Opistholebes diodontis n.sp. is described from the intestine of the porcupine fish, Diodon hystrix, and development from the metacercaria to the adult is outlined. In that process, the ventral sucker shifts from near the mid-level to the posterior end of the body and other rearrangements occur as a result of differential growth. Hence in the Opistholebetidae and probably the Gyliauchenidae and Cephalo-poridae as well, the amphistome plan is considered to be secondary in phylogeny. None of those families belongs in the Paramphistomoidea. Instead, each is believed to be a distinct family of aberrant trematodes derived from a separate line of more generalized distomes. Representing one such line of descent are the Opistholebetidae and Opecoelidae which are more closely related to each other than either is to any other group. Hence a superfamily, Opecoeloidea, is proposed to include the two families and set them apart as a natural group distinct from the remainder of the allocreadioid complex which is discussed. The genus Pachycreadium is transferred to the Opistholebetidae, family diagnoses are revised, and a key to the species of Opistholebes is given.

This contribution is intended to be the last of this series. As one who may claim to have revived interest in these crustaceans from this region since the days of Bassett-Smith just prior to the beginning of the century, I find it particularly gratifying that several Indian zoologists are now steadily engaged in adding to our knowledge of these forms. Altogether some forty-three genera containing about 120 species have now been recorded from the Indian Ocean. This must form only a small fraction of the total number of forms that actually occur on the wide variety of hosts inhabiting this region.

The alimentary tract of female Culicoides nubeculosus is simple. There is much similarity between it and that of other nematoceran and brachyceran bloodsucking flies.

Among the interesting features observed are the following:

1. The mid-gut is composed of two portions, a tubular anterior segment and a dilated stomach, which show structural differences.

2. The peritrophic membrane, which is absent in the unfed female, forms around the ingested blood. Its substance is secreted by the epithelium of the stomach.

3. A simple rectal valve guards the opening of the small intestine into the rectum.

4. The Malpighian tubules are two in number, and they give attachment to muscular branches from different sources.

5. There are six accessory glands arranged in a rosette around the mouth of each salivary gland proper. Some of them may sometimes function as reservoirs for the secretion of the salivary glands proper.

In one specimen, tumours at the base of the stomach and abnormal features in the Malpighian tubules were observed.

The microsporidian described in this paper is parasitic in the fat-body of Gono-cephalum arenarium (Order Coleoptera; Family Tenebrionidae), a root-feeding beetle which is usually found sheltering under stones or fallen leaves and in lawns in South Africa. During the present investigation only adult insects were available for study as the earlier phases of the life history doubtless occur beneath the surface of the soil.

1. An account is given of the history, synonymy and host-range of Toxocara cati.

2. Eggs of T. cati were fed to various animals, and the second-stage larvae were found in the tissues of earthworms, cockroaches, chickens, mice, dogs, lambs and cats.

3. Cats were successfully infected by feeding eggs of T. cati and by feeding mice harbouring larvae in the tissues.

4. In egg-infected cats the larvae were found in the liver, lungs, muscles and tracheal washings as well as in the digestive tract, indicating that they migrate through the tissues.

5. In mice the larvae were found in the liver, lungs and muscles, but did not reach the alimentary tract. Changes occurred in the mouth region of the larvae in mouse tissues between 8 and 11 days after infection, but no actual moult was observed. No significant growth of the larvae was observed in mouse tissues.

6. In mouse-infected cats the larvae were mostly confined to the wall and contents of the digestive tract.

7. It appeared that the second moult occurred in the stomach wall of egg-infected cats at 3–10 days after infection; moulting larvae measuring 0·353–0·423 mm. in length; in a mouse-infected cat moulting larvae (length 0·459–0·765 mm.) were observed at 6 days after infection.

8. In both egg- and mouse-infected cats the third-stage larvae grew in the stomach wall and moulted for the third time at a length of 0·9–1·2 mm. This occurred at about 10 days in mouse-infected cats and at about 19 days in egg-infected cats.

9. The smallest fourth-stage larvae were found in the stomach contents whence they passed into the intestine. Probably they are at first attached to the wall and later become free in the intestinal contents.

10. The fourth-stage larvae have lips resembling the adult; sexual differentiation occurs during this stage, being first evident at 1·5 mm.

