1. Family BALANIDÆ, (or Sessile Cirripedes).

Cirripedia without a peduncle; scuta and terga furnished with depressor muscles; other valves united immoveably together.

TABLE OF CONTENTS.

Page

Structure of shell34

”of the individual compartments43

”of the radii45

”of the alæ47

”of the sheath48

”of the basis49

”of the opercular valves (scuta and terga)51

Growth of whole shell and microscopical structure54

Muscles of sack61

Branchiæ63

Thorax and body65

Muscular system68

Movements and muscles of the cirri71

Mouth74

Cirri81

Caudal appendages85

Alimentary canal85

Circulatory system87

Nervous system88

Eyes and vision93

Acoustic organs95

Olfactory sacks97

Male organs of generation97

Female organs of generation100

Metamorphoses and homologies102

Larva, first stage103

Larva, second stage109

Larva, last or pupal stage110

Act of metamorphosis126

On the homologies of the carapace131

Cementing apparatus133

Affinities, classification, variation152

Rate of growth, exuviation, &c.156

Geographical range and habitats159

Geological history172

Almost every one who has walked over a rocky shore knows that a barnacle or acorn-shell is an irregular cone, formed generally of six compartments, with an orifice at the top, closed by a neatly-fitted, moveable lid, or operculum.[19] Within this shell the animal’s body is lodged; and through a slit in the lid, it has the power of protruding six pairs of articulated cirri or legs, and of securing by their means any prey brought by the waters within their reach. The basis is firmly cemented to the surface of attachment. The whole shell, basis, and operculum consists, as we have already seen, of the first three segments of the head, modified into a singularly constructed carapace, which encloses the mouth and rest of the body. The anterior extremity of the shell is situated in the centre of the basis, where indeed, by due care, the antennæ of the pupa may be always detected; and the posterior extremity is directed vertically upwards.

Structure of Shell.

When the shell of a sessile Cirripede or barnacle, for instance, of a Balanus, is first examined, the structure appears extremely complicated; but this can hardly be considered as really the case. The structure will, I think, be best understood by recalling to mind that of Pollicipes,—the oldest known genus, from which, in one sense, all ordinary Cirripedes, both sessile and pedunculated, seem to radiate. I must premise, and the fact in itself deserves early notice, that the homologies of the several parts in the pedunculated and sessile Cirripedes admits of no doubt,—that is, if amongst the pedunculated, the genus Pollicipes, or certain species of Scalpellum, be taken as a standard of comparison.[20] The peduncle corresponds with the basis, as may be clearly seen, if a Pollicipes with a short peduncle, and a Balanus, with a deep cup-formed or cylindrical basis be compared, for the contained parts are similar, and both grow at their upper edges upwards and outwards. Secondly, the valves round the lower part of the capitulum of a Pollicipes, though generally much more numerous, and forming more than one whorl or circle, and not so closely packed together, answer to the compartments forming the shell of a sessile Cirripede; this is shown by their lateral and downward growth, by their upper ends generally projecting freely above the cavity in which the animal’s body is lodged; and in the case Pollicipes mitella, by an actual resemblance in outline, some being triangular, some broad at the upper end, and some sub-rhomboidal, and, lastly, in the manner in which they slightly overlap and indent each other: moreover further resemblances in the relative position and even in the size of the several valves, will hereafter be pointed out between certain sessile genera amongst the Chthamalinæ and certain genera of the Lepadidæ. Thirdly, the scuta and terga in Pollicipes, so strikingly resemble in manner of growth in position relatively to the animal’s body—in shape—and even in being articulated together, the valves which form the operculum or lid of sessile Cirripedes, that their identity is at once obvious.

It may be well here further to premise, that apparently none of the sutures in the shells of Cirripedes correspond with the articulations between the three archetype cephalic segments, of which the whole shell is formed; or with the eight elemental pieces, of which each separate segment in the archetype crustacean is known to consist. But, as I believe, the several valves in the shell of a Cirripede are homologous, or at least analogous, with the sclerodermic plates,[21] of which the carapace of the Podophthalmia is formed; with this difference, that in the latter they become, after their first formation, united together into a single piece, and are thus moulted as a whole; whereas in Cirripedes, the valves or sclerodermic plates are not moulted, but continue to be added to throughout life.

In Pollicipes, there is no difficulty in understanding the growth of the lower valves of the capitulum, especially if a species be taken in which these valves stand a little way apart: at each period of growth, they are added to at their basal edges and a little way up both sides; at the same time, a new membrane connecting them together is formed, the old membrane disintegrating, or being left hanging in tatters to the last zone of growth. Now if we look at the shell of a sessile Cirripede, there is no essential difference in the growth of the compartments or valves; all grow downwards and laterally; but they overlap each other much more laterally than in Pollicipes, and the connecting membrane is in most parts reduced to a mere film jammed in between the valves; but, in the case of the opercular membrane, it still remains wide, and is periodically moulted.

In the annexed woodcut (fig. 1), of the rostrum of Balanus Hameri, the downward growth and the lateral growth on both sides is plain. The modified sides (rr) for convenience sake, have been called the radii; they invariably overlap the adjoining compartments. The middle part (p), has been called the wall, or parietal portion: in the specimen figured, the walls and radii are distinctly separated, but in some cases, especially amongst the Chthamalinæ, the lines of growth are absolutely continuous from one to the other. In fig. 2 of a Lateral compartment of the same Balanus, we have the same essential structure; but the left side (a) is more protuberant, and is hollowed out in its lower half; it is, also, more distinctly separated from the parietal portion: this side has for convenience been called the ala; it is invariably overlapped by the adjoining compartment: in some few cases, as in Pachylasma, the ala is not hollowed out in its lower part, and from being added to in a straight line along its whole edge, with the lines of growth continuous with those on the wall, it differs hardly at all in appearance from a radius. Lastly, in fig. 3 of the carina, or compartment facing the rostrum, we have alæ (aa) on both sides; these being, as in all cases, overlapped by the adjoining compartments.

Fig. 1.

Fig. 2.

Fig. 3.

p, p, Parietes; r, r, Radii; a, a, Alæ.

Fig. 1, Rostrum with two radii, serving in the Chthamalinæ for rostro-lateral compartments.

Fig. 2, always serving for lateral and carino-lateral compartments.

Fig. 3, Carina, serving in the Chthamalinæ, also, as a rostrum.

Now, the compartments in the shell of every sessile Cirripede, are without exception constructed on the above three simple patterns. In number, they are 8, 6 or 4, or all confluent together.

Considering this simplicity in growth and form of the separated compartments, it seems at first surprising that the construction and enlargement of the whole shell in Balanus, should long have been viewed as a difficulty. But the radii, from growing against rectangular indentations, or rather furrows, in the opposed compartments, come to be set a little inwards; and their external surfaces assume a very different appearance from the wall-portions of the compartments, which grow against the surface of attachment. In different species, the summits of the radii (and of the alæ) grow either very much more obliquely than in the species figured, or more squarely—that is, they extend from tip to tip of the adjoining compartments, parallel to the basis. In this latter case, and when the surfaces of the radii differ considerably in appearance from the walls, as in Balanus tintinnabulum (Plate 1), I am not at all surprised that the radii should have been described as separate elements, and called “areæ interjectæ,” or “compartments of the second order:” for the shell of this Balanus seems to be composed of six wedges with their points upwards, namely, the parietal portions of the compartments, and of six other narrower wedges, the radii, with their points downwards; and the fact that these latter wedges consist simply of the sides of the parietal portions, modified by growing against the adjoining compartments, is completely masked. I should add, that sometimes the radii are not developed, which simply means that the overlapping lateral edges of the compartments have not been added to during growth.

The alæ are originally developed at the period of the metamorphosis, as slight lateral protuberances in the upper part of the compartments; from being overlapped, and therefore not exposed to external influences, and from growing (as in the case of the radii) against rectangular indentations or furrows in the adjoining compartments, they generally assume an extremely different appearance from the parietes, and might naturally be thought to have a very different nature. But the alæ in all cases (as is obvious in Pachylasma) are nothing but the protuberant lateral edges of the compartments, rendered thin and modified during growth. In order that the margins of the alæ should be received in an indentation, the upper internal surfaces of the walls of the recipient compartments are thickened all round, excepting where they receive the alæ. This thickened, upper, internal portion of the walls or shell, together with the alæ themselves, form the part called the sheath. The sheath sometimes blends insensibly into the lower parts of the compartments, and then perhaps it would not be thought to be a distinct element; but often it is abruptly separated by an overhanging edge (see Pl. 9, fig. 5 b, 9 b; Pl. 20, fig. 1 b; Pl. 25, fig. 1, K′) from the lower part, and then the sheath greatly complicates the internal appearance, but not the essential structure of the shell. The sheath acts beautifully, like an internal hoop, in strengthening the shell round the orifice, where it is naturally weaker than at the lower or basal end, where it adheres to the surface of attachment: in the upper part of the shell, moreover, the sutures between the compartments do not go straight through, but owing to the alæ projecting and being overlapped, are extremely oblique; or the joints, in the language of carpenters, may be said to be broken.

There is one other point of structure in the shells of the Balanidæ, more especially of species like Balanus tintinnabulum, which adds to their apparent complexity, namely, that the rim or orifice of the shell formed by the upper ends of the compartments, projects considerably above the opercular valves. In a young Balanus, immediately after the metamorphosis, the operculum is attached by the opercular membrane all round to the summits of the compartments, and there cannot be said to be any orifice to the shell itself, but only an orifice or slit between the opercular valves; but during growth, as the compartments are added to at their basal edges, their upper ends are deserted, and cease to enclose the sack, within which lies the animal’s body. Hence the upper ends come to project freely, either quite separately as in some species of Pollicipes, where they cannot be said to form an orifice; or more or less united into a ring so as to form an orifice, as in the different species of Balanidæ. It follows, that to understand the real shape of a Balanus, or rather of the cavity enclosing the animal’s body, all that part of the shell which projects above the opercular membrane, may, in imagination, be removed as something extraneous, like so many spines; not that I mean to say that these points of shell are dead; on the contrary, they are often porose and penetrated by numerous threads of corium. This upper part of the shell, thus produced so as to form an orifice, no doubt serves to protect the less strong and moveable operculum.

