The difference in the appearance of dew, when deposited on tin and on glass, is sufficiently remarkable to arrest attention, not only when the moisture remains uncrystallized, but also when it is frozen. In an example that occurred of the latter case, a decrease in the magnitudes of the frozen particles could be traced from its edge to the dry and unfrozen margin surrounding a parcel of wool, placed on the middle of the plate, as represented in fig. 13: the appearance of the frozen atoms partaking, in some degree, of the lustre of the tin. The parcel of wool, in the interval from nine P. M. to midnight, gained four grains of moisture; and from the last mentioned hour, to six the next morning, thirty-two grains; thus gaining, in a double time, an eight fold quantity of moisture. The wool was frozen to the tin; and when the rays of the sun fell on the metallic surface, the crystalline particles became detached from it, and were readily collected together. The dew deposited on the glass presented an irregular fibrous appearance, its colour partaking of the greenish hue of the crystal. The icy particles on the tin were first deposited as dew, and frozen before they had collected in sufficient numbers to run into each other, and form an uniform crystalline surface. But the dew on the glass being formed at an earlier period of the night, a sufficient quantity was deposited to cause the particles to mingle with each other, and thus to present to the action of the freezing temperature a wide spread surface of water. The unequal action of the glass, combined with the law which regulates the crystallization of water, communicated to the frozen surface of dew, the fibrous and irregular character represented in fig. 12. Soon after the solar rays had impinged on the glass, filaments of ice were detached from both its surfaces, that from the upper side being much the thickest.

Plymouth, December 12, 1823.

ART. II. Description of Two New and Remarkable Fresh Water Shells: Melania setosa and Unio gigas. By William Swainson, Esq., F.R.S., L.S., M.W.S., &c.

[Communicated by the Author.]

THE attention of several conchologists has been excited by a new and most extraordinary fluviatile shell, belonging to the genus Melania, recently brought from the Mauritius. Having been favoured with its examination, I now lay before the public the following description of the shell, drawn up from the only specimen which its discoverer, Mr. Warwick, was able to procure, after diligent and often repeated searches in the same locality. I take this opportunity also of recording the characters of another freshwater shell of gigantic dimensions, equally unknown and interesting to naturalists.

MELANIA. LAM: CUV.
Specific character.

M. testâ ovatâ, ventricosâ, spinis tubularibus seta bina porrecta basi connexa emittentibus coronatâ. Shell ovate, ventricose, coronated by tubular spines, each sheathing the base of two protruded horny bristles.

DESCRIPTION.

Length, one inch two tenths, of which the spire occupies very nearly one half. In habit the shell resembles Melania amarula, (Helix amarula, Lin.) but the basal volution is more ventricose, the spire more conic, and the tip acute; it is also much thinner, and may be termed subdiaphanous; the whole shell is covered by an olive brown epidermis; the spiral volutions are angulated, and marked by from three to four transverse elevated striæ; the basal volution is without any indication of plaits, but is slightly impressed by narrow, transverse grooves, which are wide apart; these are crossed by very delicate and close-set longitudinal striæ; but whether these last are only external and belong alone to the epidermis, could not be ascertained without injury to the specimen. The most extraordinary characteristic of this shell, I shall now proceed to detail. On the upper part (or shoulder as it is sometimes called,) of the body whorl, is a row of coronated spines,

perfectly tubular; these spines are very thin, and are placed parallel with and very near to the aperture; their summits are obtuse and their length variable, probably owing to some having been injured through their great delicacy; the longest measured nearly oneeighth of an inch; from the summit of each spine emerges two. stiff erect acute bristles; closely adhering together, and projecting about two-tenths of an inch. The colour of these bristles is black, their surface polished, and their substance horny. They likewise possess some degree of elasticity, being easily bent by a slight pressure applied laterally; although I doubt whether they would have sustained such pressure had it been applied horizontally. These bristles it will be perceived, are completely sheathed at their base by the tubular spines, but these latter are so thin that the lower part of the bristles are distinctly seen through them; rooted, as it were, in the substance [of the shell. I know not, positively, whether each spine contains two distinct bristles; or only one, forked or divided at about half its length, as this fact could only be ascertained by removing one of the spines, and tracing how far the division extended; but that portion which forms the lower half (and is enclosed within the spine) is so thick, as to favour the supposition of their being in pairs. These spines are continued round the middle of each volution of the spire to its apex; but they are more remote, and the bristles much shorter, than those on the body whorl; sometimes, indeed they hardly project beyond the spines. The direction of the whole is slightly incurved. The aperture is pale; and, at the top of the outer-lip, is an indented sinus similar to that seen in M. amarula, Lam.