11. The fourth moult occurs in the intestine at a length of approximately 4·5–5·5 mm.

12. In the adult stage the lateral alae gradually disappear and the cervical alae reach the adult form at a length of approximately 45 mm. The smallest female observed containing eggs in the vagina measured 55 mm. in length. Eggs were first observed in the faeces at 56 days after infection by eggs.

13. The evidence from experimental and natural infection indicated that infection of cats with T. cati always takes place after birth.

14. No larvae were found in the alimentary tract of dogs following experimental infection either with eggs or with infected mice.

15. The relationship of the life history of T. cati to that of other ascaridid species, to the feeding habits of the definitive hosts, and in relation to the occurrence of human infection with this parasite is discussed.

1. Suspensions made from infected caecal material, derived from chickens which had been artificially infected with Histomonas meleagridis by the administration of embryonated Heterakis ova, can be used to transmit the organisms to other chickens when administered per os.

2. Infections with histomonad suspensions can only be obtained in chickens that have been starved, or in chickens that are feeding but have received an alkali mixture immediately before the suspension is administered. The highest incidence of infections is obtained by administering suspensions to starved chickens that have received 1 g. alkali mixture beforehand.

3. Successful infection appears to depend upon the pH of the gizzard and, possibly, of the upper intestine. The incidence of lesions produced by feeding suspension is increased as the contents of the gizzard become alkaline. It is also suggested that the consistency of the food may play a part, as coarse food, along with the infective material, is retained in the gizzard until it is completely ground up and mixed with the gastric juice.

4. In one series of passages the suspension became attenuated after fourteen passages through chickens. In other cases suspensions became attenuated after only six passages through chickens.

5. The courses of infections produced by histomonad suspensions and Heterakis ova are compared.

6. The advantages of the method of using histomonad suspensions are discussed in their relation to the screening of chemotherapeutical agents.

While human schistosomiasis is hardly a problem in India, schistosomiasis in domestic animals is a problem of considerable importance. Blood-flukes infect almost all the species of Indian domestic animals, and often seriously undermine their health and usefulness. Seven species of schistosomes are known to occur in Indian mammals, viz. Schistosoma indicum Montgomery (1906a), S. spindalis Montgomery (1906b), S. incognitum Chandler (1926), S. nasalis Rao (1933), Bivitellobilharzia nairi (Mudaliar & Ramanujachari, 1945) Dutt & Srivastava (1955), Orientobilharzia dattai (Dutt & Srivastava, 1952) Dutt & Srivastava (1955), and Orientobilharzia bomfordi (Montgomery, 1906b) Dutt & Srivastava (1955). The reported occurrence of Schistosoma bovis (Sonsino, 1876) Blanchard (1895) in India needs confirmation.

1. Of thirty-four Apodemus sylvaticus examined from in and near Exeter, thirteen were infested with varying numbers of the dicrocoeliid trematode, Corrigia vitta.

2. The worms live in the interlobary canals of the pancreas, sometimes singly, but up to eight have been found in a single mouse.

3. The dimensions of the more important organs are stated and compared with the measurements provided by other authors.

4. The taxonomic status of the species is briefly considered.

1. The egg-shell in many trematodes and pseudophyllidean cestodes is a quinone-tanned protein; and histochemical methods for polyphenol oxidase or for phenols may be used to demonstrate regions producing or containing shell material (i.e. the vitellaria, its ducts and the uterus) in whole mount preparations.

2. Specimens so treated may be further stained by routine whole mount methods. Gower's carmine is particularly recommended.

3. 70% alcohol was found to be the best general fixative for these methods.

4. These histochemical methods—with certain modifications—may also be used for demonstrating the egg-shell producing regions in sections.

5. Because of their affinity for basic proteins, bromo-phenol-blue and malachite green are suitable for staining these regions in sections. These stains are unsuitable for whole mounts.

6. There is some evidence to suggest that all trematodes and pseudophyllidean cestodes do not form and harden their egg-shell by quinone tanning; consequently, some forms may give negative or unsatisfactory results with these methods.

1. Cysts, formed in a strain of E. invadens growing with a mixed bacterial flora, were sterilized of all but one species of bacterium (organism x), by treatment with chemicals. Strains of E. invadens derived from these cysts were prepared, growing with organism x and either Bact. coli or B. megatherium.