Fig. 4. Octomeris.

Fig. 5. Chthamalus.

Fig. 6. Chamæsipho.

Fig. 7. Balanus.

Fig. 8. Tetraclita.

a, Rostrum; b, Rostro-lateral, c, Lateral, d, Carino-lateral compartment; e, Carina.

Horizontal sections through the Shells of the principal genera of Balanidæ, showing the arrangement of the Compartments. Genera 4, 5, and 6 belong to the Chthamalinæ; 7 and 8 to the Balaninæ.

Number and Arrangement of the Compartments.—I have already stated that the shell, in every one of the Balanidæ, consists of eight, six, or four compartments, or of all fused together into a single piece; and that the compartments themselves are all constructed on the three simple patterns of which woodcuts (figs. 1, 2, 3) have been given. They are arranged in a certain definite order. The type arrangement is found amongst the Chthamalinæ, as might have been expected, inasmuch as this sub-family is so closely related to the ancient genera Pollicipes and Scalpellum, whence all the Thoracic Cirripedia may be said to radiate. In Octomeris (fig. 4) the type-arrangement of the compartments, eight in number, is well shown; the rostrum and carina resemble each other, and have alæ on both sides, and therefore are overlapped on both sides: the rostro-lateral compartments have radii on both sides, and therefore overlap the adjoining compartments on both sides; the lateral and carino-lateral compartments have radii on their carinal, and alæ on their rostral sides; and therefore overlap on one side, and are overlapped on the other side. Now the shell of every other sessile Cirripede differs, I believe, from that of Octomeris, only in the fusion together or abortion of some of the eight normal compartments: in one genus, however, Catophragmus, outer whorls of small compartments, arranged like the lower valves in the capitulum of Pollicipes, are superadded. The genus Chthamalus (fig. 5) differs from Octomeris only in the carino-lateral compartments being aborted, (as will presently be discussed), and hence has six compartments. Chamæsipho (fig. 6) differs from Chthamalus only in the rostro-lateral and lateral compartments being fused together; and hence has only four compartments. In Balanus (fig. 7) and the whole sub-family of the Balaninæ, the rostrum is compounded of the true rostrum, as seen in the type Octomeris, and of the two rostro-lateral compartments; hence this compounded rostrum has radii instead of alæ on both sides, and there are only six compartments. Tetraclita (fig. 8) and Elminius differ from Balanus only in having the carino-lateral compartments absent, and probably aborted; hence there are only four compartments. Lastly, in Pyrgoma, all the compartments are blended together into a single piece.

In Pollicipes, the old type-form of the whole order, and in Scalpellum, we have four valves, (answering to the operculum), surrounding the aperture leading into the sack, and the valves below are arranged in successive whorls, with a strong tendency to alternation. For, the rostrum alternates with, that is faces the interval between, the two scuta; the carina alternates with the two terga; and the upper lateral valves alternate with the scutum and tergum on each side. These four valves, namely, the carina and rostrum, which resemble each other in structure, and the pair of upper latera, which are larger than the other lateral valves, together form the uppermost whorl, or that beneath the scuta and terga. In the next whorl we have the rostro- and carino-lateral valves, alternating with those above them; and beneath them there are generally other valves, which decrease in size and display the same tendency to alternation. The valves here just specified, namely, the rostrum, carina, and three pairs of lateral valves, in the Lepadidæ, are so much larger, and are so much more commonly present, than the other valves of the capitulum, that to them alone I affixed special names. Now if amongst sessile Cirripedes we look to that genus, viz., Catophragmus, which comes in its whole structure the nearest to Pollicipes, one of the Lepadidæ, we find (as in fig. 4), firstly, a rostrum and carina resembling each other, and a pair of lateral compartments, larger than the other lateral pairs; these four valves alternating with the opercular valves: and, secondly, we find, but forming part of the same whorl, a pair of rostro-lateral and a pair of carino-lateral compartments, which, just as in Pollicipes, are larger than the exterior and lower valves. These lower little valves, I may remark, decrease in size in the successive whorls, and tend to alternate in position, just as in Pollicipes. Observing these several striking points of correspondence in the valves, (and indeed in the whole structure), of Catophragmus and Pollicipes, one is strongly inclined to suspect that in Catophragmus, and therefore in Octomeris and other sessile Cirripedes, although the rostro- and carino-lateral compartments appear to lie in the same whorl with the rostrum, carina, and large lateral compartments, yet that they really belong, as in Pollicipes and Scalpellum, to a lower whorl. Now if a very young shell of Balanus, immediately after the metamorphosis, be examined, the carino-lateral compartments will be found not to have been developed; they first appear after two or three zones of growth have been added to the other compartments; bearing in mind that in Pollicipes and in Catophragmus the lower whorls are added successively during growth, we find in this fact strong confirmation of the view that the carino-lateral compartments normally belong to a whorl beneath that including the rostrum, carina, and lateral compartments. Whether the rostro-lateral, like the carino-lateral compartments, are developed subsequently to the others, I have had no opportunity of ascertaining, and therefore cannot confirm the above analogical conclusion, namely, that they, also, belong to a lower whorl.

In the sub-family Balaninæ, which includes Balanus (woodcut 7), and Tetraclita (woodcut 8), the shell is characterised by not having rostro-lateral compartments, and by the rostrum having radii: now in Pachylasma giganteum, which undoubtedly belongs to the sub-family Chthamalinæ, at a very early age the rostro-lateral compartments become blended with the true rostrum, making a compound rostrum, exactly like the rostrum in the Balaninæ; distinct evidence of a similar fusion is retained throughout life (Pl. 15, fig. 1) in all three species of Chelonobia, which is undoubtedly a member of the Balaninæ. Hence, I think, we may conclude that in all the genera of the Balaninæ the rostro-lateral compartments are probably not aborted, but are blended with a normal rostrum (resembling that in woodcuts 4, 5, 6), making together a compound rostrum furnished with radii: it must, however, be observed that I could not detect any actual evidence of this fusion in Balanus, even immediately after the metamorphosis. In Chamæsipho (woodcut 6), either the rostro-lateral compartments attain a most unusual breadth, or, as is more probable, they have become confluent with the lateral compartments, which in the Lepadidæ seem to be the most persistent of all the lateral valves. In such genera as Tetraclita and Chthamalus, in which the carino-lateral compartments are absent, they may be fused with the lateral compartments or with the carina; but seeing that they are normally developed later than the other valves, it appears to be the simplest theory to assume, until the contrary be proved, that they are aborted. Finally, the somewhat unexpected conclusion that the shell (not including the operculum) of sessile Cirripedes normally consists of eight valves, four belonging to an upper whorl, and four to a lower whorl, all forced into a single ring, and often more or less fused together, though not strictly proved, is rendered highly probable. I will only further add, that the Basis perhaps represents several whorls of the small valves or scales on the peduncle of Pollicipes, fused together; the comparison of the basis with the calcareous cup, forming the lowest portion of the peduncle in Lithotrya, which has been made by some authors, I do not think is very accurate, as the cup in Lithotrya seems to have a special relation to the boring habits of that genus.

Fig. 9.

Basal edge of wall of compartment in Balanus tintinnabulum; a, a, outer lamina; b, b, inner lamina; c, c, longitudinal septa uniting the inner and outer laminæ with their ends denticulated.

Structure of the Individual Compartments.

If the basal margin of a compartment, for instance, of Balanus tintinnabulum, be examined, it appears sufficiently complicated, being composed of an outer and inner lamina, separated by longitudinal septa, which are denticulated at their bases; and the tubes formed by these longitudinal septa are crossed by transverse septa. On the other hand, in some cases, as in the genera Chthamalus and Elminius, each compartment consists of a simple shelly layer. These two extreme states graduate into each other: we have, firstly, on the internal surface, quite irregular points and ridges; these become regular, causing the internal surface to be longitudinally ribbed; then these ribs themselves become finely furrowed on their sides and at their lower ends, producing sharp, minute ridges, the ends of which I have called the denticuli; and, lastly, some of the denticuli on the adjoining longitudinal septa become united into a solid layer, forming the internal lamina of the wall. These denticuli do not generally cover the whole surface of the longitudinal ribs, but leave a portion near the outer lamina of the compartment smooth. The denticulated ends of the longitudinal septa project beyond the basal edge of both the outer and inner laminæ, and enter the mouths of the tubes (where such occur) in the basis, and thus strengthen the shell. The whole of the internal lamina generally is more or less striated longitudinally, thus displaying its origin from the union of the inner edges of the longitudinal septa. I need only further remark that on the internal surface of the outer lamina, between the main longitudinal septa, there are generally (as in the woodcut) smaller longitudinal ridges, which do not reach the inner lamina, and on this account alone are not called septa.

Tubes are formed by the longitudinal septa, between the outer and inner laminæ. These tubes are almost square, and are occupied by threads of corium, which enter at pores left open between the edge of the compartment and that of the basis on which it rests. The tubes extend high up the compartments; but in the uppermost part they are generally cut off by thin, transverse, calcareous septa, deposited by the ends of the threads of corium; a cancellated structure being thus produced. Or the uppermost part of each tube becomes filled up solidly with compact shelly layers, which are always first thrown down on the side of the tube facing the outside, and thus greatly strengthen the shell: in several instances, as in Balanus perforatus and Tetraclita porosa, in which the disintegration of the upper part of the shell is a necessary element in its growth for the enlargement of the orifice, these filled up tubes become exposed. In Coronula and Tubicinella, the tubes in their upper parts are, I believe, crossed only by transverse membranous septa.

