the moon, by noting the exact times of contact of the limbs of the sun and moon. The beautiful corona was watched with awe and admiration, and a few sketches were made of its form, -but there the study of an eclipse ended. In fact, an eclipse was watched only if the shadow happened to cross the observer. So little interest was taken in the phenomena, so few investigations were planned, that no expeditions were sent out.

How different is the scientific attitude in the twentieth century! In the year 1901, the writer of this article traveled halfway round the world to the far-off Dutch East Indies in order to observe the total eclipse of May 18 of that year. In other words, he went as far from home as it was possible to go, and the purpose of this trip was to make observations which were concentrated within the time of six short minutes.

The writer regards himself as very fortunate in having been selected four times to become a member of the party of the United States Naval Observatory, and he has thus seen the eclipses of 1900, 1901, 1905, 1918, and altogether has traveled about 40,000 miles for this purpose.

As a matter of fact, an eclipse is not of the rare occurrence that the foregoing remarks might lead one to believe. Each and every year there must be two eclipses of the sun, and there may be even more. Somewhere on the earth each year two eclipses of the sun may be observed, but usually these eclipses are partial eclipses, the sun being only partly obscured. Since few scientific facts can be learned at a partial eclipse, the astronomer takes little interest in them. It is only when the sun's surface is wholly covered up that the matchless corona may be seen; it is only at the time of a total eclipse that there is furnished the unusual opportunity of investigating the sun's surroundings when the brilliant glare of the sun itself is absent.

About once every two years a total eclipse may be seen somewhere on the earth's surface, but as some of these eclipse tracks lie almost wholly on the water surface of the earth, or fall upon inaccessible portions of the globe, it is only on an average of about once in three years that a total eclipse falls at a habitable spot on the earth, even though that location, as in 1901, may be so far away. On the average a total eclipse lasts for about two minutes, so that in a century, about sixty minutes, or one short hour of time, is given to the astronomer for his investigations. Yet in spite of the brevity of time. afforded, some very startling results have been accumulated!

As is well known, an eclipse takes place when the sun, earth, and moon are in a straight line, an eclipse of the sun occurring when the moon comes between the sun and the earth, or when the earth passes into the shadow cast by the moon. The earth makes an annual journey about the sun, traveling in the ecliptic at the speed of more than eighteen miles a second, and accomplishing its journey in 3654 days. The distance from the sun is on the average of ninety-three millions of miles, but the earth's orbit is not a circle but an ellipse, so that the distance from sun to earth may vary one and a half million miles on either side of the mean. Once a month, the moon revolves about the earth, but it likewise does not move in a circle so that the distance from earth to moon varies considerably on either side of the average of 239,000 miles. Moreover, the moon's path is not exactly in the plane of the ecliptic, but is inclined to the ecliptic by a small amount, a little more than five degrees of angle. An eclipse of the sun can take place only at the time of new moon, so that manifestly it is only at the time of new moon, when in addition the moon is near the plane of the ecliptic, that an eclipse of the sun can take place.

Although the average motions of the moon have for some time been so well known that the general time and locations of eclipses may be predicted at long range with a considerable degree of accuracy, still it may be truthfully said that the moon has given the mathematical astronomer more work and worry than all the millions of stars of the universe, with the result that to predict the time of coming of an eclipse at any one locality exactly to the fraction of a second taxes the ingenuity of the astronomer even today. It is no wonder, therefore, that man should always have regarded the moon as of the feminine gender!

The distance and dimensions of the sun and moon being known, it is comparatively easy to find out the diameter of the moon's shadow intercepted by the earth. The maximum width of the shadow is 168 miles, and when all conditions are most favorable, the total eclipse may last for somewhat more than seven minutes. Under average conditions, the region on the earth where the total eclipse may be observed is less than one hundred miles in width, and the average duration of totality is about two minutes of time. The chance that the stay-at-home might see many total eclipses in his lifetime is very limited. As a matter of fact, in London before the eclipse of 1751, there had not been a single total eclipse of the sun visible for more than six centuries. At any one location, an inhabitant would see many more total eclipses of the moon than of the sun. When the moon passes into the shadow of the earth and is eclipsed, then wherever upon the earth's surface the moon is visible, the eclipse may also be seen. The result is that each total eclipse of the moon is visible over more than half the earth, while on the other hand the total solar eclipse is visible only over a narrow track.

Ordinarily a total solar eclipse attracts astronomers from all quarters of

247

the globe for the purpose of making observations. Thus in 1901, in fardistant Sumatra, in addition to a large party from the United States, there were gathered astronomers from England, France, Germany, Holland, and Japan. For the eclipse of 1905, which took place in Europe, there were congregated in the eclipse track, hundreds of astronomers, professional and amateur, from every civilized nation of the world. The trip in 1901 was a most fascinating one, including as it did a journey across the continent to San Francisco; from the Golden Gate to Manila, stopping en route for three days at Honolulu; and ten or a dozen days' stay in Manila while waiting for the United States gunboat which took the party the remaining 2200 miles along the coast of Palawan and Borneo, across the equator, and through the Strait of Sunda to the west coast of the Island of Sumatra. A stay of eight weeks in the interior of the island was necessary in preparation for the eclipse, a site having been chosen at the terminus of the government railroad. The country was picturesque, the manners and customs of the people most interesting, for, belonging as it does to the Dutch, who have peculiar ideas of their own regarding colonization, few foreign influences had been allowed to disturb the primitive lives of the natives. Indeed, ten miles due east of the eclipse camp so little is known of the country that it is said cannibals are still in existence there.

In 1905 there was another attractive trip, when a voyage was made across the Atlantic aboard the U. S. S. "Minneapolis" which was the flagship of Rear Admiral Chester, then Superintendent of the United States Naval Observatory. At Gibraltar, we had the pleasure of viewing the British Mediterranean fleet with Admiral Lord Beresford in command. Eclipse observations were made from the little town of Daroca in the interior of Spain