Naturalists have agreed to designate as the cardinal teeth those solid projections which arise on the edge of the hinge. These projec tions on the one valve are for the most part 4ccompanied with corresponding depressions one or two in number, and remote from the centre of the hinge, they are named lateral teeth. Of these lateral teeth one is an terior (c, fig. 365), the other posterior (d, fig. 365). The anterior lateral tooth is com monly situated at the extremity of the lunule, and the posterior lateral tooth at the extre mity of the ligament. The cardinal teeth, properly so called, vary in number. When there are but two, the one is anterior, the other posterior ; when there me three or more, those in the middle are entitled median teeth. If the hinge be composed of a great number of teeth, it is said to be serial (b, b, fig. 366).
The teeth are commonly simple and conical ; occasionally they are flattened either lengthwise or transversely. In a considerable number of species they are grooved to different depths on their summits, and the teeth are then said to be bifid (e , .fie,r. 365).
There are other parts still which present themselves upon the cardinal edge, and of which it is important to have a sufficient know ledge,—namely, those destined for the implan tation of the ligament when it is external ; to these parts the name of nympliw is given. These form two callosities more or less promi nent, which are seen along the posterior and superior edge of the shell. When the ligament is internal, it rests upon a cavity generally pro minent towards the interior of the valves, and designated by the name of cuilleron or spoon shaped cavity. This cuilleron is generally situated in the centre (c, fig. 367) of the hinge ; sometimes, however, it becomes a little oblique, elongated, narrower, and runs in the direction of the posterior and.superior edge.
When we direct our attention to the external forms of the bivalve shells, we observe numerous modifications, of the principal of which it is necessary to take some notice. In a consi derable number of species the two valves are alike, when the shell is said to be equivalved. When one of the valves is larger than the other it is of course inequivalved ; to constitute it so it is not necessary that the shell should be irregular. A regular shell is that which at liberty always presents the valves alike in all the individuals of the species ;. ari- irregular shell is not only inequivalved, but' farther, the whole of the individuals of the same species are not exactly of the same form, and want the same peculiarities of external conformation generally. The Oysters are inequivalve and irre
gular shells ; the Corbules are inequivalve and regular shells; the Venus and many others are perfectly equivalved and regular ; the Placunes, to choose a particular example, are in like manner equivalved but irregular. The length of a shell is always calculated from the summit of the hooks to the inferior edge. All that are of greater length than breadth are entitled longitudinal, ( Mytilus, Pinna, &c. fig. 353) ; and all that are of greater breadth than length are named transversal : the breadth is estimated by a line passing from the anterior to the posterior extremity, and cutting the posterior axis of the shell at a right angle, ( Solen, Tel lina, &c. The number of transverse bivalve shells is very great : fig. 367). If the position of the hooks with relation to the transverse and longitudinal lines be considered, the shell is said to be symmetrical, when, the hooks being in opposition, the anterior segment is equal to the posterior, and of the same form in consequence of this symmetry ; a perfectly symmetrical bivalve shell might in fact be held to be composed of four similar parts ; but this perfection of symmetry, which exists in many Brachiopods, never appears among the conchifera properly so called, even those which are the most symmetrical in external character, as certain Petuncula, fig. 358, would be more correctly designated as sub-synzme trical. When the hooks are inclined to one side of the shell, and divide it into two equal parts, it is said to be equilateral. But if the hook be carried further forwards than back-, wards, so that one of the sides of the shell then becomes larger than the other, it is said.to be inequilateral. In the greater number of the conchifera the two valves of which the shell consists join each other accurately around their whole circumference, in which case the shell is said to be shut or closed. When, on the contrary, the two valves present a vacancy, between them in some part of their circum, ference, when they are approximated as nearly as possible to one another, the shell is said to be patulons. This open space is vari ously situated in different species, sometimes in the anterior surface, rarely in the inferior edge, but pretty frequently in the posterior edge, especially in those species of the class whose mantle is prolonged on this side into one or two syphons.