Hexactinellidae. In this group of the siliceous sponges the skeleton is composed of six-armed spicules, the rays of which are almost invariably at right angles to each other (fig. 28, D). In the centre of each spicule are three canals, cutting each other at right angles and forming an axial tube. The spicules become very commonly fused together by amorphous silica, so as to form a trellis-work of rectangular or polyhedral meshes, the individual spicules of which are only recognisable by the persistence of their axial canals (fig. 28, C). The "fiesh-spicules" are fundamentally six-armed, but give off secondary branches so as to form a rosette.
Among the living Hexactinellidae, the Venus' Flower-basket (Euplectella) is one of the most familiar forms. In this exquisitely beautiful sponge, the skeleton-spicules are of large size, and the entire skeleton is at first flexible and soft, the spicules being free. Ultimately, the spicules become cemented together by a coating of vitreous silex, so as to form a ladder-like trellis-work. There is a single terminal osculum, provided with a porous lid; the sponge-body is rooted in the mud of the sea-bottom by a beard of long siliceous fibres; and the entire skeleton in the living state is completely concealed by a thick covering of brown sarcode. Another very interesting Hexactinellid sponge is the Hyalonema or "Glass-rope Zoophyte," long supposed to be a kind of coral. In this singular type, there is a comparatively small sponge-body, which is rooted to the mud of the sea-bottom by a long rope of delicate siliceous fibres. In addition to this skein of "anchoring-fibres," there are branched spicules, which are four-armed or five-armed in the recent forms, but are hexradiate in fossil examples. Other well-known living Hexactinellidae are Aphrocallistes, Farrea, Dactyl-ocalyx (fig. 28, A), etc. All the known forms are marine, and are inhabitants of deep water.
Lithistidae. These are siliceous sponges in which the spicules are essentially quadriradiate, three of the four arms being so disposed as to come together at an angle of 120o, while the fourth arm lies in a different plane to the others, and forms a cylindrical shaft from which the latter spring. The extremities of the arms of the spicules are divided into processes (fig. 28, B), and by the interlocking of these, contiguous spicules are united into a continuous skeleton, the meshes of which are more or less irregular and curvilinear.
Like the Hexactinellids, the Lithistidae are all marine, and inhabitants of deep water; Discodermia, Coralliste, M'Andrewia, Azorica, and Leiodermatium being well - known recent genera.
The reproduction of sponges may be effected either asexually or sexually, the following being a brief outline of the phenomena which have been observed in the common fresh-water sponge (Spon-gilla), in which the process was first accurately noticed.
In the first or asexual method of reproduction, which takes place in the winter, the deeper portions of the sponge are found to be filled with small seed-like rounded bodies, termed "gem-mules "or "spores," each of which possesses a small aperture or "hilum" at one point (fig. 29, h). Each gemmule is composed of an outer coriaceous capsule surrounded by a layer of peculiar asteroid spicula, resembling two toothed wheels united by an axle, and termed "amphidiscs" (fig. 29, b, c). These amphidiscs are embedded in sarcode, whilst their inner surfaces rest upon the tesselated capsule already mentioned. In the interior of the capsule thus formed is a mass of protoplasmic cells, which, on the coming of spring, is extruded through the hiliform opening of the capsule into the water, and becomes developed into a young Spongilla.
Fig. 29. - a Gemmule of Spon-gilla; h Hilum; b Diagrammatic section of the gemmule, showing the outer layer of amphidiscs and the inner mass of cells; c One of the amphidiscs seen in profile.
In the second or sexual method of reproduction, certain of the sponge-particles or "sarcoids" separate themselves and become nucleolo-nucleated, thus constituting ova. At the same time other sarcoids become motionless, and their contents become molecular, and are finally converted into spermatozoa. By the rupture of these, and by the consequent contact of the different elements (fig. 30, A), embryos are produced, which are at first ciliated and move about freely, becoming eventually stationary, and developing into new individuals.
As regards the development of the sponges, the impregnated ovum (fig. 30, A) cleaves, by the usual process of "segmentation," into a mass of primitive cells, sometimes containing centrally a primitive and temporary cavity (fig. 30, B). These cells are divisible into two distinct groups, one of which ultimately forms the external layer (ectoderm), whilst the other forms the internal layer (endoderm). As described by Metschnikoff in the embryo of Sycon, these groups of cells at first form the two poles of the larva, the cells of the endoderm being ciliated (fig. 30, B, c), and enabling the organism to swim actively through the water, whilst the cells of the ectoderm are non-ciliated (fig. 30, B, b). In the process of growth, the ciliated endodermal cells become gradually retracted into the interior of the larva (fig. 30, C and D), till the body becomes completely invagtnated upon itself. In this condition (fig. 30, E) it forms what Haeckel terms a "gastrula," and consists of two layers of cells, an outer and an inner, enclosing a central cavity, which communicates with the outer water by a single primitive opening. This aperture is formed by the invagination of the body, and not by rupture of the walls of the central cavity. The skeleton is formed in the ectodermal layer, and the primitive opening into the body-cavity becomes finally effaced. In its further development, the young sponge, now consisting of two cellular layers surrounding a closed central cavity (fig. 30, F), fixes itself by one extremity to some foreign object; a primitive "osculum" is developed at the free extremity of the larva ; and the walls become perforated with numerous small apertures, which ultimately become the inhalant openings or "pores" of the adult. It should be added that the account of the development of the Calci-spongiae given by Haeckel differs in some points very materially from the above.
Fig. 30. - Development of Calcispongiae. A, Ovum in the act of being impregnated by the spermatozoids. B, Free-swimming embryo of Sycon, showing the non-ciliated ectodermal cells, and the ciliated endodermal cells, the latter enclosing a temporary "segmentation-cavity" (a). C, The embryo further advanced, with the ciliated half of the body reduced in size. D, The embryo at a later stage, showing the primitive spicules, and the commencing body-cavity (d). E, Unattached larva, without the skeleton; the ciliated endoderm has now been withdrawn within the non-ciliated ectoderm, and the primitive opening into the body-cavity (e) has been formed by invagination. F, Young Sycon, six days old, showing the skeleton. b Non-ciliated ectodermal cells ; c Ciliated endodermal cells. (A is after Haeckel; B, C, D, E, and F are after Metschnikoff.)
It should be added, further, that according to the researches of Mr Saville Kent, the so-called "ciliated embryo" of the sponges is not composed of "cells," in the ordinary acceptation of this term, but is "a spherical or ovate aggregation of typical collar-bearing Monads or spongozoa, connected laterally and by their bases with one another, and with their anterior flagellate collar-bearing extremity directed outwards."