packets, attached as ovisacs or egg-strings to the genital openings, or enclosed in a dorsal marsupium, or deposited singly or occasionally in bundles. The youngest larvae are typical nauplii. The next, the copepodid or cyclopid, stage is characterized by a cylindrical segmented body, with fore- and hind-body distinct, and by having at most six cephalic limbs and two pairs of swimming feet.
The order thus defined (see Giesbrecht and Schmeil, Das Tierreich, 1898), with far over a thousand species (Hansen, 1900), embraces forms of extreme diversity, although, when species are known in all their phases and both sexes, they constantly tend to prove that there are no sharply dividing lines between the free-living, the semi-parasitic, and those which in adult life are wholly parasitic and then sometimes grotesquely unlike the normal standard. Giesbrecht and Hansen have shown that the mouth-organs consist of mandibles, first and second maxillae and maxillipeds; and Claus himself relinquished his long-maintained hypothesis that the last two pairs were the separated exopods and endopods of a single pair of appendages. Thorell’s classification (1859) of Gnathostoma, Poecilostoma, Siphonostoma, based on the mouth-organs, was long followed, though almost at the outset shown by Claus to depend on the erroneous supposition that the Poecilostoma were devoid of mandibles. Brady added a new section, Choniostomata, in 1894, and another, Leptostomata, in 1900, each for a single species. Canu in 1892 proposed two groups, Monoporodelphya and Diporodelphya, the copulatory openings of the female being paired in the latter, unpaired in the former. It may be questioned whether this distinction, however important in itself, would lead to a satisfactory grouping of families. In the same year Giesbrecht proposed his division of the order into Gymnoplea and Podoplea.
In appearance an ordinary Copepod is divided into fore- and hind-body, of its eleven segments the composite first being the head, the next five constituting the thorax, and the last five the abdomen. The coalescence of segments, though frequent, does not after a little experience materially confuse the counting. But there is this peculiarity, that the middle segment is sometimes continuous with the broader fore-body, sometimes with the narrower hind-body. In the former case the hind-body, consisting only of the abdomen, forms a pleon or tail-part devoid of feet, and the species so constructed are Gymnoplea, those of the naked or footless pleon. In the latter case the middle segment almost always carries with it to the hind-body a pair of rudimentary limbs, whence the term Podoplea, meaning species that have a pleon with feet. It may be objected that hereby the term pleon is used in two different senses, first applying to the abdomen alone and then to the abdomen plus the last thoracic segment. Even this verbal flaw would be obviated if Giesbrecht could prove his tentative hypothesis, that the Gymnoplea may have lost a pre-genital segment of the abdomen, and the Podoplea may have lost the last segment of the thorax. The classification is worked out as follows:—
1. Gymnoplea.—First segment of hind-body footless, bearing the orifices of the genital organs (in the male unsymmetrically placed); last foot of the fore-body in the male a copulatory organ; neither, or only one, of the first pair of antennae in the male geniculating; cephalic limbs abundantly articulated and provided with many plumose setae; heart generally present. Animals usually free-living, pelagic (Giesbrecht and Schmeil).
This group, with 65 genera and four or five hundred species, is divided by Giesbrecht into tribes: (a) Amphaskandria. In this tribe the males have both antennae of the first pair as sensory organs. There is but one family, the Calanidae, but this is a very large one, with 26 genera and more than 100 species. Among them is the cosmopolitan Calanus finmarchicus, the earliest described (by Bishop Gunner in 1770) of all the marine free-swimming Copepoda. Among them also is the peacock Calanid, Calocalanus pavo (Dana), with its highly ornamented antennae and gorgeous tail, the most beautiful species of the whole order (fig. 4). (b) Heterarthrandria. Here the males have one or the other of the first pair of antennae modified into a grasping organ for holding the female. There are four families, the Diaptomidae with 27 genera, the Pontellidae with 10, the Pseudocyclopidae and Candaciidae each with one genus. The first of these families is often called Centropagidae, but, as Sars has pointed out, Diaptomus (Westwood, 1836) is the oldest genus in it. Of 177 species valid in the family Giesbrecht and Schmeil assign 67 to Diaptomus. In regard to one of its species Dr Brady says: “In one instance, at least (Talkin Tarn, Cumberland) I have seen the net come up from a depth of 6 or 8 ft. below the surface with a dense mass consisting almost entirely of D. gracilis.” The length of this net-filling species is about a twentieth of an inch.
