The fact that all the principal groups of the Actiniaria, except, perhaps, the group that includes those forms with only eight mesenteries, the Edwardsiæ, have representative genera or species in the great depths of the ocean, points to the conclusion that the sea anemones have migrated from the shallow waters in comparatively recent times, and that the migrations have been successive, each period of their history sending some specimens to survive or to become extinct in the struggle for life in the deep sea.
Of the Madreporarian corals, several genera are now known to inhabit very deep water, but they do not present many very remarkable points of divergence from the shallow-water forms.
It is true that as we pass from the shallow waters, of those parts of the world where the great colonial madrepores build up the greater part of the vast coral reefs, into the deeper water beyond them, the solitary forms become relatively more abundant, but no new groups characterised by any special deep-sea attributes make their appearance. We must remember, not only that a great many solitary corals occur in shallow water in different parts of the world, but that some colonial forms, such as Lophohelia prolifera for example, are found only in very deep water.
Until quite recently it was usually stated in works dealing with the structures of coral reefs that the so-called reef-building corals, that is to say the large madrepores, astræids, and others, are confined to water not deeper than thirty fathoms. This limit must now be somewhat extended, in consequence of the discovery by Captain Moore of an abundance of growing coral at a depth of forty-four fathoms in the China seas; but, nevertheless, it is perfectly true that the corals do not grow in such profusion in very deep water as to form anything that can be compared with the reefs of the shores. It is quite possible that the advantages afforded by the light, warmth, and abundance of food of the shallow water may account for the luxuriance and vigour of the reef corals, and that where the food is scarce, and the water cold and dark as it is below fifty fathoms, the power of continuous gemmation is lost, and the rapidity of the growth and reproduction of the individual polyps is considerably diminished.
The fact remains, however, that, as with the sea anemones, so with the madrepores, nearly all the great divisions have a few isolated representatives in the abyss, and that no great family occurring in large numbers has yet been discovered peculiar to this zone.
The Alcyonaria, on the other hand, do present us with at least one example of a true deep-sea family. This great class of Anthozoa, distinguished from the Zoantharia by the presence of not more than eight tentacles and mesenteries and by the pinnate character of the former, falls into four principal divisions. The Stolonifera, the Alcyonidæ, the Gorgonidæ, and the Pennatulidæ. The first three of these divisions principally inhabit the shallow water. Each of them sends a few representatives into the great depths, but by far the greater number of the genera and species are to be found between tide-marks or in depths of less than fifty fathoms.
The Pennatulids, on the other hand, are rarely found in very shallow water, and nearly half the known genera live in deep water. At least two families may be said to be characteristically abysmal. These are the Umbellulidæ and the Protoptilidæ.
The Pennatulidæ are regarded by naturalists as the most complicated or highly organised group of the Alcyonaria. Three different forms of polype build up the colony or sea-pen as it is called. There is a single very much modified and enormously large polype, without tentacles, forming the axis, a large number of ordinary Alcyonarian polypes (autozooids) arranged in the form of leaves, or simply scattered irregularly on the surface of the central polype, and a number of very small undeveloped polypes (Siphonozoids) without tentacles, whose function seems to be to pump water into the canals of the colony, and thus to keep up the circulation of water.