Anomalies and exceptions.—In Tetraclita (Pl. 10, fig. 1 g, 1 h) from the branching of the longitudinal septa, several irregular rows of tubes are formed. In certain varieties of Balanus balanoides (Pl. 7, fig. 2 b), and in B. cariosus (Pl. 7, fig. 3 b), slight branching ridges on the internal surface of the walls, seem to answer to the longitudinal septa, and produce, during the downward growth of the shell, extremely irregular cells, and short tubes. In Balanus vinaceus (Pl. 2, fig. 7 d), the internal lamina, instead of being solid, as in every other species, is left cancellated, and thus betrays, much more plainly than usual, its origin in the united denticuli of the adjoining longitudinal septa. In Balanus porcatus, between the main longitudinal septa, there are (Pl. 6, fig. 4 e) what may be called rudimentary and disconnected longitudinal septa. In Coronula and its allies (Pl. 16, fig. 6, and Pl. 17, fig. 4 c) it is the outer lamina of the compartment which is anomalous; for in the two or three lower zones of growth, it forms only a ledge on each side of the longitudinal septa; which ledges, higher up, become confluent, and so form an ordinary outer lamina. In Coronula, also, the wall of each compartment (see transverse section, Pl. 16, figs. 5, 7) is very remarkable from being deeply folded, the folds being on their internal faces firmly calcified together, and on their external faces closely pressed together (often with a neatly serrated suture), so that the whole nature of the shell might be, as has happened, easily quite misunderstood; and the walls be considered as very thick, instead of being, as is really the case, very thin. In Chelonobia (Pl. 15, fig. 1), however, the walls are truly of such great thickness, that the nature of the relative parts might likewise be misunderstood; in this genus the ovarian tubes enter the walls, extending up between the longitudinal septa, or, as they may here be more naturally called, the radiating septa. I will specify a few more peculiarities worthy of remark:—in some species of the sub-genus Acasta, clefts are left, covered only by membrane, on the lines of suture (Pl. 9, figs. 7 a, 8 a), between the compartments, just above the basis; and in other species the basis is perforated by numerous membrane-covered, minute orifices. In Platylepas, each compartment has one deep inward fold (Pl. 17, fig. 1), somewhat analogous to the three folds in Coronula; this fold is produced into an internal midrib, supporting and rendering convex the membranous basis; in this genus, also, the rostrum, owing to its midrib, is generally thrust a little on one side, and the shell thus rendered asymmetrical. In Chamæsipho scutelliformis the shell is symmetrically perforated (Pl. 19, fig. 4 a) by four apertures. Lastly, in Chthamalus Hembeli and intertextus, after a certain age, the basal edges of the walls become inflected, and continue to grow inwards till they entirely take the place of the true membranous basis.

Structure of the Radii.

Radius.—This term, as we have seen, is applied to that side of the compartment, the growth of which is modified, by abutting against and overlapping the adjoining compartment. Hence the structure of the radius is essentially the same with that of the parietal portion of the compartment. When best developed, as in Balanus tintinnabulum, the radius consists of an outer and inner lamina, separated by denticulated septa, extending in horizontal lines parallel to the basis, and is consequently perforated by minute tubes or pores. The tubes become filled up solidly much more commonly than do the parietal tubes; and the inner lamina, in such cases, is hardly distinct from the outer lamina. The denticuli often fail, or are present only on the lower sides of the septa; and very frequently the edge of the radius can only be said to be crenated. Notwithstanding these frequent anomalies, if a series of species and genera be taken, it is certain that there is, as might have been expected, a close relationship in internal structure, between the radii and the parietes. The edge of the radius is received in a slight furrow (generally marked like a seal, with the impression of the denticulated septa) in the opposed compartment: sometimes the outer edge or lid of the recipient furrow, is so broad as to give the false appearance of a radius having been developed, at least in the lower part of the shell, on both sides of the suture. A crest of corium runs into each suture between the edge of the radius and the furrow in the opposed compartment; and when the radius is permeated by pores (as in woodcut 10), threads of corium branch off this crest, and enter the pores. In the lower part of the shell, these crests of corium project from the corium forming and surrounding the sack; but in the upper part of the shell, above the opercular membrane, and therefore above the sack, the corium is produced up each line of suture as a separate ribbon. In proportion as these ribbons extend more or less near to the summit of the shell, so do the edges of the radii continue to be added to, to a greater or less height from the basis; and consequently their summits become less or more oblique.

Fig. 10.

Edge of the radius of Balanus tintinnabulum. a, outer lamina; b, inner lamina; c, denticulated septa, uniting the two laminæ.

Peculiarities in the Structure of the Radii.—In some of the species of Tetraclita, in which genus the walls consist of several rows of tubes, the radii are likewise perforated by several rows; and in some of the other species (Pl. 10, fig. 1 h), the edge, or disarticulated surface of the radius, is marked by irregularly branching ridges; and these evidently correspond with the branching septa or ridges of the wall. In Chelonobia, the outer lamina of the radius, as well as of its recipient furrow, is of extraordinary thickness; and this lamina, in C. testudinaria (Pl. 14, fig. 1 a, 5, b, and Pl. 15, fig. 1, f), is modelled into sharp transverse ridges and valleys. In the Chthamalinæ, the radii, like the parietes, are simply solid; and apparently in consequence, for the sake of strengthening the sutures, the edges of the radii, and of the recipient furrow in Octomeris (Pl. 20, fig. 3 a) and in Chthamalus dentatus and Hembeli (Pl. 18, fig. 3 b, 5 a), are neatly dentated. In some other species of Chthamalus (Pl. 19, fig. 1 a), the radii present a slight modification of this structure, the sutures being formed by oblique interfolding laminæ. In the radii of Coronula and Tubicinella, there is a peculiarity, in apparent connection with the fact, that in these genera the parietal tubes are not crossed by transverse calcareous septa, namely, that the pores by which the radii are permeated keep unclosed throughout their length, and open into a special longitudinal tube (Pl. 16, fig. 7, d′), which runs along that margin of the wall, whence the radius arises. In Coronula the wall is of extreme thinness, and in conformity so is the true radius, but that the shell might not thus be rendered very weak, complementary or pseudo-radii are developed on their inner sides (Pl. 16, fig. 7, adjoining the true radii A d, C d, and shaded by distant convex lines). Even in the allied genus Xenobalanus, in which the whole shell tends to become rudimentary, traces of these pseudo-radii (Pl. 17, fig. 4 b, d) can be detected. In Coronula, though the radii (Pl. 16, fig. 7, A d, C d) are, by the above special means, rendered thick, and though the alæ also are thick (C a′, D a′), yet together they do not equal in thickness the folded walls; and consequently, there is left between the radii and alæ square chambers (v), occupied by the branching ovarian tubes.

Structure of the Alæ.

These project, generally abruptly, from the sides of the upper part of the compartments; they appear from the first growth of the shell; they are overlapped by the radius and by part of the wall of the adjoining compartment; they are thinner, and have, owing apparently to being overlapped, a very different aspect from the parietal portion; but they do not differ from it in essential nature. They are solid, that is, they are never permeated by pores; but their edges are generally crenated, and there is, in some cases, as in Chelonobia, sufficient evidence that these crenations answer to the horizontal septa on the edges of the radii (also often reduced to mere crenations), and consequently, likewise, to the longitudinal septa of the parietes. In Coronula the edge of each ala consists of a medial ridge, sending off denticulated septa on both sides, and is therefore anomalous as compared with the alæ in other genera, but corresponds in structure with the similarly anomalous radius of Coronula. In order to allow of the growth of the edge of the ala, a fine thread of corium runs up the narrow furrow in which the edge is lodged, proceeding from the corium of the sack. In proportion as this thread runs up higher or lower, so are the summits of the alæ rendered, during growth, less or more oblique.

Structure of the Sheath.

As the compartments overlap each other, the edges of the alæ would have projected, and the inner surface of the orifice of the shell would not have been smooth and rounded, had not that part of each wall, which does not overlie an ala, been thickened so as to allow of the formation of a shoulder or indentation, against which the edge of the ala fits and abuts. The thickened portions, and the alæ themselves, together form the sheath, of which the use seems to be to strengthen, like a broad internal hoop, the upper part of the shell round the orifice, where naturally it is weak. The sheath is composed of successive, fine, shelly layers, which extend, as the shell is added to at the basal margin, lower and lower down on the inner surface of the walls. The lower edge of the sheath either simply projects a little inwards, or more commonly is formed into a sharp depending ridge, as represented in fig. 1, K′, Pl. 25. In some species of Pyrgoma (Pl. 13, fig. 2 b), the sheath reaches nearly to the bases of the compartments; and in Chelonobia (Pl. 14, fig. 4 e c e), the inner layer of shell surrounding the sack, which seems to correspond more nearly to the sheath than to the inner lamina of the walls, actually rests on the basal membrane. The opercular membrane is generally, but not invariably, attached only a little way above the lower edge of the sheath: at each exuviation, a new opercular membrane is formed, and is attached to the next lower zone of the sheath; the old membrane being cast off, but a circular slip of it is left, investing the last zone. Hence the whole upper part of the sheath above the opercular membrane, comes to be thus invested; and is marked by circular lines, one above the other, caused by the successive exuviations. This investing membrane often supports rows of minute bristles, directed upwards. Generally, a film of shell is deposited, at the period of the formation of each new opercular membrane, on that part of the sheath which lies immediately beneath. This innermost film or thin layer of shell, on the lines of suture between the compartments, breaks joint, at least in some cases, with the underlying shelly layers,—that is, the suture in this last-formed film does not lie exactly over the suture in the subjacent layers of the sheath. In Tubicinella, the sheath extends down close to the basis; and what is unique in this one genus, the opercular membrane, gradually thinning out downwards, closely adheres to the whole inner surface of the shell. In Tubicinella and in Xenobalanus (Pl. 17, fig. 4 b), the sheath separates easily into separate successive rims of shell; and this structure evidently is for the sake of facilitating the breakage of the upper end of the shell, which, as we shall presently see, is necessary to allow of the increase in size of its orifice.

