The cephalopodous Mollusca form no exception; their bone, composed of two or three calcareous plates, being found fully developed in the egg of the Cuttle-fish some time before the young animal is hatched.

These observations are directly at variance with the theory maintained by the late Sir Everard Home,* viz., that the shell of the Vermes Testacea is formed after the animal has quitted the egg; and as regards the Cuttle-fish, they are opposed to the remark, made by the Baron Cuvier, that the young Cuttle-fish, when first hatched, has only a cartilaginous plate like the Loligo.

The shell when first observed on the embryo (even of the animals of spiral shells) forms a short, blunt, more or less curved, subcylindrical cone, covering the hinder part of its body as the organization of the embryo becomes developed, and the hinder part of the body extended, the shell increases in size, till the body and shell together occupy nearly the whole of the egg. While inclosed in the egg, the embryo shells are generally of a pale horn colour, and destitute of markings: when, therefore, they remain attached to the apex of the spire of adult shells, they may be easily distinguished by their appearance from the part formed after their exclusion; and as, in such cases, they offer some characters of importance, it has been proposed to designate them by the name of the nucleus of the shell.

The effect of the atmosphere on the shell is almost instantaneous in some young Helices and in a species of Voluta in my collection, the very first line of calcareous matter deposited after their exclusion from the egg is marked nearly as the adult shells of the species.

The nucleus may be generally distinguished from the perfect shell by the rapid enlargement of its whorls, by its extreme tenuity, by its want of colour, and by the great obtuseness of that part which is earliest formed and constitutes the extremity of its first volution. It is necessary to pay attention to these particulars, inasmuch as the nuclei of some large species have been mistaken for full-grown shells, and vice versa. Thus, the Murex decollatus of Pennant is the just-hatched shell of the Fusus despectus; Risso's genus Orbitina is established on the nuclei of two land shells; and the genus Vitrina was regarded by Montagu as the nucleus of the common snail. In some instances the first half-whorl of the nucleus (the part first formed on the embryo), instead of being regularly curved, is bent across the tip of the other

* Philosophical Trans. 1817, p. 229.

whorls, as in the Pyramidellæ, or placed in an oblique position with regard to the succeeding one, as in Voluta papillosa and some other species.

The nuclei of many shells of different genera have not the same characters as their parent shells; thus, the nucleus of the Tritons has a short anterior nick instead of an elongated canal, and is very like a minute Buccinum. Some retain the generic, but not the specific character of the group to which they belong; thus the nuclei of the Volutes in general have the pillar slightly plaited, but the young of Voluta musica has only two or three plaits on its pillar, while the adult has many.

The nucleus forms the original apex of the valves of all shells, whatever may be their form, and frequently remains attached to them during all their periods of growth; this is particularly the case with the Volutes, in which, from its large size and rounded shape, it has been called the Nipple. It is also conspicuous on most of the species of Dolium and on some Fusi, among the univalves; and on the apices of the valves of the Cyclades, and Chamæ, among the bivalves.

This part of the shell has not received the attention that it deserves. It is largest in those shells the animals of which are viviparous; and is consequently very distinct in the Volutæ, Paludinæ and Cyclades. In the oviparous species it agrees in size with the egg of the animal; thus Achatina octona, which has an egg nearly equal in dimensions to the mouth of the shell, and Bulimus ovatus and B. bicarinatus, which have large eggs, have large nuclei, the magnitude of the nucleus in general rendering the top of the spire blunt. Some shells on the contrary, those, for instance, of the genus Stylina, generally, and of the Pupa purpurea, have a very long, slender, acute, turreted nucleus, but the form and size of the eggs are in these cases unknown.

The nucleus is found on examination to consist of two very distinct parts or coats, the outer of which is membranaceous or horny, and called the Periostracum, and the inner hard and calcareous, and constitutes the shell.

These two coats may be observed in all the stages of the shell they are generally very thin in the nucleus, and the outer one is rarely distinctly visible in that state; but it is to be clearly seen (covered with five or more bands of hairlike processes) in the very young Paludinæ. In such shells as are enveloped in the mantles of their animals, as the Dolabellæ, Aplysia and Bullææ, the outer coat or periostracum is very thin; it is, however, to be found in all shells, and

may also be observed on the shelly plates of the Cirrhipedes. In some instances, as in the Cowries and Melons, the outer coat of the shell is covered in the adult age with a deposition of shelly matter, which entirely conceals it from view. A few shells, as for example those of Loligo and Aplysia, contain so little calcareous matter, as to appear to be formed entirely of periostracum.

2. The External Form of Shells, and their Variations.

Each valve of a shell, according to the manner in which it is first formed and subsequently increases, is a more or less depressed or lengthened cone. The apex of this cone is always oblique: in all the shells with which I am acquainted, it is excentric; and in most of the univalve shells, whether they be simply conical, involute or spiral, it is directed from the head of the animal towards the hinder part. The only exception, as far as I know, to this rule, occurs in the genera Patella and Lottia, in which the apex is directed from the hinder part towards the head; and this is the more remarkable, as in the Chitons, the animal of which so much resembles that of Patella, each of the valves takes the usual direction. The similarity of direction in the two genera above named is still more curious, as their animals bear scarcely any resemblance to each other.

