[Ill.u.s.tration: FIG. 13.

A, _Cerianthus solitarius_ (after A. Andres).

B, Transverse section of the stomodaeum, showing the sulculus, sl, and the arrangement of the mesenteries.

C, Oral aspect of _Arachnactis brachiolata_, the larva of _Cerianthus_, with seven tentacles.

D, Transverse section of an older larva. The numerals indicate the order of development of the mesenteries.]

The order ANTIPATHIDEA is a well-defined group whose affinities are more obscure. The type form, _Antipathes dichotoma_ (fig. 14), forms arborescent colonies consisting of numerous zooids arranged in a single series along one surface of a branched h.o.r.n.y axis. Each zooid has six tentacles; the stomodaeum is elongate, but the sulcus and sulculus are very feebly represented. There are ten mesenteries in which the musculature is so little developed as to be almost indistinguishable. The sulcar and sulcular pairs of mesenteries are short, the sulco-lateral and sulculo-lateral pairs are a little longer, but the two transverse are very large and are the only mesenteries which bear gonads. As the development of the Antipathidea is unknown, it is impossible to say what is the sequence of the mesenterial development, but in _Leiopathes glaberrima_, a genus with twelve mesenteries, there are distinct indications of an Edwardsia stage.

[Ill.u.s.tration: FIG. 14.

A, Portion of a colony of _Antipathes dichotoma_.

B, Single zooid and axis of the same magnified. m, Mouth; mf mesenterial filament; ax, axis.

C, Transverse section through the oral cone of _Antipath.e.l.la minor_, st, Stomodaeum; ov, ovary.]

There are, in addition to these groups, several genera of Actinians whose mesenterial arrangement differs from the normal type. Of these perhaps the most interesting is _Gonactinia prolifera_ (fig. 11, B), with eight macromesenteries arranged on the Edwardsian plan. Two pairs of micromesenteries form couples with the first and second Edwardsian pairs, and in addition there is a couple of micromesenteries in each of the sulculo-lateral exocoeles. Only the first and second pairs of Edwardsian macromesenteries are fertile, i.e. bear gonads.

The remaining forms, the ACTINIIDEA, are divisible into the Malacactiniae, or soft-bodied sea-anemones, which have already been described sufficiently in the course of this article, and the Scleractiniae (= Madreporaria) or true corals.

[Ill.u.s.tration: FIG. 15.--Corallum of _Caryophyllia_; semi-diagrammatic.

th, Theca; c, costae; sp, septa; p, palus; col, columella.]

All recent corals, as has already been said, conform so closely to the anatomy of normal Actinians that they cannot be cla.s.sified apart from them, except that they are distinguished by the possession of a calcareous skeleton. This skeleton is largely composed of a number of radiating plates or _septa_, and it differs both in origin and structure from the calcareous skeleton of all Alcyonaria except Heliopora. It is formed, not from fused spicules, but as a secretion of a special layer of cells derived from the basal ectoderm, and known as _calicoblasts_.

The skeleton or corallum of a typical solitary coral--the common Devonshire cup-coral _Caryophyllia smithii_ (fig. 15) is a good example--exhibits the followings parts:--(1) The _basal plate_, between the zooid and the surface of attachment. (2) The _septa_, radial plates of calcite reaching from the periphery nearly or quite to the centre of the coral-cup or calicle. (3) The _theca_ or wall, which in many corals is not an independent structure, but is formed by the conjoined thickened peripheral ends of the septa. (4) The _columella_, a structure which occupies the centre of the calicle, and may arise from the basal plate, when it is called essential, or may be formed by union of trabecular offsets of the septa, when it is called unessential. (5) The _costae_, longitudinal ribs or rows of spines on the outer surface of the theca. True costae always correspond to the septa, and are in fact the peripheral edges of the latter. (6) _Epitheca_, an offset of the basal plate which surrounds the base of the theca in a ring-like manner, and in some corals may take the place of a true theca. (7) _Pali_, spinous or blade-like upgrowths from the bottom of the calicle, which project between the inner edges of certain septa and the columella. In addition to these parts the following structures may exist in corals:--_Dissepiments_ are oblique calcareous part.i.tions, stretching from septum to septum, and closing the interseptal chambers below. The whole system of dissepiments in any given calicle is often called _endotheca_. _Synapticulae_ are calcareous bars uniting adjacent septa.

