the hinge will permit. This admirable structure was first pointed out by Dr. Leach, in 1818.*

The muscular scars and submarginal impression are present in every bivalve shell, but the pallial sinus,+ being marked out by the foot, is of course absent in all the genera which are apodous. There is besides to be observed, in many bivalves, an impression on the posterior slope, forming one or two parallel and adjacent shallow furrows; this is called the siphonal scar, for it is impressed on the shell by the branchial and anal siphons. Hence the presence of this impression is a proof that the animal had these organs, combined or adnate when the furrow is single, and separate when the furrow is divided by a raised central line. The length of the siphons is indicated by the length of the furrows.

The inner surface of the bivalves is always smooth, glossy when fresh, frequently nacred or pearly, generally white, but in many tinted with rose-colour, yellow, orange, purple or blue. This colouring is produced, not by glandular secretion like the colours of the external surface, but by the contact of the inner layers with a similarly coloured viscus of the animal, by whose excretion it is stained. Hence the character furnished by this colouring is of little value in distinguishing genera, or even species.

The margin of the valves is variously fashioned, but there is no difficulty in understanding the discriminating terms. It is thin and acute, or thickened; even or undulated or sinuated; smooth or serrated. If the little denticles which, by their manifold and equal repetition, constitute the serrated character, are made larger and fewer, then a crenulated or crenated margin is the result; and a toothed margin is one with still larger and few projections. The margins of freshwater bivalves are never properly toothed or serrated. The only exception is found in Unio sulcatus of Lea, in which the margin approaches to the dentate character.

In the preceding explanations the Conchifera have been kept exclusively in view, but the Brachiopoda possess likewise a bivalve shell, and it is now necessary to notice a few terms more peculiar to that singular order.

The valves of the shell of the Conchifera, when viewed in their relation to the animal in its natural position, are right and left; but in the Brachiopods one valve is superior and

Gray in Zool. Journal, i. 217-220.

+ Mr. Sowerby, in his Genera of Recent and Fossil Shells, always names this the "impressio muscularis pallii,"-the "muscular impression of the mantle."

the other inferior, the superior overlying the dorsal, and the inferior sustaining the ventral aspect of the animal's body.* In a conchiferous Mollusk, on the contrary, the back looks to the superior or cardinal margin, and the belly to the inferior margin or gape.

The shell of the Brachiopods is almost always regular and symmetrical, viz. a line drawn longitudinally from the summit to the middle of the inferior margin divides it into four equal areas.

Fig. 89.

The muscular impressions are more than two and faintly marked. There are usually three or four on each valve. The mantle also adheres throughout very firmly to the inner surface, and leaves no marked impressions (Fig. 89).

When the valves are unequal, as they mostly are in this order, the largest valve is the dorsal (Fig. 89). Its apex is more or less elongated, and usually perforated. It is patelliform in the Crania and Orbicula, and more or less pyramidal in Calceola and some Terebratulæ, in which the inferior or ventral valve is reduced almost to be operculiform.

The beaks of the valves, when equal, are always opposite and look to each other; when unequal, the beak of the dorsal valve is the largest and most developed.

The beak of the dorsal valve offers to our notice certain modifications which have been used to distinguish groups in the order. It is sometimes short and almost obsolete, but, through a series of species, it is seen gradually to enlarge and to project beyond the valve, sometimes in a pyramidal shape, and sometimes more or less spiral, rounded entirely or, not unfrequently, flattened on the side of the hinge. The most important character, however, is its perforated or entire state. The perforation is either a hole or a deep sinus or notch. (See Fig. 89). Usually the perforation is on the summit of the beak, which is there observed to be truncated; but it is often found between the summit and the hinge.

"The perforated valve (Fig. 89) is the upper or dorsal one, while the other is the lower or ventral (Fig. 90); this last being usually furnished with an appendage, assuming various forms in different species, for the support of the parts of the body of the animal. When the shell is placed on the lower valve with the hole or gap towards the observer, the sides of the shell will correspond with his own."-J. E. GRAY, in Zool. Journ. i. 209.

