed that a dark eclipsing satellite revolving in a nearly circular orbit round Algol, in the period indicated by the light variation, would explain the observed phenomenon within the limits of errors of observation, and he pointed out that the orbit of the bright star might be determined by spectroscopic observations without any knowledge of the star's distance from the earth.

Assuming the correctness of this hypothesis, and taking into consideration the observed diminution of light at minimum, Mr. Maxwell Hall computed that the density of Algol is only one-fourth that of water. From spectroscopic observations made by Professor Vogel at Potsdam in 1888 and 1889, he concludes that the decrease of light is really due to an eclipsing satellite. He found that before the minimum of light the star is receding from the earth at the rate of 24 miles a second, and, after the minimum, approaching with a

1 Mr. Fowler has quite recently found that the bright star Vega is also a close double, with a period of only 24 hours 41 minutes, and a mass about 22 times that of the sun.

velocity of 28 miles. The observations also show a motion of translation of the system in space at the rate of about 2 miles per second, towards the earth. Assuming the orbit to be circular with its plane passing through the earth, Professor Vogel computes the diameter of Algol at 1,061,000 miles, and that of the dark companion 830,000 miles, with a distance between them of 3,230,000 miles. He makes the mass of Algol four-ninths of the sun's mass, and that of the companion two-ninths, or a combined mass equal to two-thirds of the mass of the sun. Taking the sun's density as 144, and its diameter 866,000 miles, I find that the above dimensions give a mean density for the components of Algol of about one-third of that of water, not differing much from Maxwell Hall's result, and showing the correctness of his conclusion that, "in the case of the components of Algol, as Mr. Lockyer argues in the case of the sun, we are undoubtedly dealing with masses of gas." The spectrum of Algol is of the first or Sirian type, all the spectral lines being faint except those of hydrogen, a type of spectrum which indicates that the star is very hot, and therefore probably in the gaseous state. This confirms the conclusion as to its density derived from the spectroscopic evidence of its orbital motion, and proves the correctness of the hypothesis that the variation in its light is due to a dark eclipsing satellite.

Professor Vogel assumes that both the components of the Algol system have the same density. But if this be so, we have the curious case of two bodies not differing largely in volume, of which one is intensely hot, and the other nearly a dark body. Vogel does not, however, consider it necessary to assume that the satellite is absolutely dark. It may be still in a very heated condition, but to agree with the observed variation the light of the companion cannot be greater than one-eightieth of that of Algol itself. As the spectrum of Algol is of the first type, we may conclude, I think, that the intensity of its light is greater than that of our sun. The light emitted by the satellite may therefore possibly be equal to several thousand times the light of the full moon without interfering with the hypothesis. Professor Vogel refers to the parallel case of Sirius and its comparatively dark companion.

The brightness of Algol and its comparatively small mass might be taken to indicate a relative proximity to the earth; but if its parallax were even one second of arc (a highly improbable value), the greatest distance between the components would amount to only one twenty-ninth of a second, a distance quite beyond the dividing power of even the largest telescopes.

VOL. CCLXX. NO, 1922.

It is to be hoped that the spectroscopic method may be applied to other stars of the Algol type, but some of these are so faint that a very large telescope would be required for the purpose. The following are, however, sufficiently bright to repay examination with telescopes of moderate power: Lambda Tauri, magnitude 3, and Delta Libræ, of the 5th magnitude. The others we must leave to the great Lick telescope or Mr. Common's 5-feet reflector.

J. E. GORE.

SOME MORE CURIOSITIES OF

EATING AND

DRINKING.

as an experiment, lived for several weeks on sixpence a day and fared sumptuously. Unfortunately I did not fully carry out the well-known advice of a famous surgeon of another generation, for though I lived on sixpence a day, I did not at the time even earn that miserable sixpence. Although I can most comfortably go inany hours without food, I do not deny the extreme fascination of dainty fare and of everything relating to it; I have made an exhaustive study of the subject, finding endless delight in it, though often wondering whether I could not occupy my time more intelligently and instruct my neighbours better in other ways. But one must write what the world will read, and, indeed, the subject is not without its historic value; you learn so much when, as in the case of the Great Napoleon, you are informed that he was greedily fond of mutton and garlic; then "there was Hallam with his mouth full of cabbage and contradiction," as Sydney Smith said of him when describing a dinner-party; and Humboldt, his plate piled up with all the delicacies on the table, too intent on getting still more to eat what he already had.

What an insight the following passage from one of Froude's most charming "Short Studies" gives into the character of the great Dutch man who seemed to hold the key of the position in the stirring Reformation days, but whose extreme vacillation and constitutional timidity made him drift hither and thither aimlessly, like a ship without rudder or pilot :

Erasmus moved about in Switzerland and on the Upper Rhine. The lakes, the mountains, the waterfalls, the villas on the slopes delighted him when few people else cared for such things. He was particular about his wine. The vintage of Burgundy was as new blood in his veins, and quickened his pen into brightness and life. The German wines he liked worse--for this point among others, which is curious to observe in these days. The great capitalist wine. growers, anti-reformers all of them, were people without conscience and humanity, and adulterated their liquors. Of course they did. They believed in nothing

02