Marine, rarely freshwater, Coelenterata; free or fixed; simple or colonial. There are two forms of zooids, the Hydroid and the Medusa, the former asexual except in Hydra, the latter alone, or one of its degenerate forms, sexual. Both forms of zooids usually possess tentacles: the mouth is circular or square: the gastric cavity either simple, or provided with four ridges, or in the Medusa partly obliterated, leaving pouches or canals radiating from a central cavity. Sensory cells and ganglion cells may be found in the Hydroid; aggregated sense-cells, ocelli and auditory organs, nerve rings or ganglionic centres in the Medusa. The generative organs are either ecto- or endo-dermic. The sexual zooid is developed from the asexual, either directly by metamorphosis, or indirectly by gemmation or fission, thus giving rise to an Alternation of Generations.

There are two principal types of the Hydroid. One, the Hydromedusan or Craspedote type, consists typically of an oral and stomachal region (hydrocephalis), with or without tentacles, borne upon a peduncle (hydro-cope). The tentacles are variable in shape and disposition, rarely tubular, usually solid. The mouth is placed at the extremity of an oral cone, and is sometimes extremely dilatable; the gastric cavity is simple, but its endo-derm cells are sometimes thrown into ridges by the contraction of the musculature. The second or Acrasped type, the Scyphostoma, has a squarish mouth in the centre of a peristome, which is fringed by a circle of solid tentacles; a somewhat flask-shaped body and a short peduncle. Its gastric cavity is traversed by four equidistal longitudinal ridges, into which the mesoglaea enters. The Hydroid is (I) a permanent locomotor sexual form, multiplying by gemmation, but only temporarily colonial, - Hydra: (2) a larval form which passes by a metamorphosis into a Medusa, which may multiply by gemmation but is only temporarily colonial, its hydriform progeny alone in some instances attaining the Medusa-stage, - Trachyme-dusae, Acrasped Pelagia: (3) A non-sexual but permanent form, sometimes solitary, usually however multiplying by gemmation, which is rarely discontinuous, but as a rule continuous, giving origin to colonies, - most Hy-droidea, Acraspeda: (4) a locomotor sexual form, probably derived by the specialisation of (3), and never multiplying by gemmation - Acrasped De-pastridae and Lucernaridae.

The Medusa1 is a bell or disc-shaped organism with the mouth at the apex of a manubrium, which depends from the centre of the concavity of the bell or disc. The mesoglaea of the aboral aspect of the bell is much thickened to form the umbrella s. exumbrella, that of the oral remains thin, and is known as subumbrella. The rim of the bell may carry hollow or solid tentacles, and the mouth may have solid marginal tentacles. The gastric cavity in the bell is generally partially obliterated. Organs of special sense are generally present. Sexual organs are developed on the walls of the manubrium, gastric cavity, or radial canals, from the ectoderm, or in the gastric cavity from the endoderm. Asexual reproduction by gemmation, very rarely by fission, occurs in some Craspedota. The Medusa swims by alternate contractions and expansions of the bell. It is derived (i) by direct metamorphosis from a larval Hydroid, - Trachymedusae, Acrasped Pelagia: (2) by gemmation from a hydroid colony, - Uydroidea: (3) by the multiple transverse fission of a Hydroid, - most Acraspeda, (4) by gemmation from a Medusa, - some Craspedota.

1 See p. 247, ante, and Fig. 11.

The hydroid of the Craspedota may become polymorphic (pp. 757-8), and the Medusa be degenerate (pp. 762, 768) or devoid of mouth, and simply a locomotor or hydrostatic organism (pp. 771-2).

In one order, the Siphonophora, precocious gemmation takes place in the embryo to form colonies containing polymorphic hydroid individuals, and generally medusoid as well. The Siphonophora are consequently to be regarded as the most specialised group in the class.

Skeletal structures in the Hydrozoa are confined to the Hydroid, and are secreted by the ectoderm. They occur in two forms, most commonly as a chitinoid investment, the perisarc, more rarely as a calcareous coeno-steum. The axial endoderm cells of the solid tentacles have a tough cell membrane and vacuolated contents.

The sexes are, as a rule, separate in the individual. Segmentation is total, and generally equal: the larva a free swimming planula1.

The Hydrozoa are widely distributed and almost exclusively marine; a few are fresh-water, but some of the marine forms are tolerant of brackish or even of fresh-water (p. 748). They are essentially carnivorous. Some of the Campanularian Hydroidea, some Craspedote and Acrasped Medusae, and especially among the latter the genus Pelagia, some Siphonophora, e.g. Diphyes, A by la, etc, are phosphorescent. Fossil forms are rare. Two extinct groups, the Silurian Graptolithidae, and the Silurian and Devonian Stromatoporidae, are generally considered as Hydroids. A few Medusae have been described from the Jurassic Solenhofen slates, and one from the Chalk. One or two of them appear to belong to the Trachymedusae, the remainder to the Acraspeda.

There are two sub-classes, the Craspedota and Acraspeda.

1 Metschnikoff has classified the formation of the endoderm in this class as follows: (A) Multipolar, and then (1) by primary delamination, i. e. transverse fission of. the cells of the single layered embryo (Geryonidae, Eudendrium); (2) by immigration of cells on all sides (Aeginopsis); (3) by secondary delamination, i. e. arrangement of the cells (Aglaura, Rhopalonema, most degenerate Hydroidean Medusae); (4) by a combination of the foregoing (Polyxenia leucostyla). (B) hypo-tropic, and then by (1) immigration at one pole (Hydroidean Medusae); (2) by invagination (many Acraspeda). See Embryol. Studien an Medusen, Wien, 1886, pp. 70, 71.