Structure of the Basis.

This, in several genera and species, is composed of simple membrane, and consists of successive, concentric, circular slips, added round the outside, at each period of growth. In some species of Tetraclita and Balanus the basis is calcareous, but diaphanous, very thin, smooth, or somewhat granulated. In other cases it consists of a single calcareous lamina, either smooth, or with ridges radiating from its centre; it is formed of two laminæ, (as is most usual in Balanus,) separated by radiating septa. These septa, as well as the radiating ridges in the case of the single lamina, are homologous with the longitudinal septa of the parietes. The denticulated ends of the latter enter the mouths of the tubes formed by the radiating septa of the basis: threads of corium pass between the denticuli of the parietal septa, and thus enter the basal tubes. The ends of these threads of corium generally deposit transverse calcareous septa, exactly as within the parietal tubes. When the basis is thick the septa themselves (ccc) between the proper basal tubes, become porose, (or rather cancellated,) and they sometimes expand into a very thick, cancellated layer, separating the outer lamina (a) of the basis from the proper basal tubes, which always lie close under the inner lamina (b). This structure differs only slightly from that seen in the parietes of Tetraclita, in which the branching of the longitudinal parietal septa, produces thick walls, formed of several rows of tubes or pores. With respect to peculiarities in structure of the basis, Balanus lævis offers the most remarkable case; for here, in specimens which have grown crowded together, the whole interior appears sometimes to have become too much elongated or too deep for the animal’s body, and consequently the lower part of the deeply-concave basis has been filled up (Pl. 4, fig. 2 a) by thin, irregular, calcareous diaphragms. In elongated specimens, also, of Balanus balanoides, the shell sometimes appears to have grown too long for the animal’s body; but in this case the membranous basis becomes extremely convex inwards; it still reaches the basal edges of the parietes all round, but in the middle it is raised high above the surface of attachment; yet sometimes threads of the cementing tissue depend from the middle part to the surface of attachment. In Balanus terebratus (Pl. 8, fig. 2 a, 2 b), and in some species of Acasta, the basis is riddled, as previously stated, by numerous, minute, membrane-covered orifices. In B. declivis the membranous basis is always extremely oblique, owing to the rostral end of the shell being twice as high as the carinal and opposite end.

Fig. 11.

Portion of edge of basis of Balanus tintinnabulum, a, a, outer lamina; b, b, inner or upper lamina; c, c, c, porose or cancellated radiating septa.

Regarding the very remarkable means by which the basis of sessile Cirripedes is cemented to the surface of attachment, it will be convenient to defer for a little the description, on account of its necessary length.

Structure of the Opercular Valves (Scuta and Terga).

These are situated on each side of the slit or orifice leading into the sack; from their shape, their powers of movement, their separation by flexible membrane from the shell, to which they serve as a lid, they appear at first as if they constituted an element very distinct from the shell itself, but this is not the case. They are, together with the opercular membrane, as essentially as the whole of what is externally visible, a part of the modified carapace, of which they occupy the upper or posterior extremity: from tracing the metamorphoses, or even by comparison of a Balanus with Pollicipes, there can be no doubt of the truth of this conclusion. The opercular valves are four in number,—a pair of scuta and a pair of terga; but the latter in Coronula diadema and reginæ, are either aborted or represented by a mere rudiment; and in Xenobalanus both scuta and terga are quite absent. In several cases, more especially in the genus Pyrgoma (Pl. 13, fig. 1 b), the scutum and tergum on each side are calcified together, so that sometimes not even a trace of the line of junction can be discovered. In most cases the scutum is firmly united, being articulated in a manner presently to be described, to the tergum; but in Coronula, Tubicinella (Pl. 17, fig. 3 c), and Platylepas, the ends of these valves are simply approximated.

Scuta.—These valves are important, inasmuch as the animal’s body is attached to them; in Pl. 25, fig. 1, the broken line, surrounding a, b, shows where the body has been cut, in removing the scutum on the near side, the other scutum, S, being left articulated to the tergum, T. In shape the scuta are generally sub-triangular; but in some species of Pyrgoma and in Chelonobia, &c. they are much elongated. The lines of growth are usually prominent; and along the occludent margins the alternate, or sometimes every third or fourth line, is developed into a knob, which produces a serrated edge, serving to lock the two opposed valves together; there is, however, no trace of this structure in Coronula and Tubicinella. In some species of Pyrgoma, a ledge of considerable breadth (Pl. 13, fig. 3 e, &c.) is developed along the occludent margins of the two scuta, as well as of the two terga, giving them an anomalous structure. The Terga differ considerably in outline in the different genera and species: their shape approaches more nearly to a triangle than to any other regular form; but there is generally a projection or spur on the basal margin, on the side towards the scutum. In some species of Pyrgoma, the tergum is of so irregular a shape as to defy description. In most cases, a longitudinal depression or furrow runs down the valve, from the apex to the extremity of the spur; and it not rarely happens that the sides of this furrow become folded inwards and almost closed. The spur probably answers to the basal point of the usually sub-rhomboidal tergum in Pollicipes and Scalpellum.[22] The tips of the terga in some species of Balanus, &c., are specially modified into sharp points or beaks (Pl. 2, fig. 3 b, 3 d), bowed a little inwards, and projecting considerably above the tips of the scuta; this is effected by the medial, uppermost part of the valve being internally thickened and hardened, and then, by the disintegration of the two margins and the external surface, the internal modified portion becomes exposed. The whole valve, also, at least in such cases as in Balanus psittacus, appears to be forced slowly upwards in the articular furrow of the scutum. I am assured, by a competent observer, that the beaks of the terga in B. porcatus can give an object placed within the orifice of the shell a sharp tap.

The scutum and tergum, with the few exceptions above stated, are articulated together at a large or open angle. The articulation (see Pl. 11, fig. 5 b, c, d, and fig. 6 b, c) is effected by the margin of the tergum being a little inflected, and lodged in a furrow in the margin of the scutum. This furrow in the scutum has its further border generally prominent and often reflexed or curved over; I have called it the articular ridge; it, also, is lodged in a furrow in the upper part of the tergum, which again is bordered by a ridge, viz., the articular tergal ridge. So that in both scutum and tergum there is an articular furrow, bordered in each case, on one side by the margin of the valve, and on the other side by the so-called articular ridge. In Chelonobia (Pl. 14, fig. 1 b) the articular ridge of the scutum is horny. When, as often happens, the scuta and terga have been much worn, the manner of their articulation (Pl. 18, fig. 1 a) is pretty well shown even from the outside; in this case their external appearance is very different from what it is in those individuals (fig. 1 c) of the same species, which have not suffered disintegration. This articulation of the scuta and terga is prefigured amongst the Lepadidæ, in Pollicipes mitella, and in Lithotrya.

The scuta are brought together by a short, strong, straight, adductor muscle (Pl. 25, fig. 1, a); its attachment leaves (with very few exceptions, as in Tubicinella) a rounded impression, or even pit, on the under side of the valve in its upper part. This pit is frequently bounded, on its lower side, by a sharp ridge, which, though not in actual connexion with the adductor muscle, I have, for convenience sake, called the adductor ridge; it serves apparently to give support to the animal’s body; in some few cases (as in B. psittacus, Pl. 2, fig. 3 c) it is confluent at its upper end with the articular ridge, and converts the whole basi-tergal corner of the valve into a deep cavity. In some of the species of Pyrgoma (Pl. 12, fig. 5 c, 7 b), and in some varieties of Creusia, this adductor ridge is enormously developed, so as to depend far beneath the true basal margin, or that to which the opercular membrane is attached. At the basi-tergal corner of the valve, there is generally a small pit or impression, and sometimes distinct crests, for the attachment of the lateral depressor muscle. At the rostral end there is, also, a small cavity formed by the overfolding of the occludent margin (rarely furnished with crests) for the attachment of the rostral depressor muscle. In the Terga, at the basi-carinal corner, there are usually crests, though sometimes feebly developed, for the attachment of the tergal depressor muscle. But in Chelonobia, Coronula, Tubicinella, Platylepas, and in some other cases, there are no crests. The crests, when well developed, are furnished with rectangular sub-crests or denticuli on both sides; in fact they resemble, and are probably homologous with, the denticulated ribs or septa in the parietes, radii, and basis. Altogether the scuta and terga are attached, as far as muscles are concerned, to the shell and sack, by three longitudinal pairs.

Growth of the Whole Shell, and Its Microscopical Structure.

The opercular valves are added to along their basal margins alone;[23] the animal’s body, together with the several muscles, becoming attached at each period of growth lower and lower down to the valves; this no doubt is effected by the absorption of the upper surfaces of the muscles, and the formation of new fasciæ on their lower surfaces. The opercular membrane, which, though thin and flexible, forms part of the general outer surface of the animal as much as does any portion of the rigid shell, with which indeed it is strictly homologous, is periodically moulted, together with the integuments of the whole included animal. The new opercular membrane is of course each time formed a little larger than the old one. In Coronula and Tubicinella, however, several successive opercular membranes are preserved one over the other, and the outside membrane gradually disintegrates; in these cases the undermost opercular membrane is formed wrinkled and considerably too large, so as to allow of being stretched, before it is finally cast off. In Tubicinella, the opercular membrane runs down, adhering to the inner surface of the shell, to nearly the basis, and hence during the diametric growth of the shell, it is longitudinally split, and is repaired by slips of new membrane, which resemble the radii in form and in direction of the lines of growth.

