In the preceding pages I have described the principal features presented by comets as they approach and pass away from the neighbourhood of the sun. The various hypotheses which have been put forward to account for these peculiarities must now for a brief space claim our attention. Although we are far from being in a position to theorise with any confidence respecting the nature of comets, and still less as to the purposes which they subserve in the economy of nature, yet the observations made upon the second comet of the year 1868 have resulted in a positive discovery which may serve as a stand-point, so to speak, whence we can examine somewhat more confidently than of old, the various theories which have suggested themselves to those who have studied cometic phenomena.

In considering these hypotheses we have to distinguish between the views which have been entertained respecting the nucleus and coma, and those which regard the less intelligible phenomena presented by the tail. This remark may seem trite and obvious, but in reality the two classes of hypotheses are found singularly confounded together in many works on popular astronomy. Let it be understood then, that when, in speaking of an hypothesis respecting comets no special mention is made of the tail, it is to be assumed that the hypothesis applies solely to the head of the comet. The same holds, by the way, with reference to the phenomena presented by comets. For instance, when we said in the paper on Comet I. that comets grow smaller as they approach the sun, the remark was to be understood to apply to the volume of the head, not to the whole space occupied by the head and tail. In fact, it would have been impossible to assert anything with respect to the volumes of comets tails, inasmuch as the apparent extent of these appendages varies according to the atmospheric conditions (humidity, clearness, and so on) under which the comet is observed, and also according to the light-gathering power of the observer's telescope.

To return, however, to the theories which have been formed respecting comets.

It has been commonly admitted that the substance of which comets are composed is either wholly or principally gaseous. In no other way, it should seem, can the remarkable variations of appearance which comets present as they approach the sun or recede from him be reasonably accounted for.

Kepler held that comets are wholly gaseous, and that they are liable to be dissipated in space by the sun's action. He supposed that the process of evaporation which thus led to the destruction of a comet was carried on through the medium of the tail. It need hardly be said that modern observations are completely opposed to this view. Comets have been seen to return again and again to the neighbourhood of the sun without any apparent diminution of volume, although at each return a tail of considerable length has been thrown out. For a long time, indeed, it was thought that Halley's comet was gradually diminishing in volume; but at the last return this magnificent object had recovered all its pristine splendour.

Newton held, on the contrary, that comets are partly composed of solid matter. He supposed that only the gaseous matter was affected to any noteworthy extent by the action of the sun's heat. Raised from the solid nucleus the vaporised particles passed first into the coma, he imagined, and were thence carried off into space to form the comet's tail. Others so far modified Newton's views as to suggest that the vaporised matter is not wholly carried off but partially re-precipitated upon the head of the comet, just as the vapours raised from the ocean are precipitated upon the earth in the form of rain.

We have seen that a comet diminishes in volume as it approaches the sun. It will be noticed that both the theories which have been described would account satisfactorily for the observed decrease of volume. But neither of them gives any satisfactory explanation of the fact that a comet recovers its original volume as it departs from the sun's neighbourhood. Newton, indeed, put forward certain views respecting the emission of smoke from the nucleus during perihelion passage, and he surmised that the true dimensions of the comet might in this manner be veiled to a certain extent: but this part of his theory has the disadvantage of being almost unintelligible, besides being wholly insufficient to account for the regular diminution and increase which attend the approach and recession of a comet.

A theory has lately been put forward by M. Valz which accounts for the variation of a comet's volume by the supposition that the solar atmosphere exerts a power of compression, which, varying with that atmosphere's density, is most effective in the sun's neighbourhood. We know, for instance, that a balloon must not be fully inflated at first rising, because when it reaches the upper regions of air, where there is less compression, the enclosed gas expands and would burst the silk if the balloon had been fully filled at first. And certainly, on the somewhat bold assumption that the solar atmosphere extends outwards to those regions in which the observed change of volume takes place, and on the additional and equally bold supposition that comets are surrounded with a film of some sort performing the same function as the silk of the balloon (or that in some other way the substance of the comet is prevented from intermingling with the substance of the solar atmosphere) the theory of M. Valz would have a certain air of probability. Even then, however, it would be insufficient to account for the enormous extent to which the variation has been observed to proceed.

The only probable explanation of the variation in question is that which is put forward by Sir John Herschel in his admirable work on the southern heavens. During his stay at the Cape of (rood Hope he had an opportunity of observing the recession of Halley's comet, and he discusses the phenomena with admirable acumen and judgment. The result at which he arrives appears to afford a simple and rational explanation of the observed phenomena. He supposes that as a comet approaches the sun the action of the solar heat transforms the nebulous substance of the comet into invisible vapour. This action progressing from without inwards, of course produces an apparent diminution of volume. The diminution continues as long as the comet is approaching the sun, and for yet a few days after perihelion passage; but soon after the comet has begun to leave the sun's neighbourhood the transparent vapour begins to return to its original condition, the solar action being insufficient to keep the whole of the vaporised matter in the gaseous state. Thus the comet gradually resumes its original apparent dimensions.