of another kind of ornamentation found on the Upchurch pottery, formed by parallel intersecting lines. In its shape, this vessel has much the appearance of a sepulchral urn. A considerable quantity of this pottery is without ornament at all. Among this unornamented pottery are found especially jug-shaped vessels, commonly with a handle, like that represented in the last figure in our plate. Two similar vessels are represented in our woodcut (8 and 9); in which I give also a curiously-shaped plain urn (7), and an unornamented vessel of another form (6).

At different spots over the locality which was covered by these potteries, Mr. Roach Smith has found remains which indicate the former existence of kilns, and further researches will most probably bring to light some of the kilns themselves. Traces have also been found of the residences and of the graves of the potters. There appears to have been a more extensive settlement a potters' village, or little town-on the higher ground, bordering the marshes at Halstow. "In the Halstow marshes," Mr. Roach Smith observes, "I noticed, at a particular spot, a considerable quantity of tiles and stones, which I could not positively identify as having been used in buildings; but adjoining the church, near the creek, there are abundance of fragments of tiles of various kinds, that clearly show the locality to have been the site of buildings, which, if we may judge from their débris, must have been tolerably extensive. On the sides of the church, facing the creek, an embankment has been thrown up to protect the land from the sea; this defence is filled with broken tiles and pottery, which also literally cover the shores. The church itself, probably of Saxon origin, has a large quantity of Roman masonry worked into the walls, and in a field west of the church, in the side of a well sunk for water, for purposes of brick making, I noticed a tier of Roman tiles, which appeared to be part of a hypocaust."

CHACORNAC ON THE VOLCANOES OF THE SUN.

WE have, in former numbers, made mention of M. Chacornac's solar observations at Ville-Urbanne, and of the theories which he has been led to form, and we are now indebted to him for a long paper on "The Structure and Origin of Solar Volcanoes." In the present condition of the sun, in which the production of spots is at a minimum, he informs us that the time is favourable for the study of little spots which frequently appear entirely destitute of a penumbra so long as their largest dimensions do not exceed five or six seconds of an arc, but which acquire one as they grow bigger. Referring to earlier observations, M. Chacornac informs us that he has drawings of 740 isolated spots not exceeding the dimensions just stated, and all destitute of any certain traces of a penumbra; while other spots, in closer aggregation or united by fissures, usually exhibit a portion of a penumbra generally corresponding with that part of the spot which is darkest and most deeply excavated. In this category must not be comprehended simple superficial fissures, which are themselves only isolated penumbra, and which appear of different degrees of shade from the intensity of certain nuclei to that of the general surface of the orb, or about a sixtieth below the surface. These are only variations in the level of the photosphere, and seem to occur in the most external layers of the immense luminous envelope of the sun. The causes which produce them appear to reside in the superficial layers, as it is only changes in the layers which underlie the photosphere that can give rise to such phenomena, and this fact, taken in connection with the rapid changes of form that are noticed on the solar surface, leads to the conclusion that we have to do with an immense atmosphere enveloping a central nucleus.

The phenomenon which should occupy the foremost place in any hypothesis is without doubt the rapidity of the precipitation of spots one in the other. Hence comes the relative movements of groups in different latitudes, and the appearance of fusion which characterises the formation of spots. All astronomers, for example, might remark the regular form of the great spot which entered the visible hemisphere during the night of the 7th-8th of last July, and the persistence of this form, which was nearly circular and without diminution, all the while the spot appeared, which was up to the 20th July at 6 P.M.; and they could also notice the irregularities of shape and the rapid changes exhibited by a single group which occupied the centre of the disc on the day last named.

The very rapid changes in this last group, M. Chacornac

thinks not consistent with the supposition of a liquid medium. It was evidently a cloudy or gaseous one, able, by sudden condensation, to form a relative vacuum, and occasion the precipitation of fresh atmospheric layers upon the photosphere, and the consequent formation of spots. Spiral gyratory movements indicated this kind of action. Moreover, the spots exhibited a disposition more or less striking to form gulfs, into which the photospheric matter descended.

In order to have superficial currents in a fluid spherical mass, it seems necessary that there should be a break in the decreasing density of its layers, and that there should be an abrupt transition, as in the case of the atmospheres of Jupiter, Saturn, or the Earth, which rest upon a solid surface, and likewise a cause for the variation of the temperature of its layers. From these and other considerations, M. Chacornac concludes that it is an atmosphere with which our observations have to do.

The nuclei, in which great and feebly-luminous depths are seen, disappear as if by evaporation-they dissipate themselves like atmospheric clouds. The same phenomena are exhibited in deep as in superficial layers, and who does not see in this order of facts an explanation of the immense cavities into which faculæ are precipitated in torrents. Nothing is more simple than these fusions and rapid falling of little spots into big ones. In the region in which a vacuum is made, that is to say in a spot, the cloudy layers are precipitated in the liquid form on to the surface of the central body, or perhaps these vapours in a vesicular state are vaporized and dissolved so as to occasion a vacuum in which the superior or adjacent layers are swallowed up.

