The invariable presence of a rickettsia in the gonads of the mosquito, Culex pipiens, is confirmed. This rickettsia is found in the cytoplasm of every germ cell in both sexes at all stages of development from larva to adult. No other species of mosquito is known to be similarly infected.

The morphology of the organism shows the following features. In addition to “typical rickettsiae”, i.e. minute rods and coccoids which multiply by binary fission, there is a variety of other forms which are interpreted as a series of stages, usually in the following sequence: small coccoids become enlarged, and instead of dividing by binary or multiple fission, undergo a condensation of their stainable substance into points or rods which ultimately appear as various combinations of sharply curved rods and coccoids retained as in a matrix within the body of the original or “parent” coccoid. As the matrix disappears, they become the irregularly curved forms and compact aggregations of “daughter” forms characteristic of this rickettsia.

The pathological germ cells previously noted in the testis by cytologists were found to occur in both sexes at all stages, and to be apparently the result of infection with this rickettsia.

The problem of classifying the rickettsiae is discussed. The similarity of the rings and curved forms of various rickettsiae to corresponding forms, as yet little studied, in other groups of organisms is pointed out.

The name Wolbachia pipientis (gen. et sp.n.) is proposed for the rickettsia of Culex pipiens.

Large, homogeneous or granular cell inclusions with an affinity for neutral red, termed NR bodies, were found in the gonad wall of nearly every specimen of Culex pipiens, but were not found in C. territans. They resemble the inclusion bodies of certain viruses and at times closely simulate masses of rickettsiae, but are apparently distinct from the species here described. Their nature is unknown but their appearance and behaviour are consistent with their being living agents or the manifestations thereof. A degenerative process, in which NR bodies replace primary ovarian follicles in whole or in part, is described.

Inoculation and cultivation experiments resulted negatively for both rickettsiae and NR bodies.

1. Cotylaspis sinensis and Lophotaspis orientalis are described as new species of trematodes from the small intestine of the turtle, Amyda tuberculata, from Foochow, Fukien Province, China.

2. Stichocotyle cristata is described as a new species from the spiral valve of the cow-nosed ray, Rhinoptera quadriloba, from Biloxi Bay, Mississippi.

3. A new subgenus, Multicalyx, is created for the species Stichocotyle cristata, while the species S. nephropis appropriately belongs to a new subgenus Stichocotyle.

4. Evidence is provided that aspidogastrid worms are referable neither to the Monogenea nor the Digenea, but belong to an intermediate subclass group, Aspidogastrea, n.n.

5. The genus Stichocotyle has been removed from the family Aspidogastridae Poche, 1907, and placed in a new family, Stichocotylidae.

The earliest stages found are groups of small bodies, 1·8 or 2·0μ in diameter. These groups are probably taken up by phagocytic cells and pass through the epithelium of the host into the connective tissue, where a cyst is formed by the host. Within the cyst the bodies grow. In many cysts there is a differentiation into two types: the larger, when mature, attain 40μ in size, are shaped like a pear, and contain at the broad end a chromatic rod. The small bodies reach 3Oμ in length, are also somewhat piriform, and contain two similar structures attached to their nuclei. In some cysts large bodies alone are found.

Development is periodic, the youngest stages being found from January to March and the parasites not reaching maturity until September–November.

Type species P. sociabile, with the features mentioned above: found near the ventral surface of the foot of Buccinum undatum.

1. The life history of Encyrtus (Comys) infelix (Embleton), a hymenopterous parasite belonging to the chalcid family Encyrtidae, is described. It parasitises Saissetia hemisphaerica, a member of the subfamily Lecaniinae (Homoptera Coccidae). Its respiratory relationships with the host are of a quite extraordinary character.

2. The egg is provided with a hollow stalk which is left protruding from the posterior dorsal body wall of the host. The first three larval instars are metapneustic, the spiracles being placed at the tip of a pair of long caudal processes which are inserted in the hollow stalk. It has been proved that these spiracles are open and in actual communication with the atmospheric air, which enters through the pedicel.

