The ovaries are two and simple (Acoela, Rhabdocoele Macrostomidae); they are numerous and simple in Polycladida. In some Rhabdocoela, e. g. Prorhynchus, the two glands, though simple externally, are functionally divisible each into a germarium and vitellarium, and this division becomes marked externally, and the two parts have their separate ducts in other Rhabdocoela; whilst in Tricladida there is on each side of the body a single anterior ovary connected to a long oviduct, which is beset on its outer side by follicular or branching vitellaria2. Many Acoela and Alloiocoela have no specialised ducts, but the ova like the sperm pass into spaces in the parenchyma. A network of ciliated tubules connects the ovaries of Polycladida. They unite into a canal on each side of the body. Each canal is dilated into a uterus, to which accessory glands are attached. The two uteri unite into a common canal, which is dilated and receives a number of shell-glands. Its termination is often muscular, forming a female copulatory organ.

The male and female ducts open into a genital atrium in all Tricladida and most Rhabdocoelida. To this atrium is commonly appended a uterus, and in some Rhabdocoela a bursa copulatrix and receptaculum seminis3. Or in other Rhabdocoelida, and in Polycladida with the exception of two genera, the two ducts open separately, the male usually in front of the female, and both behind the pharynx. In Prorhynchus (Rhabdocoele) and Stylostomum (Polyclad) the penis opens into the mouth; and in many Rhabdocoela the penis with its chitinous hooks is employed to catch and retain the prey. Macrorhynchus helgolandicus has a poison dart in connection with the atrium genitale.

1 The structures called by Geddes Pulsatella Convolutae, and found in the Acoelan Convoluta Schultzii, are perhaps not parasitic organisms, but indications of an excretory system. They occur in very young specimens indeed. See Delage, A. Z. Expt. (2), iv. p. 150. According to Francotte, the cross anastomosis of the lateral canals in front of the mouth has in Derostomum Betie'deni an external opening. As to Gunda, Lang states that the dorsal and lateral vessels are connected in the septa of the body in which they are convoluted; the efferent branches and pores are similarly situated. He also observed in the same animal that some of the flame-cells were connected by basal processes to the cells of the intestinal epithelium, and that others were actually situated among them.

1 Iijima denies this connection between the testes and the vasa deferentia in the fresh-water Tricladida which he examined. He thinks that the sperm is set free into the spaces of the parenchyma, and is taken up thence by lateral openings in the walls of the vasa deferentia. But as the latter do not extend far forwards, the sperm of the anteriorly placed testes never reaches them at all. See Z. W. Z. xl. pp. 405-408. When ducts are absent, as in Acoela, the sperm traverses the parenchyma.

2 Iijima states that the vitellaria do not communicate directly with the oviducts in the fresh-water Tricladida which he examined. The ducts have openings into the spaces of the parenchyma, and in these spaces the vitelline cells accumulate just as do the spermatozoa. The openings in question are closed each by a peculiar large cell before the vitellaria ripen (Z. W. Z. xl. pp. 415-16).

3 The uterus of the fresh-water Tricladida does not serve as a store-house of ova as do the uteri of the Polycladida. It is single, and is probably a gland secreting the material for the cocoon. Albumen glands usually open into its duct ( = so-called vagina) or into the oviducts. A structure called ' muscular glandular organ' opens into the atrium genitale, but it appears to be variably present. Its walls are stout and muscular, and Iijima believes that it receives the ducts of various glands. See Z. W. Z. xl. pp. 419-26.

The form of the spermatozoon is very variable, especially in Rhabdo-coelida. There is a reciprocal copulation, but self-fertilisation occurs in the case of the summer ova of many Rhabdocoela. The ovum is contained in a shell, or a number of ova within a common shell or cocoon, as in Tricladida, some Rhabdocoelida, and some Polycladida. The last-named usually lay their ova in rows united by a cement. In some R/iab-docoela (certain fresh-water Mesostomidae) there are thin-shelled summer and thick-shelled winter ova. The egg shell, or cocoon, is formed either by the atrium genitale, by the uterus, or by special glands. In those forms where there are vitellaria, it contains vitelline-cells in addition to the germ-cell. The vitelline-cells retain their integrity, and are devoured by the young Turbellarian at a certain stage of growth. Segmentation is total but unequal, and there is a gastrula formed either by invagination or overgrowth. Development takes place within the parent in the case of summer ova (supra). There is a metamorphosis in many Polycladida, the larva having a praeoral ciliated ring, which is produced first into four ( = Gotte's larva) then into eight ciliated processes (=Muller's larva), subsequently absorbed.

A few viviparous species are known 1.

Asexual reproduction by posterior gemmation occurs in the Rhabdo-coele family Microstomidae; chains of sixteen individuals may be thus formed in Microstoma lineare. Transverse fission has been observed in the Triclads Planaria subtentaculata, Planaria (sp. ?), and Polycoelis cornuta.

The Turbellaria are carnivorous. They inhabit fresh brackish and salt water, and damp earth. A few Rhabdocoeles are parasitic, the genera Graffilla and Anoplodium entirely so, the former in marine Mollusca, the latter in or on Holothurians; so too the Triclad Bdelloura on Limulus.

Two curious Polyclads have recently been described with certain Ctenophore-like characters. These are the Coeloplana of Kowalewsky (Z. A. iii. 1880, p. 140) and the Ctenoplana of Korotneff (Z. W. Z. xliii. 1886). They resemble Polyclads in the flattened aspect of the body, the general ciliation of the ectoderm, the branched digestive system with ventral mouth. Ctenoplana has an outer longitudinal layer of muscles, an inner transverse, as well as dorso-ventral bundles. Both resemble Ctenophora in the presence of two solid retractile tentacles, one on each side, at right angles to the plane of the stomach. There is an aborally placed otolith-mass. In Ctenoplana this lies, supported by the ends of stiff hairs, in an open depression, at the sides of which the nervous system is situated (?), and there are eight short dorsal rows of ctenophoral plates, arranged as in Ctenophora, and lodged in grooves which can be depressed by special muscles. Two branches arise from the stomach in Coeloplana, and pass aborally but end blindly. A right and left opening near the plane of the tentacles is found in the same animal, but what they lead to is uncertain.