allies, the spectroscope, the polariscope, and photography-respecting the aspect and general condition of the Sun's photosphere, the coloured prominences, the corona, and so on. The physical condition of the Sun, the amount and probable source of his heat, and other like questions, will next be dealt with. And, finally, we shall consider the system of suns, and the position which our Sun holds in that system, so far as the researches yet made by astronomers enable us to deal with this noblest of all subjects of study.

Thus we have before us a sufficiently wide range of research. Our progress will lead us to consider some of the most successful attempts yet made by man to resolve the mysteries of the universe. We shall have to deal with much that invites reflection and speculation-with much that may be explained by the thoughtful study of evidence already obtained;-but also with much that continues, and may perchance continue for many years, altogether perplexing. We have, in fine, to deal with a subject which is full of interest, but whose real grandeur and significance, as well as its vast difficulties, are but now beginning to be rightly understood.

7

CHAPTER I.

THE SUN'S DISTANCE AND DIMENSIONS.

THE DETERMINATION of the Sun's distance is not only an important problem of general astronomy, but, so far as the subject of this treatise is concerned, it may be regarded as the very foundation of all our researches. For until we know the Sun's distance we can determine neither his bulk nor his weight; and our views even as to his physical condition will be found to depend in an important degree on the estimate we form respecting those two elements. A minute error in the solution of the problems on which the determination of the Sun's distance depends would not only result in adding or withdrawing hundreds of millions of cubic miles from the Sun's volume, and many multiples of the Earth's mass from his weight, but our conceptions of the size of solar spots, the height of the coloured prominences, the velocity of solar currents and cyclones, and many other such matters, would be rendered proportionately erroneous. It is, therefore, of the utmost importance that we

should have accurate views respecting the modes of research by which the Sun's distance has been estimated, and that we should know the probable limits of error in the resulting determinations. The subject has a special interest at the present time, because preparations are even now being made for the application, in the winter of 1874, of one of the most effective methods at the disposal of astronomers for the solution of this most noble problem.*

Let us consider the general nature of the problem, in order that we may the better appreciate its difficulties. Many are disposed to wonder that astronomers should not long since have mastered what seems to be an almost elementary problem of the science; and it has appeared as a blot on the fair fame of astronomy that errors and differences of millions of miles should have. followed the attempt to solve this particular problem. Then, too, when the resulting errors in the determination of the distances of the outer planets, and the still larger errors in the determination of the distances at which the fixed stars lie from us, are considered, the inquiry is suggested, Where is the boasted accuracy of the most exact of the sciences?' We shall see as we proceed that the great wonder is, not that the estimates made by astronomers should differ, but that even the vaguest ideas should have been formed respecting the Sun's distance. The problem, as presented in its simplicity (the simplicity of perfect difficulty),

* See Appendix A, for an account of what is anticipated from the observations to be made on the transits of 1874 and 1882.

seems at first sight one that no human ingenuity could avail to solve.

In fig. 1 let s represent the Sun, and E E' E" the orbit of the Earth. Then on the scale of the figure, a dot meant to represent the Earth should have dimensions so minute that it would be altogether invisible to the naked eye. The line representing the Earth's path is broad enough to obliterate more than a hundred such dots placed side by side across its width. Now, it is the inhabitants of the globe represented by this tiny dot,

who have to measure the distance separating the inaccessible globe s from E E' E." The globe on which they live is continually rotating on its axis, it is sweeping onward with inconceivable velocity on the path E EE", yet from this rotating, onward-rushing, and relatively minute orb, the observations are to be made by means of which the vast distance of s has to be determined. Still considering fig. 1, let us, in yet another way, picture to ourselves the nature of the problem to be dealt with.

Suppose the Earth on the orbit E E' E' to be represented by a globe one inch in diameter. Then the Sun at s would be represented by a globe 9 feet in diameter, and his distance from the orbit E E' E' by no less than 320 yards. Let any one who has noticed the 300 yards' range in rifle shooting consider how minute a disc one inch in diameter would appear at that distance, and he will at once recognise how difficult a problem the astronomer has to solve in determining the Sun's distance; for that minute and scarcely perceptible disc subtends the very angle on whose exact measurement the solution of this problem depends. But even when any one has pictured to himself the difficulty of determining the exact angle subtended by an inch disc at that distance, and how easily the angle might be over-estimated or under-estimated by a considerable fraction of its real value, even then he will not have realised the actual difficulty of the problem. Let him reverse the illustration, and picture the difficulty of determining by observations made from points within this inch disc the distance of a station 320 yards off. Yet even then he will have underrated the difficulty of the actual problem astronomers have to solve. He must suppose the two sets of observations to be made by different observers, at different seasons, in different weather, with different instruments; that each set of

It is convenient to remember that a bronze halfpenny is exactly one inch in diameter; so that an exact representation of a great circle of the earth, on the scale we are now considering, can be conveniently referred to.