point properly noticed anywhere-that the seasonal changes in Jupiter correspond to no greater relative change than occurs in our daily supply of solar heat from about eight days before to about eight days after the spring or autumn equinox. It is incredible that so slight an effect as this should produce those amazing changes in the condition of the Jovian atmosphere which have unquestionably been indicated by the varying aspect of the equatorial zone. It is manifest that, on the one hand, the seasonal changes should be slow and slight so far as they depend on the sun, and, on the other, that the sun cannot rule so absolutely over the Jovian atmosphere as to cause any particular atmospheric condition to prevail unchanged for years.

If, however, Jupiter's whole mass is in a state of intense heat-if the heat is in fact sufficient, as it must be, to maintain an effective resistance against the tremendous force of Jovian gravitation-we can understand any changes, however amazing. We can see how enormous quantities of vapour must continually be generated in the lower regions to be condensed in the upper regions, either directly above the zone in which they were generated, or north or south of it, according to the prevailing motions in the Jovian atmosphere. And although we may not be able to indicate the precise reason why at one time the mid zone or any other belt of Jupiter's surface should exhibit that whiteness which indicates the presence of clouds, and at another should show a colouring which appears to indicate that the glowing mass below is partly disclosed, we remember that the difficulty corresponds in character to that which is presented by the phenomena of solar spots. We cannot tell why sun-spots should wax and wane in frequency during a period of about eleven years; but this does not prevent us from adopting such opinions as to the condition of the sun's glowing photosphere as are suggested by the behaviour of the spots.

It may be asked whether I regard the ruddy glow of Jupiter's equatorial zone, during the period of disturbance lately passed through, as due to the inherent light of glowing matter underneath his deep and cloud-laden atmosphere. This appears to me on the whole the most probable hypothesis, though it is by no means certain that the ruddy colour may not be due to the actual constitution of the planet's vaporous atmosphere. In either case, be it noted, we should perceive in this ruddy light the inherent lustre of Jupiter's glowing mass, only in one case we assume that that lustre is itself ruddy, in the other we suppose that light, originally white, shines through ruddy vapour-masses. It is to be remembered, however, that whichever view we adopt, we must assume that a considerable portion of the light received, even from these portions of the

planet's disc, must have been reflected sunlight. In fact, from what we know about the actual quantity of light received from Jupiter, we may be quite certain that no very large portion of that light is inherent. Jupiter shines about as brightly as if he were a giant cumulus-cloud, and therefore almost as white as driven snow. Thus he sends us much more light than a globe of equal size of sandstone, or granite, or any known kind of earth. We get from him about three times as much light as a globe like our moon in substance, but as large as Jupiter, and placed where Jupiter is, would reflect towards the earth; but not quite so much as we should receive from a globe of pure snow of the same size and similarly placed. It is only because large parts of the surface of Jupiter are manifestly not white, that we seem compelled to assume that some portion of his light is inherent.

But the theory that Jupiter is intensely hot by no means requires, as some mistakenly imagine, that he should give out a large proportion of light. His real solid or liquid globe (if he have any) might, for instance, be at a white heat, and yet so completely cloud-enwrapped that none of its light could reach us. Or, again, his real surface might be like red-hot iron, giving out much heat but very little light.

I shall close the present statement of evidence in favour of what I begin to regard as in effect a demonstrated theory, with the account of certain appearances which have been presented by Jupiter's fourth satellite during recent transits across the face of the planet. The appearances referred to have been observed by several telescopists, but I will select an account given in the monthly notices of the "Astronomical Society," by Mr. Roberts, F.R.A.S., who observed the planet with a fine telescope by Wray, 8 inches in aperture. "On March 26, 1873," he says, "I observed Jupiter about 8 p.m., and found the fourth satellite on the disc. I thought at first it must be a shadow; but, on referring to the Nautical Almanac,' found that it was the fourth satellite itself. A friend was observing with me, and we both agreed that it was a very intense black, and also was not quite round. We each made independent drawings which agreed perfectly, and consider that the observation was a perfectly reliable one. We could not imagine that such an intensely black object would be visible when off the disc, and waited with some impatience to see the emersion, but were disappointed by fog, which came on just at the critical time." Another observer, using a telescope only two inches in aperture, saw the satellite when off the disc, so that manifestly the blackness was merely an effect of contrast.

In considering this remarkable phenomenon, we must not forget that the other satellites do not look black (though some

of them look dark) when crossing Jupiter's disc, so that we have to deal with a circumstance peculiar to the fourth or outermost satellite. Nevertheless, we seem precluded from supposing that any other difference exists between this satellite and the others than a certain inferiority of light-reflecting power. I might indeed find an argument for the view which I have suggested as not improbable, that Jupiter is a heat-sun to his satellites, since the three innermost would be in that case much better warmed than the outermost, and therefore would be much more likely to be cloud-encompassed, and so would reflect more light. But I place no great reliance on reasoning so ingenious, which stands much as a pyramid would stand (theoretically) on its apex. The broad fact that a body like the fourth satellite, probably comparable to our moon in light-reflecting power, looks perfectly black when on the middle of Jupiter's disc, is that on which I place reliance. This manifestly indicates a remarkable difference between the brightness of Jupiter and the satellite; and it is clear that the excess of Jupiter's brightness is in accordance with the theory that he shines in part with native light, or, in other words, is intensely heated.