 |
| Fig. 4.—Calocalanus pavo (Dana). |
2. Podoplea.—The first segment of the hind-body almost always with rudimentary pair of feet; orifices of the genital organs (symmetrically placed in both sexes) in the following segment; neither the last foot of the fore-body nor the rudimentary feet just mentioned acting as a copulatory organ in the male; both or neither of the first pair of antennae in the male geniculating; cephalic limbs less abundantly articulated and with fewer plumose setae or none, but with hooks and clasping setae. Heart almost always wanting. Free-living (rarely pelagic) or parasitic (Giesbrecht and Schmeil). This group is also divided by Giesbrecht into two tribes, Ampharthrandria and Isokerandria. In 1892 he distinguished the former as those in which the first antennae of the male have both members modified for holding the female, and the genital openings of the female have a ventral position, sometimes in close proximity, sometimes strongly lateral; the latter as those in which the first antennae of the male are similar to those of the female, the function of holding her being transferred to the male maxillipeds, while the genital openings of the female are dorsal, though at times strongly lateral. In 1899, with a view to the many modifications exhibited by parasitic and semi-parasitic species, the definitions, stripped of a too hampering precision, took a different form: (a) Ampharthrandria. “Swimming Podoplea with geniculating first antennae in the male sex, and descendants of such; first antennae in female and male almost always differently articulated.” The families occupy fresh water as well as the sea. Naturally “descendants” which have lost the characteristic feature of the definition cannot be recognized without some further assistance than the definition supplies. Of the families comprised, the Mormonillidae consist only of Mormonilla (Giesbrecht), and are not mentioned by Giesbrecht in 1899 in the grouping of this section. The Thaumatoessidae include Thaumatoessa (Kröyer), established earlier than its synonym Thaumaleus (Kröyer), or than Monstrilla (Dana, 1849). The species are imperfectly known. The defect of mouth-organs probably does not apply to the period of youth, which some of them spend parasitically in the body-cavity of worms (Giard, 1896). To the Cyclopidae six genera are allotted by Giesbrecht in 1900. Cyclops (O. F. Müller, 1776), though greatly restricted since Müller’s time, still has several scores of species abundantly peopling inland waters of every kind and situation, without one that can be relied on as exclusively marine like the species of Oithona (Baird). The Misophriidae are now limited to Misophria (Boeck). The presence of a heart in this genus helps to make it a link between the Podoplea and Gymnoplea, though in various other respects it approaches the next family. The Harpacticidae owe their name to the genus Arpacticus (Milne-Edwards, 1840). Brady in 1880 assigns to this family 33 genera and 81 species. Canu (1892) distinguishes eight sub-families, Longipediinae, Peltidiinae, Tachidiinae, Amymoninae, Harpacticinae, Idyinae, Canthocamptinae (for which Canthocampinae should be read), and Nannopinae, adding Stenheliinae (Brady) without distinctive characters for it. The Ascidicolidae have variable characters, showing a gradual adaptation to parasitic life in Tunicates. Giesbrecht (1900) considers Canu quite right in grouping together in this single family those parasites of ascidians, simple and compound, which had been previously distributed among families with the more or less significant names Notodelphyidae, Doropygidae, Buproridae, Schizoproctidae, Kossmechtridae, Enterocolidae, Enteropsidae. Further, he includes in it his own Enterognathus comatulae, not from an ascidian, but from the intestine of the beautiful starfish Antedon rosaceus. The Asterocheridae, which have a good swimming capacity, except in the case of Cancerilla tubulata (Dalyell), lead a semi-parasitic life on echinoderms, sponges, &c., imbibing their food. Giesbrecht, displacing the older name Ascomyzontidae, assigns to this family 21 genera in five subfamilies, and suggests that the long-known but still puzzling Nicothoë from the gills of the lobster might be placed in an
