hibits much of the character of electric cumulus in its lower parts, having also a considerable volume and rotundity of outline.

The year 1872 will be long remembered both for its excessive rainfall and the number and destructive violence of its thunderstorms. It was equally remarkable for the variety and striking character of its cloud forms. The electric cumulus was scarcely absent for a day during the whole period over which the thunderstorms extended. On one occasion, when leaving Liverpool for Blackburn, I observed three or four conical piles of the cloud upon the horizon in the direction of the latter town: the weather was beautiful at Liverpool, but a storm, as I learnt on arriving at my destination, had raged there at the time I observed the cloud so many miles away.

There can be little doubt that this phenomenon may often be seen at a distance of 50 to 150 miles from the point where the storm breaks.

Another remarkable form that appeared during the summer of 1872 was the "Festooned," or Pocky cloud, as some have termed it. Professor Poey a year or two ago, alluded to it as a "new" variety. Since then, however, it has been shown to be of ancient date, and a somewhat elaborate and interesting account of earlier observations was contributed by Mr. Scott to the "Quarterly Journal of the Meteorological Society," April 1872. I have myself observed the cloud on several occasions, though only once or twice in perfect form. The illustration given by Mr. Scott in his paper was furnished by Dr. Clouston. The droplets or festoons which form the lower outline of the cloud have a semi-elliptical or egg-shaped form, were very dark, and extended over a considerable portion of the heavens. In the most perfect example noted by me (about twelve years ago) the pendulous processes or droplets had an almost perfectly 'circular outline, even, and well defined. The festoons were about 4° or 5° in apparent diameter, their altitude being about a mile. The whole mass from which the droplets depended was not large, and moved with the wind; the curves meanwhile retaining their form intact. What made the case most singular was, that the regularity of the outline could be maintained, even a few moments, under the variable atmospheric conditions of the time. Though the outlines were so regular, the surface of the cloud had a very nimbus-like appearance, and was of a lurid yellow tint. The sky was not overcast at the time.

The more usual concomitants of this form are, however, a dark turbid sky and a "pallium" of cirro-cumulus subsiding into nimbus or nimbo-stratus. Under these conditions I observed it twice during the stormy summer of last year, one of the cases being immediately after a thunderstorm. The fact of

its being emphatically a foul-weather cloud adds interest to the investigation, for heavy rain and wind usually follow its appearance. My own observations indicate, however, that rain is more certain to follow than wind. In Lancashire it is sometimes termed "Rain-ball."

The origin of these symmetrical cloud forms is very difficult to determine. Mr. Scott, in his paper above referred to, attaches considerable importance to the experiments of Sir J. Herschell and Mr. Jevons, by whom it has been shown that a similar appearance may be produced by gradually mixing two fluids of slightly different specific gravities.

The nature, however, of the forces which determine the peculiar symmetry of the forms in question seems almost equally obscure in the case of the experiments as in that of the atmospheric phenomenon they are used to illustrate. In the spherical forms observed by myself it is hard to imagine such an appearance being caused by a mere undulatory movement of the vapour arising from gravitation. We are rather reminded of the formation of bubbles; but if that be an analogous case, the tension of the vapour must be something unusual. Let the reader, however, compare this phenomenon with the peculiar rounded prominences that occasionally appear on the sides of a large mass of the electric cumulus, and range themselves in regular lines proceeding from a common origin or vertex. Must we attribute these protuberances to an electrical decomposition of vapour?

363

SEXUALITY IN PLANTS.

BY MAXWELL T. MASTERS, M.D., F.R.S.

[PLATE CIII.]

IT

T has long since become unnecessary to offer any argument or illustration in support of the doctrine of sexuality in plants. Among the so-called flowerless plants even, the evidence of the existence of distinct sexes is now so overwhelming that there are few who hesitate to believe in its occurrence, even in those cases where only one of the two sexes has up to this time been demonstrated. Linnæus, to whom, most of all, we owe the promulgation of the doctrine, though he was by no means the first to propound it, proceeded to establish it by inductive, circumstantial evidence. He brought forward comparatively little in the way of direct proof or absolute demonstration, even among flowering plants, while the actual existence of sexes in the so-called flowerless plants has been questioned up to within quite recent times. It is a singular fact, however, that the whole process of fertilisation has been, generally speaking, inore thoroughly and completely demonstrated among the so-called Cryptograms, long considered as a sexual, than among the more highly developed flowering plants. It is curious also to remark the different views now taken as to the process from those which were held originally. As soon as it was clearly perceived that the stamens and the pistils, or their contents, were the essential agents in the process of fertilisation, it was naturally surmised that the pistil of every given flower was fertilised by the pollen of the same flower. Under this supposition many curious contrivances which were observed were at once set down as so many aids and promoters of selffertilisation.

Sprengel was one of the first to show the fallacies of these observations, and to demonstrate the frequent existence of cross-fertilisation, whether effected by the agency of winds or by pollen-carrying insects.