Anthozoa which may be simple or colonial, and in the latter case furnished as a rule with either an organic or calcareous contimwus skeleton derived from the basal ectoderm. The tentacles are usually simple. They are arranged in one or more circles, the members of a circle or of different circles being often dissimilar in size. The mesenteries and the retractor muscles are never disposed as in Alcyonaria: the former are very generally paired, and then tentacles may correspond to the inter- as well as to the intra-septal spaces. The number of both tentacles and mesenteries is very generally, but not universally, some multiple of six1. One or two siphonoglyphes are of ten present. Dimorphism is known only in one instance.

1The Zoantharian mesenterial filaments are described on p. 241. It has been suggested by Wilson that the median lobe is of ectodermic, the two lateral lobes of endodermic origin. Histology, as pointed out by Fowler (Q. J. M. xxvii. p. 8), points in the opposite direction. But there is one difficulty in the way of any such derivation, viz. the filaments of incomplete mesenteries are constructed on the same type as are those of complete. As the former set of mesenteries are never in contact with the stomodaeum, it is not possible for them to acquire a band of ectoderm cells. See on Alcyonarians, p. 730, ante.

There are three distinct subdivisions of living Zoantharia, the Actini-aria, Antipatharia and Madreporaria.

The Actiniaria or Malacodermata are distinguished in the first place by the absence of an organic or inorganic skeleton. The animal is usually-solitary, and either free, or adherent to the surfaces of foreign bodies, and creeps about by means of a pedal disc, or it lives immersed in sand or mud. It is rarely colonial and fixed. There are six tribes.

(I) The Hexactiniae have all the mesenteries paired. Two pairs placed one at each end of the mouth, termed directive mesenteries, to each of which corresponds a siphonoglyphe, differ from the remaining pairs which are generally numerous, in having the retractor muscles on the interseptal surfaces instead of the intraseptal. The number of primary pairs of mesenteries is six, two directive and two lateral on each side of the body. In the Sagartidae and Amphianthidae the six pairs in question are the only complete mesenteries, whereas in other Hexactinians, so far as is known, the secondary mesenteries are also complete: the tertiary and quaternary are always incomplete. The tentacles are numerous, and frequently of great length. There is always a marginal set, and sometimes an intermediate as well as circumoral, - the two latter often termed 'accessory1.' They differ, like the mesenteries, in age, and therefore sometimes in size, but there is a general tendency to equalisation in this respect. When they are numerous they are arranged in concentric circles, the oldest nearest to the centre of the disc. Structurally speaking, they are evagina-tions of the oral disc, which correspond to both the intra- and inter-septal spaces.

They are usually contractile, and furnished with a terminal pore. They become altered in character in some deep sea forms; i.e. the terminal pore enlarges; the tentacle itself may be reduced to a short wide-mouthed tube or stomidium, or to an aperture with ring-like margin, which may almost disappear; and finally in Liponema it becomes a simple opening in the peristome. When the animal is irritated the tentacles shorten. At the same time the outermost margin of the disc is very generally strongly contracted over the tentacles and mouth by the action of an endodermal sphincter or Rotteken's muscle (p. 240), rarely completely absent. Acontia (p. 241) are present in the Phellidae and Sagartidae, and may be protruded through the mouth, by cinclides, or by rupture of the body-wall. The inner stomata are always, the outer sometimes, present. The genital products are either borne upon all the mesenteries, e. g. in Corallimorphidae, or they are confined to those of lower order, and are at least not developed on the primary, e.g. in Sagartidae and Liponemidae1. The Amphianthidae live attached to fragments of the stems of Axifera, and like Gerardia (infra, p. 737) have the sagittal or antero-posterior axis short, and the transverse elongated.

In Halcampa the aboral end of the body is rounded, and it is pierced by numerous (24?) small openings. Its mesenteries are reduced to the primary six pairs2. The Hexactiniae live usually attached to rocks, or when the base is feebly developed, immersed in sand or mud. The Minyadinae, which probably belong to this tribe, are pelagic. The base may be inverted and serve as an air vesicle, but in Nautactis is capable of eversion and attachment. It is perforated in Dactylominyas (Oceanactis) rhododactyla.

1In the family Stichodactylinae the disc is large and covered with tentacles in radial series either of one form and simple, or of two and then simple, mixed with ramose or foliate tentacles, or of various shapes. So too in the family Thalassianthinae the disc is covered with dendritic appendages. See Andres, 'Le Attinie,' Monograph ix. (I), 1884, p. 264 and p. 299.

(2) The Paractiniae of Hertwig differ from the Hexactiniae in having the mesenteries multiples of 8; Sicyonis has 64: but Polyopis 36, there being in Hertwig's opinion two redundant pairs in this case. Polyopis has the tentacles reduced to stomidia, and has a rounded aboral pole.

(3) The Monaideae, represented by Scytophorus striates, have seven pairs of complete mesenteries, one only of which is directive. The single siphonoglyphe corresponds to it. The animal is hermaphrodite: the testes are situated near to the base, the ovaries at a higher level.

(4) The Edwardsiae possess eight mesenteries, all complete, all bearing reproductive organs. There are two pairs of directive mesenteries; the remainder are not paired and have the retractor muscles turned in the direction of one of the two pairs of directive mesenteries, hence termed ventral. There are two siphonoglyphes. The arrangement is therefore neither Alcyonarian nor Hexactinian. The tentacles vary in number with age (12-36), and are more numerous than the mesenteries, a fact which probably denotes the loss of mesenteries by reduction. The body is divisible into three regions; the middle one (=scapus) is covered by a tough mucoid investment; the posterior is pointed but not perforated. The animals live immersed in sand or mud, in holes of rocks or under the roots of seaweeds.

(5) The Zoanthidae have the mesenteries paired as in the Hexactiniae, but each pair consists typically of a complete mesentery, the macro-septum, which has a mesenterial filament, and bears genital organs, and of an incomplete mesentery, the micro-septum devoid of both the structures named. The directive mesenteries are as in Hexactiniae, but one pair con1In Ophiodiscus the mesenteries of the fourth order are devoid of mesenterial filaments and muscles, and bear the genital organs, the mesenteries of higher orders being muscular and sterile. So too perhaps in Polystomidium. See also note, p. 726.

2The base is rounded also in Ilyanihus, in Siphonactinia (= Peachid) and Philomedusa, but whether perforated or not is uncertain. The mesenteries and retractor muscles of Siphonactinia are apparently restricted in number and peculiarly disposed, its base perforate according to Faurot, C. R. 98, 1884. It must be borne in mind that the present arrangement of the Hexactiniae in families, etc. is not founded throughout on anatomical structures: it can only be regarded as temporary.