_Tabulae_ are stout horizontal part.i.tions traversing the centre of the calicle and dividing it into as many superimposed chambers. The septa in recent corals always bear a definite relation to the mesenteries, being found either in every entocoele or in every entocoele and exocoele.

Hence in corals in which there is only a single cycle of mesenteries the septa are correspondingly few in number; where several cycles of mesenteries are present the septa are correspondingly numerous. In some cases--e.g. in some species of _Madrepora_--only two septa are fully developed, the remainder being very feebly represented.

[Ill.u.s.tration: FIG. 16.--Tangential section of a larva of _Astroides calicularis_ which has fixed itself on a piece of cork. ec, Ectoderm; en, endoderm; mg, mesogloea; m, m, mesenteries; s, septum; b, basal plate formed of ellipsoids of carbonate of lime secreted by the basal ectoderm; ep, epitheca. (After von Koch.)]

Though the corallum appears to live within the zooid, it is morphologically external to it, as is best shown by its developmental history. The larvae of corals are free swimming ciliated forms known as planulae, and they do not acquire a corallum until they fix themselves.

A ring-shaped plate of calcite, secreted by the ectoderm, is then formed, lying between the embryo and the surface of attachment. As the mesenteries are formed, the endoderm of the basal disk lying above the basal plate is raised up in the form of radiating folds. There may be six of these folds, one in each entocoele of the primary cycle of mesenteries, or there may be twelve, one in each exocoele and entocoele.

The ectoderm beneath each fold becomes detached from the surface of the basal plate, and both it and the mesogloea are folded conformably with the endoderm. The cells forming the limbs of the ectodermic folds secrete nodules of calcite, and these, fusing together, give rise to six (or twelve) vertical radial plates or septa. As growth proceeds new septa are formed simultaneously with the new couples of secondary mesenteries. In some corals, in which all the septa are entocoelic, each new system is embraced by a mesenteric couple; in others, in which the septa are both entocoelic and exocoelic, three septa are formed in every chamber between two primary mesenterial couples, one in the entocoele of the newly formed mesenterial couple of the secondary cycle, and one in each exocoele between a primary and a secondary couple. These latter are in turn embraced by the couples of the tertiary cycle of mesenteries, and new septa are formed in the exocoeles on either side of them, and so forth.

[Ill.u.s.tration: FIG. 17.--Transverse section through a zooid of _Cladocora_. The corallum shaded with dots, the mesogloea represented by a thick line. Thirty-two septa are present, six in the entocoeles of the primary cycle of mesenteries, I; six in the entocoeles of the secondary cycle of mesenteries, II; four in the entocoeles of the tertiary cycle of mesenteries, III, only four pairs of the latter being developed; and sixteen in the entocoeles between the mesenterial pairs. D, D, Directive mesenteries; st, stomodaeum. (After Duerden.)]

It is evident from an inspection of figs. 16 and 17 that every septum is covered by a fold of endoderm, mesogloea, and ectoderm, and is in fact pushed into the cavity of the zooid from without. The zooid then is, as it were, moulded upon the corallum. When fully extended, the upper part of the zooid projects for some distance out of the calicle, and its wall is reflected for some distance over the lip of the latter, forming a fold of soft tissue extending to a greater or less distance over the theca, and containing in most cases a cavity continuous over the lip of the calicle with the coelenteron. This fold of tissue is known as the _edge-zone_ In some corals the septa are solid imperforate plates of calcite, and their peripheral ends are either firmly welded together, or are united by interst.i.tial pieces so as to form imperforate theca. In others the peripheral ends of the septa are united only by bars or trabeculae, so that the theca is perforate, and in many such perforate corals the septa themselves are pierced by numerous perforations. In the former, which have been called aporose corals, the only communication between the cavity of the edge-zone and the general cavity of the zooid is by way of the lip of the calicle; in the latter, or perforate corals, the theca is permeated by numerous branching and anastomosing ca.n.a.ls lined by endoderm, which place the cavity of the edge-zone in communication with the general cavity of the zooid.

