1. The adult stage of Polystoma integerrimum feeds on blood which is digested by a combination of intraluminar and intracellular processes, haemolysis and partial digestion in the gut lumen being followed by absorption and completion of digestion by the cells of the gastrodermis.

2. The end products of intracellular digestion are the pigment haematin and a crystalline substance identified as haematoidin, which is chemically identical with vertebrate bilirubin, and these are passed back into the lumen to be eventually voided to the exterior.

3. It is probable that the expulsion of haematin from the gut cells results in a simultaneous release of enzymes, traces of which persist in the lumen to effect the early extracellular stages in the digestion of the next meal, so that intraluminar digestion in Polystoma appears to be an incidental extension of the primitive intracellular process arising from the blood-feeding habit and the consequent need to eliminate residues of haemoglobin breakdown from the gastrodermis.

4. Since haematin formation has been described in certain other trematodes it is suggested that their digestive processes probably follow a course similar to that described for Polystoma but an alternative method must exist within the class since other blood feeders such as Haplometra and Gorgodera do not form this pigment.

Rana temporaria tadpoles which had been naturally infected by larvae of Polystoma integerrimum were kept in the laboratory under conditions precluding the continuous re-infection which occurs in nature. One tadpole was seen to have two larvae attached to the skin of the ventral surface of the body, which were moving towards the posterior end. The tadpole measured 7·9 mm. in body length and 20·8 mm. in total length, and its stage of development corresponded to stage 5 in the table of development and metamorphosis compiled by Rugh (1951); it was collected in mid-May and had been kept for twelve days under laboratory conditions. The tadpole was fixed and dissected: the bladder had not developed, and the gills showed no indication of atrophy; the alimentary canal was long and heavily coiled, and filled with food. In addition to the two larvae observed on the skin, one larva was found in the branchial chamber, two were in the cloacal tube, seven occupied the cloacal chamber and two had travelled up the right ureter to the kidney; the alimentary canal was free of larvae. All fourteen larvae were gyrodactyloids.

Six specimens of Xenopus laevis obtained from University College, London, were examined for polystomes; one harboured an adult worm measuring 2·1 mm. in length and two immature specimens, both 0·8 mm. in length, and one was parasitized by a single mature specimen measuring 2·4 mm. in length.

Cercaria owreae (Hutton, 1954) is renamed and redescribed. It has two posterior appendages and represents a new type of distome cercaria, named “Diplocercous”. The caeca extend into the appendages, one into each, and a rosette-like group of six diverticula occurs at the anterior end of each of them. Certain peculiarities of the caecal epithelium are described and discussed briefly, but because of the immature state little information can be given about other organs. Fifteen larvae were obtained from 12 out of 181 Sagitta hexaptera collected in four Caribbean localities. Two larvae were found in the gut, the remainder in the coelom. The trematode probably belongs to the Accacoeliidae and is referred tentatively to the genus Accacladocoelium Odhner, 1928, adults of which are parasites of the sunfish Mola mola (L.).

During the course of studies on human filariasis in Fiji in the period 1954–50, two new microfilariae were found in fruit bats. The first was found in the blood of 38 bats out of 111 examined from various islands of the territory. It has recently been described and named Microfilaria fijiensis by Yeh, Symes and Mataika (1958).

Page 221 Legend to figure should read “Fig”5.

Page 223 Legend to figure should read “Fig. 3”.

Page 272 for “Morgascarida” read “Morgascaridia”.

1. At least twenty-two species of filarial worms are known to develop in mosquitoes. Filarial larvae of non-human origin must often be present in mosquitoes dissected during filariasis surveys yet the larvae seen are commonly assumed to be of human origin.

2. On Pate island off the Kenya coast where bancroftian filariasis is prevalent, several distinct species of infective larvae were found in mosquitoes both in the houses and in the bush. Seven species of filarial worms were found in the local animals and a collection of known infective larvae was made by feeding mosquitoes on these different infections. A study of the material in this “reference collection” showed that many of the larvae in “wild” mosquitoes could be identified with confidence.

3. A technique has been developed for the dissection of preserved and stained mosquitoes; this method has many advantages and can be used in routine filariasis surveys.

4. The characters which were most useful in differentiating the infective larvae were: the length; the position of the anus; and the shape of the caudal extremity. Eight species of infective larvae were available for study, these are described and illustrated: W. bancrofti, W. malayi, W. pahangi, W. patei, D. corynodes, D. repens, D. immitis and S. equina. The literature dealing with a further fourteen species is reviewed.

5. A simple key has been produced to help identify the species of infective larvae found in mosquitoes.

6. On the Kenya coast W. bancrofti infective larvae have been found only in C. fatigans, A. gambiae and A. funestus. Infective larvae of animal origin which might easily be mistaken for those of W. bancrofti have been found in Ae. pembaensis, Ae. aegypti, M. uniformis and M. africanus.

7. Formulae for transmission indices based on infection rates in insect vectors are of little value unless there is confidence in specific identification.

8. A study of the infective larvae in Simulium neavei from an onchocerciasis area in Uganda has shown that not all the larvae are of human origin. A closer examination of the morphology of infective larvae in the vectors of filarial parasites of man in other areas will almost certainly show that the interpretation of infection rates may be complicated by the presence of infections of nonhuman origin.

9. Feeding experiments were carried out with Aedes pembaensis and Ac. aegypti on a donkey infected with Setaria equina; both of these mosquitoes proved to be suitable vectors of this filarial worm.