The basis is added to only round the circumference, and hence increases in diameter, and, when concave, in depth. The compartments grow at their basal margins, where they are in contact with the basis; hence the shell is added to in height, and, owing to the outward inclination of the compartments, also, in basal diameter; but the compartments likewise, in most cases, grow along both lateral margins, that is, on the edges of the radii and alæ; and hence the upper part of the shell, also, increases in diameter. The orifice of the shell, moreover, thus becomes enlarged. In some cases the shell is destitute of radii, only sutures being present, that is, the compartments do not grow laterally; and sometimes, as in the whole genus Pyrgoma, there are not even sutures, the compartments having been fused together: in both these cases, the shell can increase in diameter only at the base; and the orifice, it might have been thought, would necessarily have remained, to the destruction of the animal, of the same minute size, as when first formed after the metamorphosis: this certainly would have been the case had not the upper ends of the compartments, surrounding and forming the orifice, been nicely adapted always to yield, in a certain limited degree, to the disintegrating influences to which every shell is exposed, but which most Cirripedes can resist; and the disintegration of the narrow end of a conical tube, of course increases the diameter of its orifice. In Tubicinella, in which the shell is furnished with narrow radii, and does increase in diameter from top to bottom, the increase is not sufficient in proportion to the continued elongation of the shell; to compensate for this, the orifice is enlarged at short intervals by the breakage of the upper end of the shell, for which purpose (as explained under the genus) it is evidently constructed. Hence we see that, in certain Cirripedes, decay or disintegration, and breakage, are necessary elements in their growth! It is a remarkable fact, which I cannot explain, that in some species in which the orifice of the shell is usually increased by disintegration, if individuals are so situated that they are not exposed to sufficiently energetic disintegrating influences, as may be inferred from the well-preserved condition of the whole surface of the shell, then the radii become developed, and the orifice is increased in size by the diametric growth of the upper part of the shell: I have seen instances of this in Tetraclita porosa, and purpurascens, and in Balanus perforatus: it appeared, but of course erroneously, as if the lateral growth of the compartments had been subjected to the will of the animal.

Considering the strength of the shell of sessile Cirripedes, the separation of their compartments one from another and from the basis, during growth, has justly been thought a surprising circumstance. In most Chthamalinæ and in some species of Balanus, however, if the shell be boiled in caustic potash, the compartments fall apart with a touch; this shows that their union is due to animal and probably to organised matter, and the growth of such matter between the opposed edges of the compartments, and their consequent gradual separation, offers no particular difficulty. But in many Balani, boiling in potash for hours does not seem even to weaken, in the least degree, the sutures, which are wonderfully strong—the shell often breaking rather than yield on these lines; if, however, the shell be dissolved in acid, the animalised tissue which is left easily separates on the lines of suture, and if this tissue be boiled in potash, the remnants of the compartments fall quite separate. These facts seem to me to show, that the compartments in such cases are joined along the lines of suture by tissue, which must be in a calcified state, but which, nevertheless, continues to grow by intersusception; in other words, I believe that the tips of the complicated ridges and points interlocking on the lines of suture, are not separated from each other by films of corium or simple animal matter, but are actually united by corium in a calcified, yet still growing condition.

In ordinary Crustaceans, the growth is periodical and sudden; a new and larger carapace, for instance, is formed under the old one, and after the exuviation of the latter, the new one soon hardens, and does not subsequently increase in size; so it is in the case of Cirripedes, with the membranes of the body, and even with certain parts, as the opercular membrane, of the external covering. But a Cirripede cannot, like a crab, crawl into some crevice and remain protected till its shell becomes hardened; hence, probably, it is that the shell is never[24] wholly moulted. Even if the margins of the opposed compartments and of the basis were to grow rapidly, the shell would necessarily be much weakened on the lines of suture, and unable to withstand the heavy breakers, to which so many species of sessile Cirripedes are exposed. On the other hand, although the margins are thus compelled to grow slowly, they do not grow continuously, as may be seen in the zones of increment on all the valves, corresponding, I believe, with the periods of exuviation of the membranes of the body. A layer of shell, often very thin, seems to be generally deposited over the whole internal surface of the several valves, at the same time that the marginal zones are added; so that the only essential difference in the growth of the external covering, in Cirripedes as compared with ordinary Crustaceans, is that the old shell is not cast off, but adheres to the outside of the new shell, and that the margins are added to (in certain definite directions) slowly yet not continuously, instead of the whole being formed at a single period.

If, now, a section of one of the shelly zones of growth be carefully examined, it can in some cases be distinctly seen to be formed of successive, excessively fine laminæ; but the animalised tissue (which differs much in amount in different Cirripedes) left after the shell has been dissolved in acid, exhibits, in most cases, neither laminæ nor any other structure whatever. The shell seems to be the actual pulpy corium, or true skin, in a calcified condition, but generally with its cellular structure modified and much reduced: I have taken a bit of recently-formed shell of Tetraclita and of Coronula, with the corium still adherent to its under surface, and after dissolution in acid, I could not distinguish the part, which had just before existed as shell, from the corium itself. In the case of Coronula, immediately prior to the period of moulting and growth, I found the unaltered corium so charged, as to effervesce, with carbonate of lime, either in a state of dissolution, or in granules too minute to be visible under the highest powers.

The sutures between the several compartments and the basis are covered by thin membrane, which is continually splitting during the growth of the opposed edges of the underlying shell; but previously to each splitting, a new slip of membrane is, I believe, already formed under the old one; so that the corium is not even momentarily exposed. Owing to this manner of growth, the slips of membrane consist of successive rims united together; in most cases, these soon become abraded from the older parts of the shell, but are sometimes preserved. The last-formed slip of membrane over a suture is homologous with the opercular membrane; and both are strictly analogous with the ring of flexible membrane, forming the joint of the leg of a crab. In the latter case, the flexible membrane and hardened crust are both moulted together: in the opercular membrane, there is a double line of splitting, one close round the opercular valves, and the other at the basal edge of the sheath, and the intermediate portion is moulted, but with a zone of membrane left adherent to the non-moulted valves and sheath: lastly, in the slips of membrane covering the sutures, there is only a single line of splitting, and no portion, I believe, is moulted; the rims of membrane on each side remaining adherent on the compartments and basis, until worn away.

The opercular membrane, when closely examined, exhibits no structure, except that it can sometimes be plainly seen to be composed of successive, numerous, excessively thin laminæ. Occasionally, however, it presents the false appearance of being permeated by parallel and anastomosing vessels: this appearance is clue to one or more of the component laminæ having been wrinkled before a succeeding lamina was thrown down and attached to its under side. If a small piece of an opercular valve of Tubicinella, with the opercular membrane adhering to it, and with the corium under both, be dissolved in acid, it may be clearly made out that the corium under the valve has gone on being converted into shell, whereas under the opercular membrane it has been converted and condensed into fine constituent laminæ of chitine. Inasmuch as the successive layers of shell, during each period of growth, go on encroaching on those of the membrane, the line of junction between the shell and chitine becomes oblique or bevelled. The membrane on this bevelled line of junction assumes a slightly different aspect to what it has elsewhere; it becomes yellowish or brown, thicker and very much tougher. In many genera it is also furnished with a row of small bristles. At the period of exuviation the opercular membrane separates just outside this modified portion, leaving the latter adherent, as a rim or slip, on the valves. If, however, the opercular membrane be rudely torn off before its proper period of exuviation, it carries with it the as yet continuous, but already modified, slip. A slightly indented line may sometimes be traced before the period of exuviation, showing where the separation will take place: what produces this line I know not. The coloured, thickened, and modified slips of opercular membrane, which are thus retained adhering to the valves, and which together form an investing membrane, have been considered by most authors as the epidermis; but they have no more right to be thus called than has any other part of the opercular membrane. Exactly similar slips of membrane are left investing the sheath. So, again, the membrane which, when well preserved, invests the walls of the shell, is made up, as already stated, of successively adherent slips, which originally covered the lines of suture.[25]

The little bristles above alluded to, which arise from the slips of membrane left adherent on the opercular valves, sheath, and walls, stand in rows; a row corresponding to each period of exuviation of the opercular membrane. The bristles are generally largest on the opercular valves and sheath; in Balanus tintinnabulum, they are from 1 to 2/1000ths of an inch in length, but they are longer in some other species. I may here mention, as showing the connexion of these bristles with the opercular membrane, that similar bristles occur in B. perforatus, scattered over the surface of that membrane, and are necessarily moulted with it. In the imbedded genera Coronula and Tubicinella, none of these bristles exist. When a portion of valve or shell, furnished with bristles, is dissolved in acid, tough, sinuous, and apparently hollow, threads are seen to run from their bulb-like bases, into and up the corresponding layer, which, before dissolution, existed as shell; and they terminate internally in very fine points, which I believe are united to the underlying corium. These threads, or tubuli,[26] as I have called them in my volume on the Lepadidæ, are, in Tetraclita porosa, about 1/5000ths of an inch in diameter, but only half that size in B. tintinnabulum. On parts of the shell where there are no bristles, similar tubuli penetrate the shelly layers, and come to the surface. The tubuli running to the lowest and last-formed row of bristles, just after a period of exuviation, are so delicate as hardly, or not at all, to be distinguished; in the row above, they are plain and longer, and for the next two or three upper rows they are, in some cases, as in Tetraclita porosa, longer and longer, having been added to during each successive thickening of the valve. These tubuli consist of chitine, and no doubt first existed as threads of corium; they are so tough that they must serve to strengthen the successive layers of shell, but I imagine their chief function is to keep up the vitality of the newly-formed layers of shell. May we not, also, venture to suppose that by their means, some degree of sensibility is given to the bristles? I need only further remark, that in some species of Balanus and of Chthamalus, the under side of the shell is penetrated by irregular pores, large enough to be visible to the naked eye, into which threads of corium penetrate; but these can hardly be said to appertain to the microscopical structure; and are more nearly related to those pores and furrows, formed by the greater or less development of the longitudinal septa, and in which the threads of corium deposit, or rather become changed into, transverse septa, or solid shelly matter, as previously described.