[Ill.u.s.tration: FIG. 18.

A, Schematic longitudinal section through a zooid and bud of _Stylophora digitata_. In A, B, and C the thick black lines represent the soft tissues; the corallum is dotted. s, Stomodaeum; c, c, coenosarc; col, columella, T tabulae.

B, Similar section through a single zooid and bud of _Astroides calicularis_.

C, Similar section through three corallites of _Lophohelia prolifera_.

ez, Edge-zone.

D, Diagram ill.u.s.trating the process of budding by unequal division.

E, Section through a dividing calicle of _Mussa_, showing the union of two septa in the plane of division and the origin of new septa at right angles to them.

(C original; the rest after von Koch.)]

A large number of corals, both aporose and perforate, are colonial. The colonies are produced by either budding or division. In the former case the young daughter zooid, with its corallum, arises wholly outside the cavity of the parent zooid, and the component parts of the young corallum, septa, theca, columella, &c., are formed anew in every individual produced. In division a vertical constriction divides a zooid into two equal or unequal parts, and the several parts of the two corals thus produced are severally derived from the corresponding parts of the dividing corallum. In colonial corals a bud is always formed from the edge-zone, and this bud develops into a new zooid with its corallum. The cavity of the bud in an aporose coral (fig. 18, A, C) does not communicate directly with that of the parent form, but through the medium of the edge-zone. As growth proceeds, and parent and bud become separated farther from one another, the edge-zone forms a sheet of soft tissue, bridging over the s.p.a.ce between the two, and resting upon projecting spines of the corallum. This sheet of tissue is called the _coenosarc_. Its lower surface is clothed with a layer of calicoblasts which continue to secrete carbonate of lime, giving rise to a secondary deposit which more or less fills up the s.p.a.ces between the individual coralla, and is distinguished as _coenenchyme_. This coenenchyme may be scanty, or may be so abundant that the individual corallites produced by budding seem to be immersed in it. Budding takes place in an a.n.a.logous manner in perforate corals (fig. 18, B), but the presence of the ca.n.a.l system in the perforate theca leads to a modification of the process.

Buds arise from the edge-zone which already communicate with the cavity of the zooid by the ca.n.a.ls. As the buds develop the ca.n.a.l system becomes much extended, and calcareous tissue is deposited between the network of ca.n.a.ls, the confluent edge-zones of mother zooid and bud forming a coenosarc. As the process continues a number of calicles are formed, imbedded in a spongy tissue in which the ca.n.a.ls ramify, and it is impossible to say where the theca of one corallite ends and that of another begins. In the formation of colonies by division a constriction at right angles to the long axis of the mouth involves first the mouth, then the peristome, and finally the calyx itself, so that the previously single corallite becomes divided into two (fig. 18, E). After division the corallites continue to grow upwards, and their zooids may remain united by a bridge of soft tissue or coenosarc. But in some cases, as they grow farther apart, this continuity is broken, each corallite has its own edge-zone, and internal continuity is also broken by the formation of dissepiments within each calicle, all organic connexion between the two zooids being eventually lost. Ma.s.sive meandrine corals are produced by continual repet.i.tion of a process of incomplete division, involving the mouth and to some extent the peristome: the calyx, however, does not divide, but elongates to form a characteristic meandrine channel containing several zooid mouths.