2. A monobacterial strain of E. invadens, growing with Bact. coli alone, was obtained by maintaining a dibacterial strain, growing with organism x and Bact. coli, for 72 days in medium containing gentian violet.

3. A monobacterial strain of E. invadens, growing with Staph. albus, was prepared from cysts, produced in the monobacterial strain of E. invadens + Bact. coli, and sterilized by treatment with chemicals.

4. An axenic strain of E. invadens was prepared by maintaining the monobacterial strain of E. invadens + Staph. albus, in saline-liver medium containing antibiotics. This axenic strain has now been maintained for over 14 months in saline containing untreated, sterile liver, but no antibiotics.

The present revision of the morphology, taxonomy and affinities of trypanosomes causing Surra in the Old World was undertaken in order to determine the status of the diverse geographical and hostal strains, the exact relation of which to Trypanosoma evansi and allied species was hitherto a subject of controversy. The material for this study was represented by thirty strains from diverse mammalian hosts and various countries, extending from Morocco to the Philippines.

One of the problems dealt with is the occurrence of polymorphism in T. evansi. When single samples of typical (monomorphic) strains from naturally infected animals were examined, only negligible numbers of stumpy forms (0·05–0·40 %) were found in a minority of cases. But in strains kept under continuous observation for several years the proportion of stumpy forms occasionally rose higher (2–25%), while one strain exhibited a remarkable and persistent degree of polymorphism in the course of 17 years, the proportion of stumpy forms—including numerous posteronuclear ( = posterior-nuclear) ones—fluctuating from 0 to 60%. It was shown that—as in the case of the Brucei group—the length distributions in polymorphic samples of T. evansi could be statistically resolved into distinct curves, corresponding to the three morphological components (stumpy, intermediate, slender).

1. Liver lesions were easily induced in hamsters with two strains of E. histolytica but not with six others.

2. Serial liver passage increased the invasiveness to the rat's caecum of one strain which had become attenuated by in vitro cultivation.

3. The invasiveness of one of the liver passage substrains was lost a second time after cultivation in vitro and a second series of liver passages restored some degree of virulence.

4. Serial liver passage did not increase the invasiveness of two avirulent strains.

5. The ease of production of amoebic liver lesions in hamsters is not correlated with invasiveness in the rat's caecum.

1. The total bacterial floras from invasive strains of Entamoeba histolytica, when added to non-invasive strains, do not increase the invasiveness of the latter.

2. The bacterial flora of non-invasive strains do not inhibit the invasiveness of virulent strains.

3. Bacteria from a virulent substrain (revived by serial liver passage) do not increase the invasiveness of the attenuated parent strain.

4. It is concluded that the virulence of a strain of E. histolytica is primarily dependent upon the invasiveness of the amoeba itself.

1. The emergence of the gametocytes of Plasmodium gallinaceum from the erythrocytes and the exflagellation of the male gametocyte were inhibited completely when infected blood was exposed to an atmosphere of 5 % CO2 and 95 % O2, or 5 % CO2 and 95 % N2. When these gas mixtures were replaced by air, the gametocytes emerged from the erythrocytes and the male gametocytes exflagellated after approximately the same interval as they did in newly drawn blood exposed to air, or in blood ingested by mosquitoes. If pure CO2 were used instead of the gas mixtures containing 5% CO2, the development of the gametocytes, after the replacement of the gas by air, was uncertain.

2. The emergence of the gametocytes and exflagellation of the male gametocytes proceeded normally when parasitized blood was exposed to an atmosphere of N2.

3. The pH of freshly drawn normal chick blood was 7·38–7·46. The pH of blood exposed to air rose over a 20 min. period of observation, the highest pH recorded being 7·97. No significant difference in pH was observed between normal and heavily infected blood. It was not possible to correlate the stages of development of the gametocytes with precise values of pH of the blood.

4. The development of the gametocytes proceeded normally when parasitized blood was mixed with isotonic alkaline Tris buffer to give a pH of approximately 8·0; but it was inhibited, though not completely, when the pH was reduced below that of freshly drawn blood by the addition of isotonic acid buffer. The inhibitory effect of the acid buffer could be completely but not invariably reversed by alkaline buffer.