Ob. 1. The extraordinary appearance of bristles protruding from the spines of a shell, a formation altogether unprecedented amongst this class of animals, might naturally excite, in some minds, a suspicion that it was an ingenious deception. But this idea, I think will be abandoned, when the peculiar construction of the spines are well considered. In the genus voluta, we have many instances of shells being crowned with thin, vaulted spines, but no example can be produced, of such coronated spines being tubular; or completely closed in their circumference, and pervious only at their

summits. Now it is obvious, that this peculiar form, is of all others the best adapted to strengthen and protect the elastic bristles which they enclose: both appendages, therefore, are in unison with each other, and leave not a doubt in my mind, (setting aside the personal testimony of its discoverer) that the whole shell is in a perfectly natural state.

It is difficult to conjecture in what way the formation of the shell accords with the economy of its inhabitant. We know that testaceous mollusca, are the food of several kinds of fish, both marine and fresh-water; may not these bristles be intended by nature to defend the animal from such enemies? they would certainly be very repulsive to the lips of any fish; and in all probability would penetrate, as deep as possible into the skin. The weapons of protection or of defence with which nature has furnished different tribes of animals, are as various as they are wonderful. In the testaceous mollusca, they are confined alone to the shelly covering of the animal, who, as long as his castle is armed and entire, with draws into its walls, secures the entrance, and remains passively

secure.

Ob. 2. Since the above was written, Mr. Broderip informs me, another specimen of this shell has come into his possession: "carefully cleared, and every vestige of bristle removed, the hollow coronations remain."

UNIO GIGAS.

Specific character.

U. testâ ovato-oblongâ, depressâ, anticè alatâ et sulcis obliquis, divaricatis subradiatâ; posticè brevissimâ; dente laterali (utriusque valvæ) solitario; umbonibus brevibus, retusis.

Shell ovate-oblong, depressed anterior side winged and marked by oblique grooves in different directions; posterior side very short; lateral teeth, one in each valve; umbones small, retuse.

DESCRIPTION.

This is truly a gigantic shell; far exceeding in size any other yet discovered as inhabiting the fresh-water, and presenting characters which leave no doubt that it has hitherto remained unknown to all conchological writers. Its extreme length is rather more

than eight inches and a half; and its greatest breadth (from the ligamental to the basal margin,) five inches three-quarters.

Its form is a broad oblong-oval; obtuse at both extremities; the anterior of which is broadest and sinuated, and the posterior rounded, and so very short as to project only three quarters of an inch beyond the outer side of the cardinal teeth. The whole shell is remarkably flat but very thick in substance; and the umbones, which are unusually small have scarcely any convexity; the ligamental margin is dilated, winged, and forming in its dilation nearly two equal sides; the horny part of the ligament itself, (with the internal plate that supports it,) extends half way between the umbo and the extremity of the wing; the exterior colour of the epidermis, is dark brown, but the umbones (in this specimen) are decorticated for a considerable space around them. The sculpture of the anterior part of the shell is very peculiar; it consists of four series of short oblique grooves, or of indented wrinkles, three of which are arranged in a direction with the umbonial slope*, the other is transverse; the first of these series consists in parallel grooves which cross the wing obliquely from left to right. The next is a range of broad and sinuated indentations, wide apart and having the same inclination as the former; the third range occupies the umbonial slope, and is formed by narrow sulcated grooves, placed nearly in a horizontal direction, and diminishing in length as they approach the umbones. The fourth and last consists of several transverse grooves situated near the basal margin, and the whole presents an appearance as if the shell had been indented, in various di rections, by some blunt instrument.

The inside is pearly, white, tinged with flesh-coloured purple, and stained (as is frequently the case, in fluviatile bivalves) with olivaceous yellow spots; in a perfect state of the shell the colours, probably, would be more brilliant.

The cardinal teeth are strong; deeply and irregularly striated and are obliquely transverse; in the right valve are two, and in the left valve, one; in each valve there is only one lateral tooth, a very unusual and discriminative character for the species; this *See Zoologiæ Illust.