When we examine the superior surface of the beak, we may observe that the perforation is completed by two small triangular pieces soldered together on the mesial line, and upon the margins of the beak itself. These pieces are apparent in almost all Terebratulæ,-in a rudimentary state in some species, but becoming large in those which have a greatly prolonged beak. From the shape and connection of these pieces excellent specific characters may be derived. They do not constitute of themselves the entire upper surface of the beak; and there are examples of brachiopod adherent shells, with the beak of the dorsal valve considerably elongated, which show no trace of them. This surface, to which M. de Buch applies the term area, has much analogy with the beak-like apex of the oysters and of the Spondyli. In the Terebratulæ we may also notice, on each side of the pieces of the beak, a more or less extended surface which corresponds to the beak of the Crania and the Thecidea; this surface is generally circumscribed externally by an angle, and it is further indicated by its lines or striæ of growth.

In Lingula the valves are equal; in Crania and Orbicula the ventral patelliform valve is larger than the dorsal; but in the majority of Brachiopods the ventral valve is smaller than the dorsal. Not only is it smaller, but its beak is never perforated, and it is often bent to bury itself underneath the cardinal hinge of the dorsal valve. This is the case in the greater number of Terebratula and of Productus, and in all the Thecideæ. Sometimes it is straight, as in Lingula; sometimes recurved as in some Terebratulæ; and in Crania and Orbicula the apex of the ventral valve is more or less elevated and subcentral.

But the most admirable part connected with the hinge of the shells of the Brachiopods is the testaceous frame-work called the "appareil apophysaire" by Deshayes, the "internal skeleton" by Owen, and intended to support the fringed arms of the animal, and to keep the valves open, or even to assist in opening them (Fig. 90), for in the Brachiopods

Fig. 90.

there is no cartilage at the hinge for this purpose. There is no trace of this apophysis to be found in the genera Lingula and Orbicula; but rudiments of it appear in some Craniæ; and it goes on through many states until it reaches very con

siderable complexity in the Terebratula, Productus, and Thecidea. The value of these modifications to the systematist is not great, for he cannot avail himself of them to separate the shells into natural families or even genera. In both Productus and Terebratula, there are species which have the apparatus well developed and spirally twisted, and others in which it is reduced into simple laminæ, more or less projecting. Rudimentary in certain Crania, the apparatus continues to be very simple in the greater number of the Terebratulæ, but it complexes itself in others. Two little ossicles spring and diverge from the hinge margin, and bifurcate towards their points; the lower prong projecting horizontally, meets that of the opposite side, and thus forms a primary arch, as is observable in a certain number of species; the other prong shoots more or less into the centre of the valve, and it is either simple or curved back upon itself to form a branch parallel to and above the first. Approaching the beak, the extremity of this branch or prong bends back horizontally, meets that of the opposite side and coalesces with it, forming in this way a large arch placed above the first.* In the fossil genus Spirifer of Sowerby, the osseous prong, which constitutes the large arch, is twisted into a pyramidal spire or cone. The figure (Fig. 90) will give you a good idea of the appareil apophysaire," in its fullest development.

66

On the inner surface of the valves of different genera of Brachiopods, near the middle, there may be observed a notable thickening or elevation. This appears under the form of an obtuse longitudinal rib in the thin valves of the Lingula; but in some Terebratulæ it becomes more developed. There is one in the ventral valve, and two in the dorsal valve, which are separate and divergent. In many species these crests are short and obtuse, not reaching beyond the centre of the valves; but in other species they reach the inferior margin, attain considerable prominence, and divide the shell into three equal parts. In the Calceola there is a single median crest with ruga on each side; and in Thecidea this crest is very prominent, and has on each side a large apophyseal apparatus analogous to that of the Spirifer.

he says,

Sowerby calls the appareil apophysaire, the "shelly processes," which are sometimes short, simple, and recurved, but sometimes of considerable length, branched and variously bent, and generally anastomising: these generally commence on each side of the hinge, but sometimes near the centre of the shells, and they are sometimes united to the shells at other points.""-" Terebratulu," in Gen. Rec. and Foss. Shells.

THE shells of the Cephalous Mollusca are either, 1, multivalve; 2, simply univalve; 3, spiral or turbinate; or 4, multilocular or chambered.

I. MULTIVALVES MULTIVALVIA, LIN.

A multivalve shell (Fig. 91) consists of eight pieces (scuta) arranged in a series along and across the back of the animal it covers (Fig. 91).* It is peculiar to a single family or section of Gasteropods.

The valves are connected together by the mantle forming a marginal band or zone, which surrounds the whole shell. (Fig. 91, z.)

The valves are imbricate when they touch each other along the transverse margin; and separate when they lie remote

* "Testæ plures, longitudinaliter digestæ, dorso incumbentes."-LIN

NEUS.