5. 5 % CO2 had no inhibitory action upon the gametocytes if the pH of the blood-buffer mixture was 7·62, but when it was 7·20 the inhibitory effect of the gas was complete.

6. The emergence of the gametocytes from the corpuscles and the exflagellation of the male gametocytes were completely inhibited when washed, parasitized corpuscles were suspended in isotonic buffer pH 7·78 or 8·20, but this inhibition was completely reversed when the corpuscles were resuspended in plasma. It is concluded that a factor present in plasma is necessary for the development of the gametocytes.

7. Further evidence that the pH of the blood, alone, is not the only factor affecting the emergence of the gametocytes and exflagellation of the male gametocytes was obtained from birds treated intravenously with Diamox. When blood from such birds was exposed to air the gametocytes emerged from the corpuscles and exflagellation took place, although the pH of the blood was as low as, or lower than, that of freshly drawn blood.

8. The emergence of the gametocytes from the erythrocytes and exflagellation of the male gametocytes were not stimulated by secretions from the stomachs or salivary glands of gorged or unfed mosquitoes.

9. The lowest measurement of pH of blood from mosquitoes' stomachs was 7·47 and the highest 7·90, the mean being 7·68, which is slightly higher than that of freshly drawn chick blood. No correlation could be found between the pH of the blood and the length of time since its ingestion by the mosquito.

The micro-glass electrode used in measuring the pH of blood from the stomachs of mosquitoes was a modification of that described by Caldwell (1954). The electrode was made of capillary tubing, drawn from Corning 015 glass, the outer diameter being 0·5–0·75 mm. and the length about 6 cm. As the electrode was used with a Pye Universal pH meter, and not with the electrometer impedance converter circuit used by Caldwell, it was important that the wall of the capillary tube should be very thin, in order that the resistance of the electrode should not exceed 500 MΩ, the maximum electrode resistance permissible with the Pye Universal pH meter.

The effect of combined doses of pyrimethamine and quinine has been studied in P. gallinaceum. No potentiation of the action of the drugs was observed.

Nine out of thirty-three specimens of the common shrew Sorex araneus, caught at Rosta in north Norway in August 1953, were found to be infected with Grahamella (Rickettsiales). Nine specimens of Clethrionomys rutilus, nine of Microtus oeconomus ratticeps and one of M. agrestis, caught at the same time and place, were found to be free of Grahamella. No other blood parasites were found in any of the animals examined.

Records by other authors of Grahamella in a variety of hosts are reviewed. The parasite is known to occur in mammalian hosts only: identifications of it in non-mammalian hosts can be discarded. Host specificity, tacitly assumed by most authors, is not proven, and therefore most of the specific names bestowed on Grahamella in different hosts are of doubtful validity. The morphology of the parasite varies little from host to host; and there is little evidence to show that its pathogenic effect on any host is serious. Grahamella has been cultured by a few authors: variations in characters of these cultures may be more reliable criteria for recognition of different species of Grahamella.

These small mammals were collected while I was a member of the University of Edinburgh Expedition to Arctic Norway, 1953. I wish to thank all who were concerned, in particular Dr D. M. Steven, with whom I did the trapping. The rest of the work was done during the tenure of a Research Studentship awarded by the Agricultural Research Council. I also wish to thank Dr P. Tate for his advice and his criticisms of the manuscript.

An account is given of some viruses affecting insects of economic importance. Twenty-six viruses are listed affecting twenty-two species of insects. Of these viruses eleven cause cytoplasmic, and eight cause nuclear, polyhedroses. Six are granulosis viruses and one virus has no intracellular inclusions. Most of these viruses are new or recently recorded. The granulosis viruses, affecting Pieris brassicae, P. rapae and Melanchra persicariae, and the diseases they cause, are described in some detail.

We are indebted to Dr L. L. J. Ossowski of the Wattle Research Institute, University of Natal, and to Mr T. H. Coaker of the Cotton Research Station, Namulonge, Uganda, who sent us, respectively, diseased specimens of the wattle bagworm and the American cotton bollworm, from which the viruses were isolated. Miss S. Vernon-Smith took the electron micrographs, and Mr Simon Frey the photographs on the optical microscope.