This completes the statement of the evidence obtained during the recent opposition of Jupiter in favour of a theory which already had the great advantage of according with all known facts, and accounting for some which had hitherto seemed inexplicable. If this theory removes Jupiter from the position assigned to him by Brewster as the noblest of inhabited worlds, it indicates for him a higher position as a subordinate sun, nourishing with his heat, as he sways by his attractive energy, the scheme of worlds which circles round him. The theory removes also the difficulty suggested by the apparent uselessness of the Jovian satellites in the scheme of creation. When, instead of considering their small power of supplying Jupiter with light, we consider the power which, owing to his great size and proximity, he must possess of illuminating them with reflected light, and warming them with his native heat, we find a harmony and beauty in the Jovian system which before had been wanting; nor, when we consider the office which the Sun subserves towards the members of his family, need we reject this view on account of the supposition—

That bodies bright and greater should not serve
The less not bright.

359

THE OUTLINE OF CLOUD FORMS: THE ELECTRIC CUMULUS, ANVIL CLOUD, AND RAIN-BALL.

BY SAMUEL BARBER, F.M.S.

Y the outline of a cloud we most readily distinguish its class, and the peculiar characteristics it may happen to possess as a species. To take the case of Cumulus. Although the term "Cumulus" (a mass) does not indicate, so well as the names of the other two types adopted by Howard, the form which it presents to the eye, yet, in fact, the varieties of this cloud, which is of the utmost value in weather prediction, are, by outline alone, more distinctly defined and easier to be discriminated than those of any other. It would, perhaps, be possible to form a classification of the various species of the cloud entirely by their outlines.

I distinguish four main varieties of cumulus, from which I select two forms, the most striking and the most valuable as weather prognostics, to illustrate the use of observing outlines. The first is the Electric Cumulus,* and the second the Anvil Cloud of Sir J. Herschell. The outline of the Electric Cumulus is very sharp and hard, not broken by large indentations, and loose feathery processes like that of the commoner kinds of cumulus, formed rapidly by exhalations from the earth, and drifting near its surface: but we have almost a continuous line, terminating perhaps in stratus or cumulonimbus below, yet rising to a vertex, conical or rounded, and consisting of a series of small curves and minute projections, representing the rounded protuberances on the surface of the cloud.

There is no cloud that possesses a more distinct outline than this, and none which exhibits more brilliant or dazzling effects of light. As a weather prognosticator it is best observed at a distance, and near the horizon, when the form of the summit may be more exactly discerned.

We now take the Anvil Cloud.

Previously described by me in the "Quarterly Journal Brit. Meteor. Soc.," Jan. 1872.

This species, alluded to by Sir John Herschell in his “Familiar Lectures on Science," cannot be regarded as so strictly a cumulus as the form we have just remarked upon, inasmuch as it exhibits at times a great tendency to the form of stratus, and at other times to that of cirrus; or, perhaps we should rather say to the form of cumulo-cirrus.* As an indicator of wind, this cloud may probably be regarded as unrivalled. It frequently appears two or three days before heavy gales, especially when they are of long duration, as about the Equinoxes. Without being hyper-critical, or, I should say, hyper-analytical, we may distinguish three varieties of the cloud, the distinction between the outlines of which is quite characteristic. The first variety possesses greater affinity for ordinary cumulus than the other two, and it often shows great resemblance in many points to the electric cumulus. It occurs for the most part in large masses, or banks, the summits of which, as it drifts along with the wind, stand out in sharp relief against the upper sky, and exhibit the most striking and fantastic resemblances to terrestrial objects, beetling crags, towers, and heads of animals-the last being of frequent occurrence. When of this variety, the Anvil Cloud appears to be highly condensed, and is of a dark blueish or slatey tint. It is generally, I believe, the precursor of heavy rain as well as wind.

The second variety of the cloud, which approaches to stratus, exhibits a more irregularly formed "anvil," the "waist " being usually much more conspicuous on one side than on the other. This is also a dense cloud, without much marking on its surface, the outline being not nearly so irregular as that of the variety just described. It is of a lighter tint, grey or muddy blueish, and does not, I think, occur in an isolated form, but rises from a bank of lower stratus or cumulo-stratus. It is also a wet-weather cloud.

The third variety is apparently much thinner in texture than the two preceding, being, however, of a double character: cirrus in the upper part, condensed cumulus, or cumulo-stratus in the lower. This is the least common of all three varieties, but is an excellent indicator of wind.

There is another form of cloud which has a close affinity for this last-mentioned species; a kind of aggregated cirrus, generally of a dark tint and seen at high altitudes, in windy weather. This lacks the consistency and volume of cumulus, and mostly spreads itself out in a sheet, with rounded edges, and without surface marking; whereas the third form of anvil cumulus ex

* This cloud, also an excellent indicator of wind, does not appear to have received the attention it deserves from meteorologists. The reader must not confound it with cirro-cumulus.