Corals have been divided into _Aporosa_ and _Perforata_, according as the theca and septa are compact and solid, or are perforated by pores containing ca.n.a.ls lined by endoderm. The division is in many respects convenient for descriptive purposes, but recent researches show that it does not accurately represent the relationships of the different families. Various attempts have been made to cla.s.sify corals according to the arrangement of the septa, the characters of the theca, the microscopic structure of the corallum, and the anatomy of the soft parts. The last-named method has proved little more than that there is a remarkable similarity between the zooids of all recent corals, the differences which have been brought to light being for the most part secondary and valueless for cla.s.sificatory purposes. On the other hand, the study of the anatomy and development of the zooids has thrown much light upon the manner in which the corallum is formed, and it is now possible to infer the structure of the soft parts from a microscopical examination of the septa, theca, &c., with the result that unexpected relationships have been shown to exist between corals previously supposed to stand far apart. This has been particularly the case with the group of Palaeozoic corals formerly cla.s.sed together as _Rugosa_. In many of these so-called rugose forms the septa have a characteristic arrangement, differing from that of recent corals chiefly in the fact that they show a tetrameral instead of a hexameral symmetry. Thus in the family _Stauridae_ there are four chief septa whose inner ends unite in the middle of the calicle to form a false columella, and in the _Zaphrentidae_ there are many instances of an arrangement, such as that depicted in fig. 19, which represents the septal arrangement of _Streptelasma corniculum_ from the lower Silurian. In this coral the calicle is divided into quadrants by four princ.i.p.al septa, the _main septum, counter septum_, and two _alar septa_. The remaining septa are so disposed that in the quadrants ab.u.t.ting on the chief septum they converge towards that septum, whilst in the other quadrants they converge towards the alar septa. The secondary septa show a regular gradation in size, and, a.s.suming that the smallest were the most recently formed, it will be noticed that in the chief quadrants the youngest septa lie nearest to the main septum; in the other quadrants the youngest septa lie nearest to the alar septa. This arrangement, however, is by no means characteristic even of the Zaphrentidae, and in the family _Cyathophyllidae_ most of the genera exhibit a radial symmetry in which no trace of the bilateral arrangement described above is recognizable, and indeed in the genus _Cyathophyllum_ itself a radial arrangement is the rule. The connexion between the Cyathophyllidae and modern Astraeidae is shown by _Moseleya latistellata_, a living reef-building coral from Torres Strait. The general structure of this coral leaves no doubt that it is closely allied to the Astraeidae, but in the young calicles a tetrameral symmetry is indicated by the presence of four large septa placed at right angles to one another. Again, in the family _Amphiastraeidae_ there is commonly a single septum much larger than the rest, and it has been shown that in the young calicles, e.g. of _Thecidiosmilia_, two septa, corresponding to the main- and counter-septa of Streptelasma, are first formed, then two alar septa, and afterwards the remaining septa, the latter taking on a generally radial arrangement, though the original bilaterality is marked by the preponderance of the main septum. As the microscopic character of the corallum of these extinct forms agrees with that of recent corals, it may be a.s.sumed that the anatomy of the soft parts also was similar, and the tetrameral arrangement, when present, may obviously be referred to a stage when only the first two pairs of Edwardsian mesenteries were present and septa were formed in the intervals between them.

[Ill.u.s.tration: FIG. 19.--Diagram of the arrangement of the septa in a Zaphrentid coral. m, Main septum; c, counter septum; t, t, alar septa.]

s.p.a.ce forbids a discussion of the proposals to cla.s.sify corals after the minute structure of their coralla, but it will suffice to say that it has been shown that the septa of all corals are built up of a number of curved bars called trabeculae, each of which is composed of a number of nodes. In many secondary corals (_Cyclolites, Thamnastraea_) the trabeculae are so far separate that the individual bars are easily recognizable, and each looks something like a bamboo owing to the thickening of the two ends of each node. The trabeculae are united together by these thickened internodes, and the result is a fenestrated septum, which in older septa may become solid and aporose by continual deposit of calcite in the fenestrae. Each node of a trabecula may be simple, i.e. have only one centre of calcification, or may be compound.

The septa of modern perforate corals are shown to have a structure nearly identical with that of the secondary forms, but the trabeculae and their nodes are only apparent on microscopical examination. The aporose corals, too, have a practically identical structure, their compactness being due to the union of the trabeculae throughout their entire lengths instead of at intervals, as in the Perforata. Further, the trabeculae may be evenly s.p.a.ced throughout the septum, or may be grouped together, and this feature is probably of value in estimating the affinities of corals. (For an account of coral formations see CORAL-REEFS.)