Muscles of Sack.

In the pupa, the thorax, as we shall hereafter more fully see, is continuous with, and opens into the large anterior end or front part of the head; but during the metamorphosis (Pl. 30, fig. 2), the thorax of the Cirripede becomes, owing to the almost transverse position occupied by the young animal within the pupa, to a great extent internally separated from the anterior end,—which anterior end forms, as we know, either the peduncle or the basis. Hence it comes to pass that the body or Thorax (Pl. 25, fig. 1) is lodged within a sack (f) within the shell. The chitine membrane lining this sack is excessively thin and transparent, but less so in Xenobalanus and Tubicinella; it is obviously continuous with that investing the body of the animal; it is also essentially continuous with the opercular valves and membrane, and consequently with the whole shell. It is periodically moulted. It is lined by corium, as is likewise the surrounding shell; hence the corium is double round the sack, as indeed might have been expected from the shell and opercular valves (at least their upper parts) being formed by the prolongation, as is obvious in the pupa, of the posterior edges of the carapace. Between the two folds of corium, which are united together by transverse ligamentous fibres, branching out at both extremities, like the roots and branches of a tree, we have the longitudinal muscles, which go to the opercular valves; and likewise a layer-like mass of branching ovarian tubes (Pl. 25, fig. 1, g): the ovarian tubes, however, are often confined to the base of the sack. In Xenobalanus, the two folds of corium are united by longitudinal membranous septa, making a series of quite peculiar, square tubes.

The above-mentioned muscles are attached at their upper ends to the opercular valves, and at their lower ends to the basis. There are, in fact, three pairs, but the pair attached to the basi-carinal angles of the two terga (Pl. 25, fig. 1, i), are almost invariably confluent, forming one great bundle; the second pair is attached to the lateral or basi-tergal corners of the two scuta, and are hidden in the figure; the third pair (h) is attached also to the scuta, to their rostral angles. These muscles can only act as depressores; they are often extremely powerful; they belong to the voluntary class, for they are transversely striped. By their action, the opercular valves are capable of varied slight movements, within the limit allowed by the width of the flexible opercular membrane. By the action of the lateral scutal depressores, the orifice leading into the sack is opened, the movement being generally aided by the protrusion of the cirri. By the sudden contraction of the rostral scutal depressores, the blows which are sometimes given by the beaked terga at the opposite end of the operculum, are probably effected. By the contraction of all three pairs of muscles, the opercular valves are held down with quite surprising force. The valves can be raised only by the action of the animal’s body against the basis.

In Coronula these muscles are more spread out, and do not extend down to the basis; their lower portions, as is likewise the case in Tubicinella, do not exhibit transverse striæ, and hence tend to pass either into the involuntary class, or into ligament. This condition of the muscles, in the above two genera, accords with the little-developed state of their opercular valves. In Xenobalanus, there is no longer any evidence of the muscles being collected into five or six bundles, for they are thinly and almost uniformly spread out, and show in no part transverse striæ. I may add that in much elongated specimens of Balanus balanoides, these muscles become in their lower part ligamentous, and destitute of striæ.

Branchiæ.

In the Balaninæ, a pair of Branchiæ is always present: they lie on each side, in a somewhat curved position, in the angle between the sides of the shell and the basis. In Pl. 25, fig. 1, they are exactly covered, on the further side, by the body of the animal. They are attached near each other at the carinal end of the sack in a vertical line, and likewise on each side in a transverse line, extending from close beneath the spur of the tergum towards the point of attachment of the body to the scutum. In Balanus, as in the figure (Pl. 25, fig. 3) of B. tintinnabulum, each branchia consists of a medial fold of skin, a little curved conformably with the sack, and slightly tapering towards its rostral and free extremity; but this fold is almost hidden by the vertical sub-folds or membranous ridges, themselves plicated and sub-plicated, which project on both sides: these vertical folds are free at their tips: at their lower attached ends, they are thickest. On the side nearest the wall of the shell, the whole branchia has a bilobed appearance, owing to a very deep indentation caused by the projection of the scutal lateral depressor muscle; the sub-folds on this side are also more plicated. The branchia essentially is an inward plicated fold of the membranes of the sack; for its outer, very thin tunic is continuous with and moulted with that lining the sack; and within it we have two layers of delicate, pulpy, transparent corium, united together (as is best seen in Coronula) by ligamentous fibres, branched at their two ends, all exactly as in the corium surrounding the sack. There are here no distinct vessels, any more than in other parts of the body, but a fluid could easily circulate in the interspaces of the corium. From the large size of this organ, and its simplicity of internal structure, being adapted exclusively to expose a great surface of skin to the water, I do not doubt that it has been correctly considered as a respiratory organ. By the voluntary movements of the opercular valves (i. e. part of the carapace) the water is constantly being pumped in and out of the sack; the movement, indeed, may be almost compared to the heaving of a man’s chest. Moreover, the branchiæ on each side are attached so closely to the spur of the tergum, that each time the latter is moved, the whole branchia must, I think, be agitated, and the folds opened, as by the action of a lever.

In our two commonest, tidal, sessile Cirripedes, viz. Balanus balanoides and Chthamalus stellatus, I have observed that, when left uncovered by water, they kept the orifice of their operculums a little open, with a bubble of air within their sacks, so that the orifice was in fact closed by a thin septum of water, with air beneath; when disturbed, they closed their operculums with force, and expelled the bubble of air with a clicking noise, which has been noticed by Dr. Coldstream,[27] and has been thought to be made by the movement of the operculum itself. Bal. crenatus, a deep-water species, when out of water, keeps its operculum closed.

In Coronula, Platylepas, Tubicinella, and Xenobalanus, each branchia[28] consists of two unequal folds, both plicated on both sides: in the two latter genera, they extend far down the deep and elongated sacks, and hence the area of surface altogether gained is extremely great. In most of the species of Chthamalus, the branchia consist of a small fillet barely plicated: in the allied Chamæsipho columna, they are rudimentary, forming a smooth little pouch only 1/100th of an inch in length: in Chthamalus scabrosus they are quite aborted, being perhaps represented by a slight hairy ridge; but in Chthamalus dentatus, and therefore within the limits of the same genus, the branchiæ (and this seems to me a singular fact) are large, each being composed of two plicated folds, as in Coronula. Tapering filaments situated near the bases of the cirri, such as those occurring in several species of the Lepadidæ, are not found in any sessile Cirripede; but I have observed nearly similar filaments, projecting upwards and inwards at the base of the sack, in several species of Balanus and in Coronula; those which I examined were simply occupied by delicate corium, and no doubt must aid in exposing a greater surface of corium to the circumambient water.

In my former volume on the Lepadidæ, I have described the ovigerous fræna occurring on the two sides of the sack, to which the ovigerous lamellæ are attached by a peculiar glandular secretion: in the Balanidæ there are no ovigerous fræna, but the branchiæ just described are identical with the fræna in essential structure and in position; differing only in being placed a little nearer to the carinal end of the sack, and in being generally (but not always) larger and more plicated: seeing this, and that in Alcippe lampas, and in some species of Pollicipes,—the genus which comes nearest to the Balanidæ,—the ovigerous fræna are large and are destitute of glands, and have therefore lost their normal function of supporting the ovigerous lamellæ, I can hardly doubt that the branchiæ in the Balanidæ are the ovigerous fræna of the Lepadidæ in a modified condition; a transformation of function not greater than that of the swimming bladder of a fish into the lungs of the higher Vertebrata.[29]

Thorax and Body.

Parts of the body included within the shell or carapace.—These parts (Pl. 25, fig. 1) consist of the prominent mouth, and of the thorax (c′), with its largely developed portion, called the prosoma (c), and with its appendages. The abdomen is quite rudimentary, being represented merely by a small portion of membrane surrounding the anus, and sometimes inserted like a wedge between the inwardly inflected posterior thoracic segments; in only two genera (Catophragmus and Pachylasma), its nature is rendered somewhat plainer by supporting caudal appendages. The probosciformed penis lies folded under the thorax; and I believe (from what is seen in the anomalous genus Proteolepas), that it normally arises from the ventral surface of the terminal point of the rudimentary abdomen.[30] The thorax is laterally compressed, the ventral surface being very narrow, with the bases of the cirri placed closely together. It consists, in appearance, of two very different portions; one a soft, more or less rounded bag, which I have called the prosoma; and the other, which supports the five posterior pairs of cirri, is narrower, invested with stiffer membrane, and is more or less distinctly composed of five segments. These segments (Pl. 26, fig. 8) on their dorsal and dorso-lateral surfaces, are generally driven like wedges one into the other, with their points directed anteriorly: on the ventral surface the articulations are transverse. The prolongation (e) of the thin membrane (a) surrounding the anus (b), that is, the rudiment of the abdomen, which sometimes carries caudal appendages, almost divides (in appearance, whether really I know not) the hindermost thoracic segment along the medio-dorsal line, into two parts. I have given the above drawing of these segments, but with the dorsal surface much flattened, in Coronula diadema; in most species of Balanus, however, the wedges formed by one segment being driven into another, are much sharper; on the other hand, in Xenobalanus they are nearly straight and transverse. The three posterior segments are always the most distinct; the two next segments are also distinct laterally, but along the dorsal surface they become, excepting in Xenobalanus and some few other cases, completely confluent. The greater distinctness of the posterior segments is conformable to what takes place in the higher Crustacea. The articulations between the segments are folded inwards, and are formed of thin membrane, which in some cases, as in Coronula diadema, forms a marked contrast with the much thicker, stiffer, and yellowish membrane of the segments themselves; in Balanus tintinnabulum, however, the whole membrane of the five thoracic segments is very thin, excepting small wedge-shaped portions along the medio-dorsal line. The infolded articulations between the segments supporting the three anterior pairs of cirri (at least in the Balaninæ), are much wider than those between the three posterior segments; the former segments, with their cirri, being consequently capable of being moved further apart from each other. Could there have been any doubt as to the distinctness and reality of the five thoracic segments, it would have been set aside by the arrangement of the muscles attached to them, as will presently be described. I need only add, that in many genera there are shield-like swellings at the exterior bases of the pedicels of the posterior cirri, which I for some time thought were the epimeral elements of the thoracic segments; but I now believe them to be parts of the pedicels of the cirri. The basi-exterior margin, moreover, of the pedicel of the third pair of cirri, in many species of the Balaninæ (Pl. 25, fig. 1), is produced as a plate, thickly fringed with fine hairs, half across the dorsal surface of the thorax; serving, apparently, as a brush to clean the sack, or to prevent the ingress of any intruding substance.

