Aquatic Arthropoda with cutaneous or branchiate respiration: with two pairs of antennae, a limb-bearing thorax, either free or united more or less to the head, and as a rule a segmented abdomen which may or may not carry limbs.

The variety of external forms in this class is very great. The head appears to consist of five fused somites, denoted by the presence of two pairs of antennae, a pair of mandibles, and two pairs of maxillae. The number of somites in the thorax and abdomen of Entomostraca is extremely variable. In the Malacostraca the former has eight, the latter six somites with a telson, except in Nebalia (Leptostracd), where there are eight. And of the eight thoracic somites, one (Thoracostracd), three or four (Cumacea), five (Stomatopoda), or all (Decapoda) may be united to the head to form a cephalothorax. The somites of the abdomen may fuse as in some Isopoda; or the abdomen may be rudimentary, e. g. Cirripedia and Laemodipoda (Amphipoda). In the Ostracoda the body is not segmented at any time; and in some fixed or parasitic forms segmentation may become obscured or lost, though in some instances shown by the presence of limbs.

Each somite is typically ring-like or cylindriform as in Copepoda, but it may be flattened dorso-ventrally or laterally. In many cases certain somites, or all of them as in most Thoracostraca, develope a lateral pleural process, extending outwards above the articulation of the limbs. A process or fold formed by the edges of the last cephalic somite constitutes the shield of Apus, the mantle (with or without calcareous pieces) of the Cirripedia, and the bivalve shell of Nebalia. United pleural processes of the fused somites of the thorax form the branchiostegite of Podophthal-mata and the bivalve shell of the Estheridae (Branchiopoda), Cladocera, and Ostracoda. In the last-named group and in Nebalia a special transverse muscle closes the shell upon the body.

The first antenna is primitively uniramose, and retains this character in Phyllopoda, Ostracoda, and Copepoda; but during growth it generally becomes two or three branched. It is minute in Cladocera and terrestrial Isopoda. In the Cirripedia its second joint carries a disc in connection with which is the aperture of a cement gland. The animal is attached by its means to some foreign object, whilst the head dilates into a broad base or elongates into a peduncle. The other appendages of the head and body are primitively biramose, except in the Ostracoda and the appendages of the thorax in Arthrostraca. The second antenna may become uniramose, or the outer branch may be reduced to a scale or squame (many Thora-costracd). It is minute in Apus and is lost in all Cirripedia and Hyperidae (Amphipodd). The mandible is reduced to the basal masticatory portion in Phyllopoda and Cumacea. In other groups the reduction may be temporary, and there are developed subsequently a number of small joints which form the palp. In many parasitic suctorial Crustacea the mandibles form a pair of stylets, inclosed in a more or less complete sheath constituted by the so-called upper and lower lips. The maxillae vary much in character, but are generally more or less expanded or foliaceous.

The outer and inner branch (=exo- and endo-podite) of the second maxilla are separated in Copepoda so as to appear like independent limbs, whilst in Cladocera the appendage is aborted. In the Copepodan families Calanidae and Pontellidae both first and second maxillae closely resemble the biramose Naapliiis appendage. The first maxillae, like the mandibles, are stylets in Argnlns, whilst the second are converted into large sucking discs. Both pairs of maxillae want the exopodite in Isopoda.

The thorax is always limb-bearing; the abdomen bears none in Entomostraca, but does so with few exceptions in Malcostraca. The primitive type of limb is probably that of the Copepoda, which closely resembles the Nauplius appendage. It has a basal stem carrying a more or less jointed or lamellate exo- and endo-podite. Such a limb is seen in the thoracic appendages of Cirripedia and of the Schizopoda among Mala-costraca, and is generally found in the abdominal region. The Phyllopod type of appendage, seen also but in a more primitive state in Nebalia, with its branchia and external respiratory plate, and its series of internal lobes or endites, is probably an adaptation of the Copepodan limb. The exopodite is lost in the ambulatory thoracic limbs of Arthrostraca and Decapoda. Of the eight pairs of thoracic limbs in Malacostraca the first pair in Arthrostraca, the three first in Decapoda, and the five first in Stomatopoda are modified into foot jaws or maxillipeds. Apus (Branchiopodd) is remarkable for having more than one pair of limbs on the posterior thoracic somites. The abdominal limbs, when present, with the exception of the last pair, never attain the same locomotor importance as do the thoracic.

The last somite of the abdomen in Entomostraca carries a pair of furcae anales, variable in character, between which the anus opens. In Malacostraca the furcae and the somite which bears them are modified into an azygos plate or telson with the anus placed on its ventral surface.

The chitinous cuticle is thin and delicate, or thick, laminate, and calcified. Glands are found in various positions opening on the integument, the limbs, upper lip, round the mouth, where they may have an amylolytic function (many Isopodd); or on the under surface of the abdomen (Becapodd), or the lamellae of the brood-pouch (some Isopodd), when the secretion probably serves to attach the ova. In some Cladocera (e. g. Sida) there is a nuchal gland, by means of which the animal can attach itself to some foreign object. A similar gland is present in the young Branchiopod, but is more or less aborted subsequently, and it exists as the dorsal organ in the embryo Arthrostracan.

The nervous system resembles that of other Arthropoda in its general features. The commissural cords connecting the supra- and infra-oeso-phageal ganglia are sometimes of great length. The first-mentioned ganglion supplies the eyes, the first antennae, and, except in certain Phyllopoda, the second antennae as well. In the order named the second antennae are innervated (? in all) by a pair of ganglia situated on the oesophageal commissures, and the right and left halves of the post-oral ganglia retain their independence and are united by transverse commissures. In other orders the halves are as a rule united closely; the ganglia supplying the oral appendages are fused into an infra-oesophageal ganglion, and concentration may proceed to such an extent that all the post-oral ganglia become united into a single mass (Decapoda Brachytird)1. A stomatogastric system originates in Apus from the first post-oral pair of ganglia; in the Decapoda from ganglia on the oesophageal commissures which are probably homologous with the first post-oral ganglia of Apus, and additional factors may be derived in the group last mentioned from the supra-oesophageal ganglion2.

1 According to Claus (Organismus der Phronimiden, Arb. Zool. Inst. "Wien, ii. p. 52) the nerves of the second antennae originate from the oesophageal commissures in the Fhronimidae, but whether the corresponding nerve-centres are supra- or infra-oesophageal in position is a point he did not determine. In some Isopoda, the right and left halves of the thoracic ganglia, and of the abdominal also when distinct, are .more or less separate, and connected to one another by transverse commissures (Huet, Journal de l'Anat. et Physiol, xix. p. 305).

2 A nerve with ganglion cells intercalated from place to place has been observed by Claus in Phronimidae running on the dorsal aspect of the heart (op. cit. supra, p. 40). So too in Stomatopoda, in which a ganglion cell corresponds to each pair of ostia (Claus, 'Circulation of Stomatopoda? op. cit. v. 1884, P IT). In the same order a ganglion lies just above the epistoma (loc. cit. p. 12). Huet has recently investigated the sympathetic system of Isopoda. He detected an azygos nerve running from one to the next succeeding ganglion of the ventral cord, as described first by Rathke, as Well as two recurrent nerves, derived one from each of the two posterior terminal nerves, and supplying the walls of the digestive tract. But he did not find the azygos 'ganglion frontale' of Leydig, placed in front of the supra-oesophageal ganglion to which it is connected; nor yet the dorsal intestinal nerve of the same observer, or of Lereboullet (see pp. 306-10, op. cit. in note 1).