In the present state of our knowledge the Zoantharia in which a primary cycle of six couples of mesenteries is (or may be inferred to be) completed by the addition of two pairs to the eight Edwardsian mesenteries, and succeeding cycles are formed in the exocoeles of the pre-existing mesenterial cycles, may be cla.s.sed in an order ACTINIIDEA, and this may be divided into the suborders _Malacactiniae_, comprising the soft-bodied Actinians, such as _Actinia, Sagartia, Bunodes_, &c., and the _Scleractiniae_, comprising the corals. The Scleractiniae may best be divided into groups of families which appear to be most closely related to one another, but it should not be forgotten that there is great reason to believe that many if not most of the extinct corals must have differed from modern Actiniidea in mesenterial characters, and may have only possessed Edwardsian mesenteries, or even have possessed only four mesenteries, in this respect showing close affinities to the Stauromedusae. Moreover, there are some modern corals in which the secondary cycle of mesenteries departs from the Actinian plan. For example, J.E. Duerden has shown that in _Porites_ the ordinary zooids possess only six couples of mesenteries arranged on the Actinian plan.

But some zooids grow to a larger size and develop a number of additional mesenteries, which arise either in the sulcar or the sulcular entocoele, much in the same manner as in Cerianthus. Bearing this in mind, the following arrangement may be taken to represent the most recent knowledge of coral structure:--

GROUP A.

Family I. ZAPHRENTIDAE.--Solitary Palaeozoic corals with an epithecal wall. Septa numerous, arranged pinnately with regard to four princ.i.p.al septa. Tabulae present. One or more pits or fossulae present in the calicle. Typical genera--_Zaphrentis_, Raf. _Amplexus_, M. Edw. and H.

_Streptelasma_, Hall. _Omphyma_, Raf.

Family 2. TURBINOLIDAE.--Solitary, rarely colonial corals, with radially arranged septa and without tabulae. Typical genera--_Flabellum_, Lesson. _Turbinolia_, M. Edw. and H.

_Caryophyllia_, Lamarck. _Sphenotrochus_, Moseley, &c.

Family 3. AMPHIASTRAEIDAE.--Mainly colonial, rarely solitary corals, with radial septa, but bilateral arrangement indicated by persistence of a main septum. Typical genera--_Amphiastraea_, etallon.

_Thecidiosmilia_.

Family 4. STYLINIDAE.--Colonial corals allied to the Amphiastraeidae, but with radially symmetrical septa arranged in cycles. Typical genera--_Stylina_, Lamarck (Jura.s.sic). _Convexastraea_, D'Orb.

(Jura.s.sic). _Isastraea_, M. Edw. and H.(Jura.s.sic). Ogilvie refers the modern genus _Galaxea_ to this family.

GROUP B.

Family 5. OCULINIDAE.--Branching or ma.s.sive aporose corals, the calices projecting above the level of a compact coenenchyme formed from the coenosarc which covers the exterior of the corallum. Typical genera--_Lophohelia_, M. Edw. and H. _Oculina_, M. Edw. and H.

Family 6. POCILLOPORIDAE.--Colonial branching aporose corals, with small calices sunk in the coenenchyme. Tabulae present, and two larger septa, an axial and abaxial, are always present, with traces of ten smaller septa. Typical genera--_Pocillopora_, Lamarck. _Seriatopora_, Lamarck.

Family 7. MADREPORIDAE.--Colonial branching or palmate perforate corals, with abundant trabecular coenenchyme. Theca porous; septa compact and reduced in number. Typical genera--_Madrepora_, Linn.

_Turbinaria_, Oken. _Montipora_, Quoy and G.

Family 8. PORITIDAE.--Incrusting or ma.s.sive colonial perforate corals; calices usually in contact by their edges, sometimes disjunct and immersed in coenenchyme. Theca and septa perforate. Typical genera--_Porites_, M. Edw. and H. _Goniopora_, Quoy and G.

_Rhodaraea_, M. Edw. and H.