The soft, rounded, bag-like portion of the body, which I have called the prosoma, is usually separated by a notch from the five posterior thoracic segments; at its upper end it may be said to carry the mouth and first pair of cirri. The prosoma includes the main part of the stomach and the broad ends of the vesiculæ seminales. It is always clothed by very thin membrane, which in Chthamalus dentatus, is hairy. In Tubicinella and Xenobalanus, the prosoma is much elongated, being produced far down the deep sack. That the prosoma is mainly formed by a great development of that segment (homologically the second thoracic segment) which carries the first pair of cirri, is certain, and I should not have hesitated to have said that it was exclusively so formed, had not the first thoracic segment in the anomalous genus Cryptophialus been developed as a distinct and free segment, not attached to the carapace; showing that possibly in other Cirripedes, the dorsal half of this first thoracic segment may be concerned in the formation of the free prosoma.

Muscular System.

Attachment of the Body to the Shell.—The prosoma which carries the posterior thoracic segments, and in appearance the mouth, is the only part of the body which is attached to the general covering (Pl. 25, fig. 1), namely, to the opercular valves. Except through the continuity of the lining membranes of the sack, the body lies free within the walls of the shell. The area of attachment (shown by a sinuous broken line round a and b) extends from about the middle of the two scuta down to their basal margins. As these valves lie obliquely across the orifice of the shell, the animal’s body comes to be suspended almost in the middle of the sack. The two scuta, as we have seen, have the power of opening and shutting a little; and are brought together by the adductor scutorum muscle (a), which is generally very powerful. The body is attached to these valves, round and beneath the adductor, so as to hide it until one of the valves be removed. The attachment is chiefly effected by three pairs of widely expanded, superficial muscles, two pairs of which are spread over the flanks of the prosoma, and the third pair over its rounded (properly dorsal) surface, which lies close to the rostral compartment (A, fig. 1) of the shell. I should have stated, that my chief examination of the attachment of the body to the scutal valves, has been made on Coronula balænaris, and less closely on Balanus tintinnabulum. Within these three pairs of superficial muscles, there are (besides the adductor) no less than five other pairs; of these one long pair is attached at one end to the basal margin of the labrum (e), and at the other end, to the under side, near to the basal margin of the scuta: two other, shorter, parallel pairs of muscles are attached at one end to the interspace of membrane between the basal edge of the labrum and the adductor scutorum muscle, and at the other end, to the under side of the scuta, above the attachment of the first pair: the fourth and shortest pair curls close under the adductor, and is there attached at both ends beneath it. The action of these four pairs of muscles must be to draw back, from the orifice of the shell, the mouth, and that interspace of body between the basal margin of the labrum and the adductor muscle. This movement I saw in living specimens. The last and fifth pair of muscles is small, but of considerable length; it is a diverging pair, attached at the converging end, above and exteriorly to the adductor muscle; and at the diverging end, low down on the under side of the scuta; I am very doubtful regarding the function of this pair. Altogether we have seen that round and within the fleshy pedicel, by which the body is attached to the scuta, there are no less than eight pairs of muscles. The central space between these muscles is hollow, and here many lacunal channels seem to converge. These muscles receive nerves from the supra-œsophageal ganglions. The interspace above alluded to, between the basal edge of the labrum and the adductor scutorum muscle, occupies a very different position according as the animal’s body is protruded as far as it can be, or is retracted. It is homologically part of the third cephalic segment; and consequently the mouth ought to have stood posteriorly (i. e. above, in the position figured in Pl. 25, fig. 1) to this interspace; yet, in fact, when the animal is retracted within its shell, the mouth usually lies almost directly beneath this interspace and the adductor scutorum muscle.

Besides these muscles of attachment, the prosoma is furnished with several other muscles. There are superficial muscles running up towards the basal margin of the sides of the mouth; and other deeper muscles, to which, I presume, the movements of the mouth, as a whole, are due. The muscles moving the gnathites do not, as far as I could make out, extend beneath the basal edge of the mouth. There are, also, powerful muscles giving movement to the basal segments of the pedicel of the first pair of cirri. Again, there are superficial muscles running to the next succeeding thoracic segment; the anterior ends of which are separated by a clear interspace from the lower ends of the above-described superficial muscles, by which the prosoma is attached to the scuta. On each flank, moreover, but more deeply imbedded, are the long flexor and extensor muscles, presently to be described, running to the five posterior thoracic segments. The last muscle which I need here mention, is a deep-seated diverging pair, attached near the upper end of the stomach, on its ventral surface, and diverging from this point to the sides of the prosoma high up beneath the mouth. The probable action of this pair, as well as of the three superficial pairs of muscles by which the body is attached to the scuta, is to draw up the whole prosoma towards or from the orifice; and likewise to contract it firmly, so as to serve as a fulcrum for the movements of the five posterior thoracic segments, together with the cirri, which they carry.

The muscles of these five thoracic segments are numerous and powerful; they are also complicated, chiefly owing to the segments on their dorsal and dorso-lateral surfaces being driven, like wedges, one into the other. As far as I could make out, there are on each side three, superficial, dorso-lateral and lateral muscles (generally, if not always, destitute of striæ), which do not cross the articulations, but extend merely from articulation to articulation; and of which the function can be only to contract each separate segment, and consequently to open out the intermediate infolded articulations; the effect of this would be to separate slightly the cirri from each other,—more especially those borne on the two or three anterior segments, between which the infolded articulations are deeper or broader. There are other more deeply imbedded, powerful, long, dorso-lateral extensor, and ventri-lateral flexor muscles, attached at one end within the flanks of the prosoma, and at the other end to the successive segments of the thorax. The action of the former is to straighten and stretch out the thorax; of the latter, or ventri-lateral muscles, to retract it. In tracing these muscles, a fascia could be seen to become attached to a segment, and then this same fascia would run on to the next succeeding segment: the effect of this must be, that each segment can be retracted and protracted either from the prosoma as a fulcrum, or from the antecedent segment as a fulcrum: we have, also, seen that each segment can, by the agency of the superficial, non-striated muscles, contract itself. Hence these thoracic segments are capable of diverse movements, as was very evident when the shell of a living specimen was opened. By one movement in common, the whole five posterior segments could be drawn back, so as to become even partly imbedded in the prosoma: lateral, twisting or wriggling movements were also quite distinct: the three posterior segments seemed to be capable of less independent movements than the anterior segments; and I observed that the more powerful flexor and extensor muscles did not run into these three posterior segments. The cirri, of course, partake of the movements of the thorax; and in watching, in an uninjured specimen, the alternate, protruding, gracefully sweeping and retracting movements of the posterior pairs of cirri, it was evident that the thorax was the chief agent in their movement. Besides the muscles now mentioned, there are some immediately to be noticed, which extend from within the thoracic segments to within the pedicels of the cirri.

Movements and Muscles of the Cirri.

Although the cirri have not been described, it will be most convenient here to treat shortly of their muscles. Each cirrus consists of a pedicel, having a long basal and a short upper segment, supporting two multiarticulate rami. The lower segment of the pedicel can be drawn forward by an adductor muscle, attached low down within this segment, and crossing at right angles (at least in the case of the anterior cirri) the corresponding muscle of the opposed cirrus, on the central, ventral surface of the thorax. This segment can also be drawn back by a muscle springing from the dorso-lateral surface of the thorax, and running only a little way within the segment: I am far from sure that the lower segment does not possess other muscles. The short upper segment of the pedicel can be moved backwards and forwards, as I saw in living specimens, independently of the lower segment; this movement being best seen in the anterior cirri, which are much more often moved independently of each other than are the posterior cirri. The rami are capable, I believe, of being moved backwards and forwards as a whole, by the movement of the few lower segments, which are generally more or less confluent. They can, also, be curled up and uncurled by the combined movement of each separate segment. The uncurling seems to separate the two rami a little laterally. Each ramus, at least in the two or three anterior pairs, can be moved to a certain extent, independently of the other ramus of the same cirrus; and the few terminal segments, either of both rami or of one ramus, are often a little moved and curled (and this is especially the case with the long anterior ramus of the first pair), without the lower segments or the pedicel being moved.