M. Chacornac observes that the descending currents in these cases must have great force to draw with them, as he has seen, matters possessing a horizontal velocity of 550 mètres in a second.

The mobility of the spots and the rapidity of their changes are regarded by M. Chacornac as opposed to the conception of a resisting medium, and still more so to rents in a solid crust.

When a great spot is forming, all the surrounding parts are dragged into the gulf, into which also fall adjacent spots of small dimensions.

If a spot is solitary it assumes the regular form of a funnel, and the flocculent matter is distributed in channels round the whirlpool orifice (orifice turbiné).

M. Chacornac does not deny the existence of ascending currents, though what is observed only proves the operation of descending currents. All spots, for example, are cavities

with the narrow orifice downwards, and the forms of streams and ripples (sillons) of flocculent matter always indicate a descent of the upper layers into the regions below.

When a funnel-shaped gulf is formed by the sinking down of the atmospheric strata, there is a disturbance in the equilibrium of the flocculent masses, precipitation of adjacent layers into the orifice, and accumulation of vaporous matter at this part of the atmosphere, as is shown by the faculæ heaped up in the vicinity of a dark cavity. It seems as if this affluence of a vaporous fluid gave rise to a heaping up (bourrelet) of atmospheric masses round a dark spot. At the base of these mountains a fusion takes place, and when the changes are very rapid, lines of dislocation are observed, circumscribing the annular mountains, and rendering them top-heavy, so that they fall and give rise to a penumbra.

M. Chacornac describes a case in which a small spot approached a large one and fell into its penumbra, without lessening it or occasioning any other change than making a new orifice in its perimeter; and he likens the currents of photospheric matter to the cascades of Niagara, in which the form is preserved. After the examination of this fact, he affirms it to be impossible to regard the spots as occurring in a liquid medium, for he says we should ask how such masses of liquid could be volatilized without occasioning an immense conflagration when they were converted into gas? Immense photospheric clouds dissolve in a few hours, though their volume may be greater than that of the earth.

M. Chacornac considers that the hypothesis of solar volcanoes arising in a liquid medium cannot be maintained, as the phenomena can only be explained on the supposition of rapid currents occasioned by the formation of a vacuum. A central body in a liquid state would not, he thinks, be in contradiction to the general phenomena exhibited by the sun. It would be the source from which emanates that cloudy mass which constitutes the solar atmosphere of which the limiting layer is in a state of lively incandescence. According to this hypothesis the temperature decreases from the circumference to the centre, as the law of densities augmenting in the same direction demands. Varying degrees of heat and density in different strata, and the effect of the sun's rotation, would give rise to currents such as are seen in the trade winds, in the belts of Jupiter, and in the motions of sun spots, provided that the atmosphere is situated upon a solid body.

M. Chacornac thinks that the sun may consist of a central liquid covered with a solid pellicle, and surrounded by an atmosphere composed of several layers, which in their normal condition appear in contact and united. The pellicle,

or crust, is ruptured by volcanic action, and eruptive currents dissolve the photosphere.

This paper is an exceedingly difficult one to give an account of; but we have endeavoured to present our readers with its prominent ideas.

THE EXHIBITION OF MINIATURES AT THE SOUTH KENSINGTON MUSEUM.

BY W. M. ROSSETTI.

(Concluded from our last.)

In resuming our notice of this Exhibition, necessarily very far from exhaustive or complete, we come first to the

BRITISH PORTRAITS OF THE STUART PERIOD AND THE COMMON

WEALTH.

Lady Arabella Stuart was, and deserved to be, frequently painted. In No. 486, by I. Oliver, she is not so pretty as in some other portraits, and tends to the Mongolian type of visage.-No. 1580, by the same, was painted at a somewhat later period of life than most of the portraits of this beautiful and ill-treated lady; here she has a melancholy look, instead of the blooming and exuberant air of the earlier likenesses, and one can guess that the iron has entered into her soul.-In No. 2169, by the same, "she is represented with her long auburn hair hanging on her shoulders, wearing a close lace falling collar or ruff, a pearl in one ear, and a black ribbon in the other."-A fourth likeness, by the same, "with a jewelled anchor as an ear-ring," is very charming.

66

Elizabeth Stuart, Queen of Bohemia, by the same, is represented young, wearing an open lace ruff, over which her light auburn hair falls upon her shoulders and down to her waist;" a very agreeable portrait.-P. Oliver also has painted a fine miniature, the face not unlike that of the greater Elizabeth, "Good Queen Bess."

Lady Shirley, by Hilliard, "represented with her hair falling loose on her shoulders, and wearing a wreath of oak leaves," was a charming woman, who pretty evidently knew that fact, without losing the grace of being natural.-I. Oliver's version is also very handsome.

Frances Howard, Countess of Essex, ascribed to P. Oliver, an excellent portrait of Sir Thomas Overbury's poisoner, shows us a fine woman, with an open, unembarrassed counte