3. The fourth and fifth larval instars are amphipneustic and the manner of respiration is entirely changed. The caudal processes degenerate and finally break away. The fourth instar larva turns round in the scale insect and becomes invested with a closely fitting transparent membraneous sheath produced by the phagocytes and fine tracheal branches of the host. This process appears to be merely a special case of the type of phagocytic activity which normally gives rise to the basement membranes and “connective tissue” membranes; such membranes are particularly conspicuous and well developed in Saissetia. The sheath becomes attached in an extraordinary manner to the main lateral tracheal trunks of the host in four (or six) places in the neighbourhood of the larval spiracles. This process is described in detail. It has been proved by an injection technique that an actual connection is established between the lumen of the host trachea and the cavity of the sheath. Bubbles of gas appear inside the sheath close to the points of attachment and near the spiracles of the larva which are thus put into functional communication with the tracheal system of the host.

4. While the production of the sheath might under certain circumstances be of value to the host the development of tracheal attachments appears to be of value to the parasite alone. The conclusion that the whole structure is an adaptation for the respiration of the parasite seems inescapable. A theory is tentatively put forward to account for the stimulation of the host tracheal epithelium by a sudden physiological change in the tension of the respiratory gases in the region of the parasite spiracles.

5. References to similar instances among the Chalcids are briefly reviewed. It is suggested that the “puparia” which have been described in the case of one or two other Chalcidoidea may be found to arise in the same way.

(i) Amoebae closely resembling Entamoeba coli (of Man) have been studied in macaques (M. rhesus, M. sinicus, M. nemestrinus) and in a marmoset (Hapale jacchus).

(ii) Pure strains of these amoebae have been isolated and cultivated, and carefully compared with similar pure strains of E. coli from Man (E. coli hominis).

(iii) The complete life-cycles of these various strains—amoebae, cysts, and all stages of precystic and metacystic development—have been studied in vitro.

(iv) The “E. coli” of macaques (E. c. macacorum) was experimentally transmitted to a man, who ingested pure culture-cysts of a strain derived from M. rhesus.

(v) The “E. coli” of the marmoset (E. c. jacchi) was likewise transferred to M. sinicus, and from this host again to M. rhesus and Man.

(vi) The natural infections observed in M. sinicus, M. rhesus, and H. jacchus all died out spontaneously after persisting for various periods (4½ years, 2 years, and 1 year respectively). The experimental infection with E. coli macacorum induced in Man also disappeared spontaneously (after about 6 weeks).

(vii) Attempts to infect kittens experimentally with E. coli jacchi were uniformly unsuccessful.

(viii) All infections with “E. coli” from every source—whether natural or experimental—have proved to be completely innocuous.

(ix) Culture-cysts of various strains of “E. coli” have been found able to live at low temperatures for all times up to about 4½ months (maximum 135 days at 1°–2° C., for a strain from H. jacchus).

(x) The “E. coli” of H. jacchus—and all its experimental derivatives—ingests human red blood-corpuscles in vitro as readily as E. histolytica.

(xi) Several strains of E. coli hominis, isolated directly from naturally infected human beings, have been found to possess the same property.

(xii) Since strains of E. coli thus exist—both in men and in monkeys—which eat human red corpuscles avidly, this faculty can no longer be considered as peculiarly characteristic of E. histolytica; and the importance of this observation, for the diagnosis of amoebic dysentery, is therefore emphasized. The statement that E. histolytica ingests red corpuscles, while E. coli does not, is untrue. It is not possible to distinguish these species by this character alone.

(xiii) As it has been found impossible to distinguish various strains of “E. coli” living in macaques, a marmoset, and men, by any recognizable specific feature—all of them being so closely alike morphologically, physiologically, culturally, and in their ability to inhabit these different hosts—it is concluded that no specific difference actually exists between any of them. The “E. coli” of the Macaques and of the Common Marmoset is probably Entamoeba coli itself—the species which naturally inhabits Man.