The flexor and extensor muscles, which, as I believe, move the upper segment of the pedicel (a and b, Pl. 29, fig. 1), are attached at their upper ends to its basal margin, and are thus enabled to draw it a little way down within the lower segment, and so move it. The short flexor muscle (c), which is attached at its lower end within the upper segment of the pedicel, and the longer extensor (d), also, attached within this same lower segment, serve, I believe, to move the lower, partially confluent segments of each ramus as a whole. In the case of these muscles, and of those last mentioned, I am surprised that the extensors (b) and (d) are not attached nearer to the exterior and dorsal surface. Other muscles (e, f) attached at their lower ends within the upper segment of the pedicel, run up each of the two rami to their tips, with some of the fasciæ terminating within each segment: of these muscles, the outer one (f, f) appears to be the extensor, and the inner one (e, e) the flexor. But besides these, there are other short flexor muscles (g, g) which run on the anterior face,[31] from segment to segment, serving to pull the front edge of one segment within the edge of the next lower segment. These muscles differ much in plainness in the several genera: they are very distinct in Coronula. In some specimens of this genus, a few of the articulations between the basal segments of the rami having been obliterated, the short muscles (g, g) running from articulation to articulation were absent, and their presence and nature in the upper segments thus rendered the plainer. The muscular system in the several pairs of cirri seems to be the same, with the exception of the first pair, in which the muscle answering, as I suppose, to (a), namely, the flexor of the upper segment of the pedicel, is much spread out at its lower end, and is there attached to the exterior surface of the lower segment.

The backward and forward movements of the segments, both in the rami and in the pedicels of the cirri, are apparently effected, as already noticed, by the outer or inner (as the movement may be) basal edge of one segment being drawn a little way down within the next succeeding lower segment. If, at the same time, both the inner and outer margins of all the segments were drawn one within the other, the whole limb would necessarily be shortened; and I distinctly saw a shortening action, with very slight movement in any other direction, in the first and second pairs of cirri; and I think it almost certain that this movement might be performed by the other cirri. If I correctly understand a statement of Milne Edwards,[32] this is an important fact, as he asserts that only the higher Crustaceans possess the power of shortening their limbs.

When a Cirripede is alive, the action of the cirri is really beautiful: from the position of the thoracic segments, the posterior cirri (three pairs in the Balaninæ and four pairs in the Chthamalinæ) form a sort of semicircle facing the mouth: the anterior cirri stand further apart, and are opposed in pairs to each other, with the first pair pointing beyond the mouth. Together the cirri form a hollow cone, not circular but elongated, with the mouth situated at the lower anterior end. The posterior cirri are protruded, by the movement of the whole thorax, curled up, close along the carinal end of the orifice; as they are protruded, they diverge, both by the movement of their pedicels, and, as I believe, by the separation of the thoracic segments. As the two rami of each separate cirrus are uncurled, they also diverge a little; as do the double rows of spines on the segments in each ramus, by their elasticity. By the movement of the thorax, the cirri are then swept towards the rostrum; and, lastly, they are brought perpendicularly down towards the mouth with a rapid movement, which would be beautifully adapted to catch any object floating or swimming in the water; hence I have called the action of the cirri, captorial. When the shell of a Balanus is broken open, the second and third pairs of cirri are repeatedly clasped over the mouth with a convulsive movement, in a manner indicating, I think, that their chief function is to seize and carry to the mouth any object entangled by the sweeping movement of the three posterior pairs. The first pair is also well adapted to aid in this seizing action; but I suspect that the long anterior ramus likewise acts as an organ of touch, warning the animal of danger. The mouth being itself moveable as a whole,—the outer maxillæ being capable of a backward and forward sweeping action, and being furnished with orifices apparently olfactory,—the inner maxillæ having more diversified movements,—the toothed mandibles overhanging the œsophagus,—and the œsophagus itself possessing a powerful swallowing movement, are all admirably adapted to secure any prey, when once entangled by the cirri.

Mouth.

The mouth, in the sub-family Chthamalinæ, cannot be distinguished from that of the Lepadidæ, which has been pretty fully described in my former volume. In the Balaninæ, however, the labrum differs considerably in not being swollen; that is, in its outer and inner fold of membrane being close together, and in having a central notch: the palpi are also larger, and the lower teeth on the mandibles, are laterally (Pl. 26, fig. 5) double, as will be more particularly stated under these two sub-families. I have given a drawing (Pl. 26, fig. 1) of the mouth, seen from above, of Balanus perforatus, with the right-hand palpus (d′) and outer maxilla (a′) cut off, in order that the labrum (e), mandibles (c), and inner maxilla (b) might be better shown; the cut-off bases (x, x) of the first cirrus on each side are also shown. In fig. 2 we have the deep supra-œsophageal cavity in Bal. improvisus torn open and laid flat, with the inner surfaces of the labrum (c) and outer maxillæ (a) exhibited, the palpi, mandibles, and inner maxillæ having been removed. Figs. 3 and 4 will presently be referred to; they are parts of the mouth, with the muscles, &c. removed, of Coronula. The mouth differs extremely little in the different genera and species of the Balanidæ, much less than amongst the Lepadidæ. In the Balaninæ, the crest of the labrum is sometimes hairy, instead of having, as is usual, from two to six teeth on each side of the central notch: in Balanus improvisus (Pl. 26, fig. 2) and eburneus, and in Chelonobia, the crest on each side of the central notch (e′) is furnished with a row of finely graduated teeth. A sub-triangular portion of the inner fold of membrane of the labrum, which overhangs the œsophagus, is always thickened and yellowish; it is also often punctured in patterns (Pl. 26, fig. 2, f), which, I believe, give attachment to little muscles that serve to open the upper end of the œsophagus. Opposite to this thickened, sub-triangular portion of membrane, the thin membrane forming the supra-œsophageal cavity (or the cavity surrounded by the gnathites) is strengthened by a pair of curved ribs (h, fig. 2) of thickened yellowish membrane, running down from the inner bases (a′′) of the bilobed outer maxillæ to the opening of the œsophagus (g): a broad branch from each of these ribs supports the sides of the orifice of the œsophagus; and this branch almost joins on to a slightly thickened rim or bar (f′), which branches off from the upper part of the sub-triangular (f) inner fold of the labrum. This structure, in Bal. improvisus, is represented in Pl. 26, fig. 2, as well as it could be, considering that the deep supra-œsophageal cavity has to be torn open; and then laid flat.

The Palpi (Pl. 26, fig. 6) differ little, except in size, in the different genera, being squarish, more or less elongated, or even approaching to club-shaped: in most of the Balaninæ they are larger even than the mandibles, of which they normally form a part. Their upper margins, especially towards their free extremities, are always thickly clothed with spines; and there is generally a single row, either short (r) or long, of spines of greater length, which arise from a little above, and stand almost in a parallel line to, the basal margin. On the internal surface, there is sometimes a row (t) of very short little spines, which overhang the crest of the labrum. The Mandibles (Pl. 26, fig. 5) have from three to five teeth; the lower point or angle is generally pectinated. In Coronula and its close allies, there are some small teeth intermediate between the four or five main teeth; and in these genera, though members of the sub-family Balaninæ, the lower teeth exhibit only rudiments of being laterally double.[33] The Maxillæ sometimes have a notch under the upper large pair of spines, and in Octomeris brunnea there is a double notch: in many species of Balanus, the inferior corner stands up like a step (Pl. 26, fig. 7, a): in many other genera and species, the whole edge is straight. In all, or almost all cases, the row of spines on the middle portion is double. The Outer Maxillæ are always bilobed on their inner faces, and are clothed with bristles. On all the gnathites, the bristles are often doubly serrated.

Muscles and functions of the Gnathites, and their confluence.—The outer maxillæ appear at first like a deeply-lobed lower lip, for they reach over almost to the labrum (Pl. 26, fig. 1), and thus partially cover the other organs; they are separately capable of a strong and rapid, to and fro movement, by which no doubt they sweep any prey, entangled by the cirri, towards the other gnathites. Each outer maxilla is furnished with a pair of muscles, apparently a flexor and extensor; there is also a little muscle between the two maxillæ, I presume for the purpose of bringing them together. The outer and inner maxillæ generally stand close together, and in several genera a little way apart, from the mandibles; but there is no trace of any labrum or true lower lip, bounding the mandibles and orifice of the œsophagus. The outer and inner maxillæ and mandibles are not opposed in pairs to each other, but against the thickened inner fold of the labrum; almost in the same manner as the posterior pairs of cirri are not opposed one to the other, but to the mouth.

I have described pretty accurately the muscles of the mandibles in my former volume, and there given a drawing (Pl. 10, fig. 1) of them. There are four muscles: first, the depressor muscle, which is the largest, and is attached, at its upper end, to ligamentous apodemes under the free toothed portion of the jaw; and at its lower end, spreading considerably out, is attached to a concavity close above the basal margin of the labrum; to understand the action of these muscles, it should be borne in mind that the mandible almost faces the labrum. In some genera, as in Coronula,[34] the swelling near the basal margin of the labrum (Pl. 26, fig. 3, k), caused by the internal concavity for the above muscle, is conspicuous. The depressor muscle is opposed by a small elevator, attached to the mandible close by the depressor; thence it runs upwards, and is united at its upper end to the base of the palpus, at the point where the latter adheres to the labrum: I have ventured to call this muscle the elevator, from being apparently so well fitted for this purpose; but I feel some little doubt, from having observed an apparent slight movement in the palpi of living Balani; and this is the only muscle entering those organs. The free part of the mandible is articulated on a square, thickened piece of membrane, forming part of the side of the mouth (Pl. 26, figs. 3, 4, c¹; and Pl. 10, fig. 1, a, b, in my volume on the Lepadidæ); to this square piece of membrane, two short muscles are attached, one above the other, and which ought, in the Plate in my former volume, to have been represented crossing the depressor muscle at nearly right angles; at their further ends they are attached to about the middle of the labrum, where, at least in Coronula (Pl. 26, fig. 3, i), a slight concavity can be detected. The action of these two muscles must be to draw the whole mandible against the labrum; and the depressor muscle might, at the same time, draw the toothed edge downwards, and thus force any prey into the œsophagus.