Various views have been held as to the relation and nature of the Medusa. That it is merely a specialised sexual organ, or that it is a colonial organism, no person with a knowledge of its development and varieties of form, would maintain at the present day. It is generally held to be a zooid, polymorphic with a Hydroid and specialised for the purpose of reproduction, this function having become restricted to certain individuals of a Hydroid colony which are detached, and have acquired special facilities for locomotion in order to disseminate the race. Another view, strongly advocated by W. K. Brooks, which appears to have much in its favour, is to the effect that the primitive Hydrozoon was a free Hydroid 1 which gradually acquired a more and more perfect organisation for swimming: that when it had thus become a larval stage in an ontogeny, it began to multiply by gemmation: that its progeny were at first all detached and became Medusae, but that at a later stage of evolution some remained in connection with one another in the Hydroid stage, whilst the development of the Medusae became accelerated in situ: that finally the Medusa has in many instances become by acceleration of the development of its sexual products a degenerate sessile organism.

The points in favour of this view, as applied to the Craspedota, are, briefly put, the following: - (I) The direct development of the Trachymedusae, see pp. 752-3. (2) The existence of a free sexual Hydroidean - Hydra j and of a free hydroid larva, the Actinula, in two genera of Hydroidea, unless the latter is to be regarded as an instance of precocious development. (3) The fact that the development of the Medusa in the Hydroidea is abbreviated as shown (a) by the way in which the bell and velum are formed not by simple growth but from an entocodon or ecto-dermic thickening2 a structure not seen in Trachymedusae, and (b) by the fact that it is so commonly degenerate (p. 762), and (c) that it is very generally produced on the hydrocephalis, a place where hydranths never bud, or on a blastostyle, a specialised Hydroid. (4) The fact that among the Hydroidea closely related genera differ greatly in the character of their reproductive zooids (p. 768), and that the development of the sexual cells is accelerated in many instances (pp. 767-8). (5) That the Medusa is not purely a reproductive organism but feeds, etc, like any other organism, and that its specialisation is as Brooks says 'to enable it to live out its own life.' (6) The fact that a Medusa may reproduce itself, but as a Medusa only, by budding.

What is true of the Craspedota appears to be true also of the Acrasfieda. Pelagia developes directly from the ovum; others, so far as their development is known, by strobilisation (p. 782), a process which has probably been attained by abbreviation, though it is perhaps not necessary to assume with Claus that the primitive Acrasped Hydroid was a colonial organism (cf. Untersuchungen uber die Organisation, etc, der Medusen, p. 18). Gotte has recently advanced the view that the subclass in question should be separated from the Hydrozoa, and that it should be united with the Anthozoa and Ctenophora in a division of Coelenterata to be termed 'Scyphozoa.' But he appears to exaggerate the degree to which the ectoderm is invaginated as a stomodaeum in the Acraspeda, and to lay too much stress on the four taeniolae and stomach pouches. The strong general anatomical resemblances between the Acrasped and Craspedote Medusae outweigh any such considerations. See his Abhandlungen zur Entwickelungsgeschichte derThiere, pt. iv. 1886.

1Brooks says 'a solitary swimming hydra or actinula' Why 'swimming' is not clear. The natatory Hydroid of the Trachymedusae is a larval form, but a free adult would more probably be a creeping form, much like Hydra. Swimming, unless effected by cilia, requires a specialised organ for the purpose.

2For an endodermic entocodon in Coryne pusilla see Weismann, Die Entstehung der Sexualzellen bei der Hydromedusen, p. 53, and for the way in which an entocodon may have originated, ibid. pp. 259-60.

Metagenesis, or Alternation of Generations in the Hydrozoa, has therefore probably arisen from the larva acquiring the power of forming a colony by gemmation, certain of its progeny only attaining a sexual development. It has itself become at the same time specialised in structure.

The genus Hydra is probably a direct descendant of the primitive Hydrozoon. Its peculiarities point in this direction. The ambulatory Medusae Clavatella and Eleutheria have been supposed to be intermediate forms between a hydranth and a Medusa, but an entocodon is present in the development of the latter by budding from the Medusa.

For the remarkable form known as Tetraplatia s. Tetrapteron volitans, see Claus, A. M. A. xv. 1872; Viguier, C. R. 100, 1885.

Toleration of freshwater by strictly marine forms, e. g. Eucope, Obelia, Sarsia, Turritopsis, Aurelia and its Scyphostoma, Q. J. M. xx. 1880, p. 483 - 4. Crambessa Tagi is estuarine. Cf. Romanes on the physiology of the freshwater Medusa, Nature, xxii. p. 179.

Origin of Medusae and the significance of Metagenesis, W. K. Brooks, Mem. Boston Soc. Nat. Hist. iii. pt. 12, 1886.

Tenns used in describing Medusae. Haeckel, Deep-sea Medusae, Challenger Reports, iv. 1882, pp. v-cv, pp. 143-154.

Fossil Medusae. List in Haeckel's System, Dk. Med. Natw. Ges. Jena, i. 1879, p. 646.

Phosphorescence, Mcintosh, Nature, xxxii. p. 477; Panceri, Atti Accad. Napoli, vii. 1875-7; Allman, Gymnoblastic Hydroids, p. 145; cf. Nature, xxx, Verrill, p. 281, Meldola, p. 289.