The nucleus of the bone of the Cuttle-fish and of the Loligo, is placed in the same direction; for it is the conical process at the end of the bone of the Cuttle-fish (called Beloptera, when found in the fossil state), which must be regarded as the nucleus of these shells. If, however, the relative position of the animal of the Nautilus, the anatomy of which has been admirably described by Mr. Owen, be correctly assigned by that author with respect to its shell, it must offer a similar anomaly with the genera Patella and Lottia. The shells of the Pteropods, as for example Hyalæa, Cleodora and Vaginella, take the same direction as the other univalves; and it was this circumstance that gave rise to the supposition that M. de Blainville, in his figure of the animal of Cymbulia, had placed it in the shell in the wrong position. The numerous specimens which are now in European cabinets have proved the accuracy of this supposition.

In bivalve shells the apex of each valve is always placed on or near the dorsal or hinge margin, varying its position on this part in the different groups. Thus, in the Pectines and other suborbicular shells, which having a very large

subcentral posterior adductor muscle, were called by Lamarck Monomyaires, the apex is generally in or near the centre; while in most of the other genera it is placed more or less towards the anterior extremity of this margin, and is sometimes incurved.

In some of these shells the apex is spirally twisted, and the spire becomes more developed as they increase in size. Now this could not take place if the valves remained inseparably united together at the same part of the dorsal margin; but it is provided for by the hinge of the shell being gradually moved backwards on the edge of the valves, the ligament separating in front of the hinge into two parts, one of which diverges along each of the umbones, and forms a spiral groove down the suture of the whorls. In Isocardia the umbones seldom make more than half a turn, but in one specimen of Chama in my collection they have made an entire revolution, and in another a revolution and a half. The valves of these shells being unequal, the spiral part of the lower or attached valve is produced into an elongated cone, while in the other it is depressed, and simply marked with a spiral groove, like that of an operculum.

In most bivalves the hinge-margin, which is deposited by a part of the mantle extended behind and between the teeth, increases in size much more slowly than the other margins of the shell; but in some free shells, such as the Arcæ, this part increases nearly as rapidly as the rest, and the umbones thus become separated from each other by a lozenge-shaped disk. In others which attach themselves to foreign bodies, as the Spondyli and Ostreæ, the hinge-margin of the attached valve only enlarges, forming a triangular flat-topped process, while that of the upper valve is scarcely increased in size. Thus the cavity of the shell, as the growth proceeds, gradually retreats from the part by which the attachment first took place.

The direction followed by the whorls in passing down the axis derives its origin from that which the shell takes in the egg; and is probably dependent upon the direction in which the embryo rotated whilst enclosed therein. In most shells they turn from left to right, and the mouth is on the right side of the axis, when the shell is in its natural position; but in others, which are called sinistral or reversed, the whorls are twisted in the contrary direction. The sinistral direction appears to be constant in many species, especially among the air-breathing Mollusca; in all belonging to the genus Clausilia, among the land ones; and in all the Physæ, Planorbes and Ancyli, among those which inhabit fresh

water. But besides these entire genera, the shells of which are invariably sinistral, there are numerous species of Bulimus, Partula, Pupa, and Chondrus, that are uniformly so twisted; and there are even some that are sometimes twisted in one direction and sometimes in the other, as Bulimus lyonettianus, B. aureus, &c. Among the marine shells the sinistral direction is much more rare, although there are a few species, such as Fusus sinistrorsus of Deshayes, F. contrarius and F. sinistratus of Lamarck, and some species of Cerithia, which are constantly so contorted. so contorted. The Pyrula perversa is as often found twisted in one direction as in the other, and its shells have even been considered as different species according to the direction of their whorls. Of other marine species liable to the same variation, I have observed Buccinum undatum, which is not uncommonly found reversed; Turbinella napus, the reversed variety of which is much sought after by the Chinese; Oliva oryza, Nassa reticulata and N. thersites; but there are some genera in which I do not recollect to have noticed its occurrence; as, for example, the Cyprææ and Ovulæ.

A bivalve shell is composed of a dextral and a sinistral valve, united together by a ligament. When the two valves are separated, and spread out on a table with the umbones above, and the front end towards the observer, the valve to the right (the left when on the animal and in its usual walking position) resembles a dextral, and that to the left a sinistral, very depressed spiral shell. This is well illustrated by comparing the left valve of an Isocardia with a Concholepas. In some very rare instances these shells are also reversed, but the fact is not easily observed, except in the unequalvalved kinds. There were formerly in the Tankerville collection two specimens of Lucina childreni, in one of which the right valve was a dextral shell, in opposition to the general structure. A much more remarkable variation is to be observed in some of those bivalve shells whose under valve is attached to foreign bodies; thus, for example, most of the Chamæ are attached by their left valve, but some species, such as Chama lazarus, are frequently attached by their right valve, under which circumstance the teeth proper to the left and usually attached valve are transferred to the right, and vice versâ.

The equality or inequality of the valves of bivalves appears to be dependent on the habitual position of the animal. Thus all the genera whose animals bore perpendicular holes in rocks, like the Pholades; or bury themselves in the mud of rivers, as the Uniones; or in the sand of the sea-coast, as