faint shadows well, would exhibit the shadow of the fourth satellite as perceptibly smaller than that of the third; whereas a telescope more efficient in this particular respect would show the shadow of the fourth satellite larger than that of the third. Or, even if only the fourth satellite and its shadow were transiting the disc, it is obvious that the former telescope would make the shadow seem smaller than the satellite, while the latter would make the shadow larger than the satellite.

It may be remarked here, in passing, that a difference corresponding to that here discussed respecting the satellites' shadows exists in the nature of the shadow of Jupiter at the parts of the cone entered by the several satellites. Thus the disappearance of IV. takes a much longer time than that of I., not merely on account of the slow motion of IV. (as is commonly stated), but on account also of the much wider fringing of penumbra in the cross-section of the cone of shadow where IV. enters it, as compared with the fringe in the cross-section where I. enters.

TERRESTRIAL MAGNETISM.

THE study of terrestrial magnetism is gradually assuming an importance which half a century ago few would have thought it could ever attain. It is being brought into intimate relationship with laws affecting not our own earth only, but the whole of the solar system. Indeed, interesting as are the bonds of union which Copernicus and Kepler and Newton have traced in the relations of our system, it would seem as though we were approaching the traces of a yet more wonderful law of association. We see the earth's magnetism responding to the solar influences, not merely in those rhythmic motions which belong to the periodic variations, but in sudden thrills affecting the whole framework of our globe. The magnetic storms which are called into action by such solar disturbances as the one of September, 1859 (witnessed by Messrs. Carrington and Hodgson), are, we may feel sure, not peculiar to our own earth. The other planets feel the same influence-not perhaps in exactly the same way, but according to the constitution and physical habitudes which respectively belong to them. So that one can scarcely conceive a subject of study at once more promising and more interesting than the science on which I now propose to make a few remarks.

To deal fully with terrestrial magnetism would require very much more space than is now at my disposal. The mere history of the science would need a volume; its relations to experimental physics would need another; and one other at least would be required to treat of the theories

which are suggested by the interesting relations which have been discovered by modern observers.

In this paper, then, I merely seek to make a few notes on certain points which have occurred to me in reading what Arago, Hansteen, Humboldt, Sabine, and others have said on the subject of terrestrial magnetism. There are certain peculiarities in the way in which the results of observation have been dealt with, which I wish in particular to comment on. I think, too, that certain general results appear to follow from what has been discovered respecting the secular variations of the earth's magnetism, which have not yet (so far as I know) been very closely attended to.

There are three features of the earth's magnetic action which are chiefly attended to by observers-the declination, the inclination, and the intensity.

The declination is the angle at which the horizontal needle is inclined to the north-and-south line. The inclination is the angle at which the dipping needle is inclined to the horizon. The intensity is the magnitude of the force with which the needle seeks the position of rest.

Now if we travelled over the whole surface of our earth, and carefully determined the declination, the inclination, and the intensity of the magnetic action at every point, we should be able to map down on a chart of the earth the relations thus presented to our notice. And speaking generally we should have the following peculiarities to deal with :

First, as to the declination. We should find that in certain regions the magnet's northern end was to the west of north, while in certain other regions the reverse was the case. If we marked in the boundary line between these regions, it is obvious that we should have traced in a line along which the needle would lie due north and south. This is what is termed a line of no declination. On charts of the earth's magnetic relations, the position of this line for about the middle of the present century is usually indicated. In some maps a set of lines used to be added, along each of which the magnetic needle had a definite declination.

T

No such lines are now marked in, however. The reason why they are omitted is that they are very complex. But I wish to call particular attention to the omission, for this reason, that the same consideration which renders the lines of equal declination unnecessary in a chart of the earth's magnetism, the same reason which gives them a complexity altogether unmeaning so far as the cosmical relations of terrestrial magnetism are concerned, renders also the line of no declination worthy of much less consideration than it has received at the hands of many eminent physicists. I shall presently show why this is so.

Secondly, as to the inclination. We should find as we travelled over the earth's surface, that the dipping needle tends to verticality at two nearly opposite points, one close to the Arctic and the other to the Antarctic circle. These are called the northern and southern inclination-poles, and must not be confounded with the intensity-poles presently to be mentioned. As we leave either inclination-pole the dipping needle leaves its vertical position, and gradually approaches the horizontal direction, until, along a curve lying midway between the two poles, the needle becomes exactly horizontal. This curve is called the magnetic inclination-equator. The present positions of the inclinationpoles are indicated in magnetic maps. The inclinationequator is also indicated. The curves which run like parallels around the two poles in such maps indicate the curves along which the dipping needle has a definite inclination (different, of course, for each curve). Another series of lines, which intersect in the poles, or at least tend towards the poles, run so as to cross the inclination-parallels at right angles. But they illustrate, properly speaking, the peculiarities of declination. They show in what direction the horizontal needle points in different parts of the earth. But it is to be noticed that as they cross the real meridians at constantly varying angles they are in no way associated with the lines of equal declination described in the preceding paragraph.

Lastly, as to the intensity. If we noticed, in every part of the earth's surface, the number of times the needle vibrated through its position of rest in a given interval (this number affording a very exact measure of the intensity of the magnetic directive action), we should find that along a curve lying near to, but not absolutely coincident with the inclination-equator, the intensity has a minimum value. This curve is called the intensity-equator. Leaving it either towards the north or south, we should find the intensity gradually increasing. We should not, however, find this increase guiding us to a northern or southern intensitypole; but we should recognise two magnetic intensity-poles in each hemisphere. The positions of these are indicated in magnetic maps. The intensity-equator shows a certain resemblance to the inclination-equator, but the two equators do not coincide.

Now in considering the various relations here presented, it is very important that we should decide which property of the magnetic needle should receive our chief attention.

General Sabine considers that the intensity is the primary quality of the magnet in all such inquiries as we are at present concerned with; that is, in all inquiries respecting the wider relations of terrestrial magnetism. He remarks that 'all that relates to the force has a more immediate bearing than what relates to the direction of the needle, either in the horizontal or vertical plane; these planes, although necessarily used by us both in observation and discussion, not bearing in themselves any direct relation to magnetism.'

Without denying the force of these remarks, I could yet venture to point out reasons why the force is a less suitable relation than the direction, where the question is one of mapping down the equator, parallels, and poles, which are to teach us something of the earth's magnetic action and its secular changes.

I must premise that the declination and the inclination may be combined in a single indication-viz., the direction,