CORALS 377 Alcyonarian and not Zoantharian. With Hcliopora mus unquestionably be placed the extinct genera Heliolites Plasmopora, Lyellia, and Polytremacis, all of which hav an essentially tubular and tabulate coenenchyma, th corallites having tabulae and more or less distinct pseudo septa. These, therefore, must also be removed to the Alcyonaria. The genus Podllopora, as shown by Verrill is a true Zoantharian, belonging to the Aporosa, aric apparently referable to the Oculinidce. Seriatopora, though still imperfectly known, may with great probability be placed in the immediate neighbourhood of Podllopora , and Rhabdopora and Trachypora may be provisionally grouped along with it. The genus Axopora, with its great columella, its reticulate sclerenchyma, and its want oi septa, is in an extremely uncertain position, but may possibly be Alcyonarian. There remain four groups of the Tabulata, which are in themselvec- well marked, but which are still in a doubtful position. The first cf these is the group of the Favositidce (comprising Favositcs, Em- Tunisia, MicJiclinia, Koninckia, Strialopora, Alveolites, &c), in which the septa are rudimentary or absent, the tabulaj are extremely well developed, and the walls are perforated more or less freely with mural pores. Their septa (when present at all) may be only pseudo pepta, like those of the genus Heliopora, and these forms may per haps be Alcyonaria. In their perforated walls, however, they closely resemble the recent genera Alveopora and Favosiloporci , and their true affinities would thus seem to be rather with the Uadreporidce in the Zoantharia pcrforata. This view is further supported by the occurrence of genera like Columnopora, in which the perforated walls and tabulate corallites are associated with com paratively well-developed septa. Another group is that of the Chcetetidae (comprising Chcctctes, Monticulipora, Dania, Conslettaria, &c., and probably Fistulipora, Callopora, and some allied forms), in which the walls of the corallites are imperf orate, the tabulaj arc well developed, and there are no traces of septa. This group is in a most uncertain position, it being an open question whether it should be referred to the Alcyonaria, thePolyzoa, or the Hydrozoa, or whether it can be retained in the Zoantharia. Though often associated with the Favositidce, it seems certain that there is little true relationship between the two groups; and the most probable view is perhaps the one which refers the Chcetetidae to the Alcyonaria. The genus Labechia has not been examined in sufficient detail to render its reference to this group at all certain. A third group is that of the Thccidce, comprising the single genus Thccia, in which the corallites have no true wall, but the coste are greatly developed, find are so filled up as to constitute a dense ccenenchyma. The affinities of this group are quite uncertain, though it may be regarded with some probability as belonging to the Alcyonaria. Lastly, there is a fourth group, comprising the genera Halysites, Syringopora, Chonostegites, and Thccostccjitcs, with probably Flctcheria snid Beaumontia, in which there are rudimentary septa (typically twelve in number), well developed tabulae, and impcrforate walls. J n Syringopora, though the walls are compact, the visceral chambers of contiguous corallites are placed in communication by means of tubular connecting processes, and there is thus a decided approach to the structure of the Favositidce on the one hand, whilst, on the other hand, the genus in some respects is nearly allied to Aulopora. The other genera of this group have no communication between their corallites. If any forms can be retained to constitute a "tabu late order of the Zoantharia, it would seem to be these ; but it is not possible at present to come to any decided conclusion on this point. (4). The TUBULOSA constitute a small group, including princi pally the genera Aulopora and Pyrgia, to which perhaps Clado- clionus^ and Stomatopora should be added. The corallum in this group is simple (Pyrgia) or compound (Aulopora), the corallites in the latter case being united by a creeping and branched coenenchyma. The theca? are trumpet-shaped, tubular or pyriform, without tabulae, and having the septa indicated by mere strife on the inner surface of the wall. The affinities of this group are very doubtful, and it seems questionable if it can be retained in the Zoantharia. In some respects the forms included under this head show points of relationship with the Cyclostomatous Polyzoa, such as Alccto ; but it seems more probable that they are really Actino- ioa, and perhaps referable to the Alcyonaria. As regards their distribution in space, the Zoantharia scleroder- mata are, like the other members of the Actinozoa, exclusively marine. They are very widely distributed over the globe, occurrino- in all seas except the coldest, but attaining their maximum develop ment in warm regions. They may conveirsntly be regarded under two heads the one comprising the "solitary corals, whether simple or compound, the other comprising the "reef-building corals. The former do not constitute the great aggregations of coral which are known as "reefs," and though some of the com- A, Portion of Hie corallum of Aulopora tuTxzformis, Goldfuss, of the natural size. B, Two corallites of the same, enlarged, showing the septa! stria;. C, Pyrgia Jfic/trlint, Edw. and H., of the natural size. D, The same, slightly enlarged, showing the interior of the calice. pound forms may, as individuals, attain a large size, they are desti tute of the loose cellular ccenenchyma of so many of the reef- builders, and thus do not tend to increase indefinitely in dimensions. The solitary corals, further, are essentially and principally deep-sea forms, only rarely found in the littoral zone, or at extreme low water, abounding most in depths of from 10 to 200 fathoms, and extending their range even to depths of from 500 to over 1500 fathoms. They may, as already remarked, be either simple or compound, and amongst the more important recent genera which are solely, or chieily, deep-water forms, may be mentioned Caryophyllia (rarely found in the littoral zone), Balanopliyllia, Flabcllum, Sphcnotrochus, Paracyathus, Desmophyllum, Lobopsammia, Thecopsammia, Dendro- phyllia, Allopora, Uculina, Lophohelia, and Amphihelia. Coral Reefs. The " reef -building " corals are essentially shallow-water forms, flourishing between extreme low-water mark and depths of from 20 to 25 fathoms. By their continued growth and aggregation they give rise to the great banks of coral which are known as " coral-reefs," the compound species often possessing a lax cellular coenenchyma, enabling them to increase almost indefinitely in size. In other cases they increase rapidly by spontaneous fission. Amongst the more important reeif-building corals of the present day may be mentioned the Astraddce, generally (Aslrcea, Meandrina, Diploria, Astrangia, Cladocora, &c.), the Madreporidce (Madrepora, &c.), the Poritidce (Porites, Goniopora, Montipora, &c.), many of the Oculinidce (Orbicella, titylaster, Pocillopora, &c.), the majority of the Funyidae (two species of Fungia inhabit deep water), and the Millepores. Though principally formed by corals belonging to the Zoantharia sclerodermata, the growth of coral-reefs is further con tributed to by various Alcyonoid corals (Heliopora, Tubipora, and numerous forms belonging to the Gorgonidoe), and by the calcareous algae (Nullipores and Corallines). The distribution of the reef-building corals seems to depend mainly upon the mean winter temperature of the sea, and they are confined to seas in which the temperature of the water during the winter does not sink on an average )elow G6 or 68 Fahr. The seas thus lircitcd may be said to be comprised within a distance of about 1800 miles on either side of the equator. Even within these limits, lowever, apparently owing to the influence of Arctic urrents, no coral-reefs are found on the western coasts of Africa and South America. The metropolis of the reef- juilding corals may be said to be the central Pacific Ocean, ivith its numerous islands and masses of continental land ;
VI. 43