To Him who is everywhere, the light-record of all that has taken place on earth is being continually conveyed, the remembrance is ever present with Him; 'the eyes of the Lord are in every place beholding the evil and the good,' 'His eyes are upon the ways of man, and He seeth all his goings.'

But, lastly, let us remember that even these thoughts, startlingly though they impress upon us the fact that nothing that is done shall be forgotten, are altogether imperfect. It is well for us to form some idea of the all-seeing vision of God, by speaking of the eyes of God, and by comparing His knowledge with that direct knowledge of events which we obtain by means of the sense of sight; but we must not forget that this mode of speaking is really as far from the truth as are the poetical expressions by which the inspired writers speak of the might of God's arm, or of His holding man as in the hollow of His hand. There is that continual record of events by means of light-waves travelling for ever and ever through space; and beyond question, the Almighty is as cognisant of those light-waves as of any event actually taking place in this world or in others. But His knowledge is infinitely more perfect and complete than any we obtain even of the simplest events by means of our senses. "God looketh to the ends of the earth, and seeth under the whole heaven. No thought can be withholden from Him.'

A CLUSTER OF SUNS.

There shall be no night there.-Rev. xxi. 25.

In the previous chapter I spoke of the existence of senses —that is, means of judging of external things-unlike any that are possessed by mankind. It appears to me that the study of astronomy is calculated to suggest ideas of the kind, thoughts of what is unfamiliar to us, conceptions of states of existence even unlike any that we know of. For in studying astronomy we find ourselves brought into the actual presence of systems where much must necessarily be quite different from what we experience here on earth. It is of a system of this kind that I propose now to speak.

When we search the heavens with powerful telescopes, we perceive besides the stars certain cloudlike objects, which have been called nebula, from a Latin word signifying a cloudlet. There are many hundreds of these objects of various orders. But my present purpose is to consider only one class of celestial cloudlets-the star clusters. For many of these cloud-like objects, when examined with suitable telescopic power, are found to consist of myriads of stars. Within a minute space of

the heavens, since even in the telescope many of these cloudlets remain exceedingly small, thousands on thousands of suns are seen, and probably many thousands of suns are there also which are unseen, because smaller than the rest. Some of these objects are amazingly beautiful and splendid, insomuch that it has been said of one of them-a cluster called 13 Messier in Hercules-that probably no one who has beheld this cluster for the first time in a telescope of great power has been able to refrain from a shout of wonder.

Now, it was formerly held that these clusters of stars are in reality galaxies like our own Milky Way, sidereal systems whose stars are suns like our own separated from each other by distances like those which separate our sun from his neighbours among the stars. But certain very simple considerations oppose themselves to this view of the nature of the star-clusters, and show that they are really distinct in their nature from that part, at any rate, of the sidereal system to which our sun belongs. When we see one of these clusters of a rounded figure as a whole, and also gathering more and more richly towards its centre, always with a uniform roundness for each order of richness, we are certain that we have in view a globular system of suns. And when we look at any globe-shaped object of any apparent size, we know within what limits of relative distance the different parts of that object lie.

To explain my meaning--if there is a soap-bubble an inch in diameter, and a yard from the eye, we know that

the farthest point of that bubble is farther than the nearest point by a thirty-sixth part of the bubble's distance; so much we could learn, let us say, by measurement: but if that bubble, so placed, looked just as large as a balloon very much farther away, then, although we could not tell by measurement how far off the balloon was, we should 'know without measurement that precisely the same relation held in the case of the balloon-namely, that its farthest point was farther away than its nearest point by a thirty-sixth part of the balloon's distance. And if the soap-bubble just concealed a star-cluster, or the sun, or moon, or in fact any globe-shaped object at any distance whatever, precisely the same relation would hold. Accordingly, though astronomers have no means of measuring the actual distance of any star-cluster-or, at any rate, have not yet succeeded in effecting such measurement—they can infer with great certainty within what limits of relative distance the parts of such clusters lie, because they can very accurately measure the apparent size of any starcluster. Only, before passing from the soap-bubble illustration, let me warn the reader not to imagine that any star-cluster is so large as that illustration might suggest— a pin's head, at a yard's distance, would look larger than most of the chief star-clusters.

Now, when such a process has been applied, it is found that the farthest star in a star-cluster of the brighter and richer order cannot be at a relatively much greater distance than the nearest star of such a cluster;

and when due account is taken of the enormous number of stars in such clusters, it follows that, on a very moderate computation, the distance separating star from star in the heart of a rich cluster must be less than a millionth part of the distance separating the cluster from

our sun.

But we can see the separate stars of these clusters, and sometimes with telescopes of no great power. We may, then, regard the stars thus seen as of at least the twentieth order of magnitude—that is, shining as brightly as stars of the first magnitude removed to but twenty times their present distance.1 If, therefore, we could approach one of these clusters until we were within onetwentieth of its present distance, we should see its leading stars shining as stars of the first magnitude. But we should still be fifty thousand times farther from the cluster than its component stars are from each other, at least in the rich central region of such a cluster. What, then, would happen if we could continue our imagined journey until we were in the very heart of the cluster of suns?

It is very easy to answer. Suppose we were then midway between two leading suns of the group, then we should be at a distance from either equal to only the one hundred thousandth part of that which had

1 These statements are all far within the truth, and are presented in their present form to avoid the necessity of a detailed account of the estimates of astronomers respecting star magnitudes.