The hydranths are usually very small in size; the largest known, a Monocaulus, does not exceed 1 1/2 in. in length. When they are attached to a branched stem, they are disposed either terminally or laterally on the branches; and in the latter case they may be pedunculate as in Tubulariae and Canipanularidae, or sessile and are then arranged either in a single row as in Plumularidae, or in a double row opposite or alternate one to reticulate hydrorhiza from which they spring; see Allman, Gymnoblastic Hydroids, pp. 54-5, and notes with Fig. 27. another as in Seriularidae. The hydranth itself varies somewhat in the shape of the hypostome, the relative proportions of the hydrocephalis and hydrocope, and the sharpness with which the latter is marked off from the former by a constriction or neck1. The tentacles of Hydra, Myriothela, and Polypodium (p. 766), are tubular, those of other Hydroids are, with a few partial exceptions, solid, possessing a notochord-like axis of modified endoderm cells generally disposed in a linear series. They are either filiform or capitate.
In the former the cnidoblasts are scattered along them in, for the most part, irregular groups containing micro- and macro-cnidia, intermingled; in the latter they are restricted to the swollen end, and contain almost exclusively macrocnidia. The tentacles are either arranged in a single adoral whirl, all similar in size, or alternately large or small, their bases sometimes united by a thin ectodermic lamella, e. g. in Campanopsis, Laomedea, etc.; or in two or more whirls of similar members, e. g. Tubularia, or of dissimilar, e. g. Stauridium, Cladonema, Pennaria, Tiarella. Or they are more or less irregularly scattered like the filiform tentacles of Clavidae, and the capitate tentacles of some Corynidae. In Cladocoryne the scattered tentacles are branched as well as capitate. They are usually numerous, but are reduced to two in Lar Sabellarum and Amphibrachium Euplectellae, to one in Monobrachium, and are absent altogether in Microhydra and Protohydra.
1 A living Hydractinia with calcareous skeleton has been described by Carter from West Africa (A. N. H. (4), xix. p. 50; ibid. (5), i. p. 300). Calcareous species from the Chalk and Pliocene, a pseudomorphic siliceous species from the Greensand are also known, together with extinct allied (?) genera: see Carter, op. cit.; Steinmann, von Meyer's Palaeontographica, xxv. 1878.
2Perigonimus Cidaritis (Weismann, Entstehung der Sexualzellen, etc. p. 116) is invested by a coat of mud, even extending on to the tentacles; P. palliatus has a gelatinous coat (Allman, op. cit. P. 325); P. vestitus a perisarc roughened by grains of sand (Allman, op. cit. p. 326). In the Tubularian Bimeria the perisarc covers the hydranth and the bases of the tentacles.
The hydranth is sometimes modified for special functions, and the following must be regarded as polymorphic forms of it. (1) The blastostyle, a hydranth comparatively rarely developed in Tubulariae, the function of which is solely to produce sexual zooids 2. It may possess a mouth and tentacles as in Podocoryne, the latter reduced to ectodermic knobs in Hy-dractinia, or tentacles and no mouth as in Myriothela; or it is, as is more commonly the case, e. g. in all Campamdariae, destitute of both structures. (2) The Dactylozooid, a mouthless hydranth, modified for solely defensive and offensive purposes. Such zooids are universal among Hydrocorallina. They are long and filiform in the Milleporidae, and bear a variable number of short scattered capitate tentacles, but in the Stylasteridae are devoid of tentacles. A circle of them surrounds each gastrozooid of Millepora and some Stylasteridae. They communicate by several apertures with the coenosarc. Under this heading should also be included the three following: (i) The spiral zooids of Podocoryne, mouthless and tentaculate, tubular in P. carnea, with a solid endodermal axis in P. Haeckelii: the similar zooids of Hydractinia, which are tubular and possess small ectodermic tentacles1.
1 The base of the hydranth of Eudendrium is surrounded by an ectodermal ring containing cnidoblasts separated by a furrow from a ring of gland cells. In E. ramosum some of the hydranths are furnished with a single basal horn-like process - the cnidophore - armed terminally with a battery of cnidoblasts, and capable of executing slow movements. See Weismann, op. cit. p. 94, and Mitth. Zool. Stat. Naples, iii. 1882. A glandular ring occurs in some Campanularidans: von Lendenfeld, Z. A. vi. p. 69.
3 According to Allman (op. cit. ante, p. 336), the blastostyle of Eudendrium is formed by the atrophy of the hydranth, every stage being present in one and the same colony. Weismann, on the contrary (op. cit. ante, pp. 97, 103), states that it differs from a hydranth in development and structure from the first. The hypostome is absent, but the female has a double circle of small tentacles and a minute mouth, organs absent in the male. Both Cladocoryne and Eudendrium ramosum (as well as other species of the latter genus, judging from Allman's figures) are instances in which gonophore-bearing hydranths and blastostyles may be found side by side. The blastostyle of the female Halecium gives origin to two hydranths at its apex (Allman, op. cit. p. 58, Fig. 29, and Hincks, op. cit. ante, i. p. 221). The blastostyle of the thecate Campanularians is expanded distally into a disc, hollow at least in some instances, in which rhythmical contractions, due to the presence of ectodermic muscles, occur in Eucopella, as in the gonophores of Coryne pusilla. The blastostyle in Eucopella and Laomedea is broken up into a number of tubes which open into the terminal hollow disc. The gonophores in Eucopella are borne upon one of these tubes, in Laomedea in connection with a central spadix.
See von Lendenfeld, Z. W. Z. xxxviii. pp. 537-544; Allman, op. cit. ante, pp. 47-8. and Fig. 20. The homology of the pedicle bearing medusae in Corymoj'pha, and Monocaulus or gonophores in Tubularia does not seem to be known. It always springs from the hydrocephalis.