The sudden death on 27 February 1963 of David Keilin deprived the scientific world of a biologist the span of whose scientific activities is unlikely to be equalled, much less excelled, in the future. These activities extended from descriptive morphology of protists, fungi and insects to the biochemistry of respiratory enzymes and metallo-protein compounds. It is, consequently, not possible for one person to deal adequately with all aspects of his work and in this appreciation of the man and his scientific achievements the main emphasis will be on his biological work. That bearing on parasitology will be considered in some detail, whereas the later, more biochemical, work will be considered briefly as fuller accounts of it have already been given by E. F. Hartree (Biochem. J.89, 1–5), T. R. R. Mann (Biogr. Mem. Fellows R. Soc.10, 185–207) and E. C. Slater (Enzymologia, 26, 313–20). In the brief survey of his biochemical work I have made much use of summaries and reports written by Keilin himself and I have often given the essence of lines of work summed up in his own words which express his conclusions with the elegance, lucidity and brevity characteristic of his style. As I was closely associated with David Keilin and in almost daily contact with him for just on forty years, it is natural that the following account will to some extent reflect my own interests and those aspects of his personality which became manifested during long and intimate contact both in private life and collaboration in scientific and administrative work.

Laboratory experiments indicated that third-stage larvae of T. axei and T colubriformis were able to penetrate soil to a depth of 12 in. or more, although most larvae remained in the upper 2 in. No geotaxis was observed.

Heavy downflows of water caused severe mortalities but did not affect the rate of movement of surviving larvae through the soil. Very limited random lateral movement occurred in turf, being more pronounced in the presence of faeces. Prolonged desiccation was extremely lethal to infective larvae. The activity of larvae was directly and linearly proportional to normal temperatures. The average activity of groups of larvae declined with time.

The author wishes to acknowledge the assistance and facilities provided by Professor B. G. Peters and the other members of staff of Imperial College Field Station, and to thank Dr C. W. R. Spedding for helpful criticism and supplying culture lambs; and Dr T. E. Gibson and Dr P. Silverman for supplying pure cultures of trichostrongyle larvae.

A standard culture technique is described for obtaining the various developmental stages of Trichostrongylus colubriformis, and a suitable technique for assessing the effect of methyl bromide fumigations is outlined.

Doses of 10, 25 and 150 mg. hr./l. were required to give an LD50 for eggs, pre-infective and infective larvae respectively. The age of infective larvae, the relative humidity and the temperature all influenced the toxicity of the fumigant. Soil, and more especially faeces, sorb the fumigant and probably protect larvae from its full effect. Infective larvae of T. axei, T. retortaeformis and H. contortus were also found to be susceptible to methyl bromide.

The activity of larvae surviving fumigation was considerably reduced, but these larvae were still capable of infecting guinea pigs.

Methyl bromide killed common pasture grasses at doses of 150–250 or more mg. hr./l., depending on the temperature. Chloropicrin was found to be highly toxic to grass and unsuitable where the pasture is required undamaged.

The results are discussed in the light of current theories of the mode of action of methyl bromide on insects and what is known at present of the physiology of nematodes.

It is concluded that under certain conditions, methyl bromide might be used to eliminate infestations of trichostrongyle larvae on pasture, without killing the pasture.

I am indebted to Professor B. G. Peters and other members of the staff of the Imperial College Field Station and Dr C. W. R. Spedding of the Grassland Research Institute, Hurley, for their help in this work and their advice and criticism.

In a small-scale trial in which boxes of grass infested with Trichostrongylus colubriformis were treated with methyl bromide, a dose of 100 mg. hr./l. prevented guinea-pigs picking up the infection when they grazed the grass at a later date. A practical method is described for fumigating small areas of pasture. In a field trial, sheep were used to test plots of pasture fumigated with doses which did not kill the grass but the results were inconclusive. These experiments are discussed and also the possible value and applications of this technique.

I am indebted to Professor B. G. Peters and to the other members of the Imperial College Field Station, particularly Dr A. B. Page and Dr O. F. Lubatti, for their help and criticism during the work. Thanks are also due to Dr William Davies, the Director of the Grassland Research Institute, Hurley, for permission to carry out the field experiment there, and also to Dr C. R. W. Spedding and Mr T. Brown for their co-operative assistance and advice throughout this work.

The route of migration of Cercaria X within the host snail, Lymnaea stagnalis, has been demonstrated by means of serial sections and a method of numerical assessment of the distribution of cercariae in the various tissues and organs of the body.

The main pathway along which migrating cercariae travel is the blood system, which differs from the route in the fish second intermediate host. The cercariae enter the main visceral haemocoel by rupture of the wall of the digestive gland and from there enter the main venous vessels deploying through the rectal sinus, sub-renal sinus, lung and heart. The cercariae travel from the heart via the mantle arteries to localize in the mantle sinuses prior to emergence.

The inner surface of the mantle, and in particular its leading edge, is the main site of emergence, and escape of cercariae is, in the final stages at least, an active process. It is suggested that passage of parasites in the blood stream is partly active.

Migration and subsequent emergence is probably facilitated by the activity of the oral cap of spines, the general activity of the cercaria and the activity of the oral and ventral suckers.

A tissue response to the presence of cercariae in the blood vessels and tissues of the mantle is evident, the cells responsible for this being, possibly, actively phagocytizing amoebocytes of the mollusc's blood. The swellings in the epidermis of the mantle in heavy infections may be the result of an accumulation of these cells.

Grateful acknowledgement is expressed to Professor J. Brough for the facilities and encouragement provided during the course of this study. This work was carried out during the tenure by A. J. Probert of a D.S.I.R. Research studentship.

The effects of parasitism by five species of larval Digenea on the digestive gland cells of Littorina saxatilis (Olivi) subsp. tenebrosa (Montagu) were investigated by comparing histochemically stained cells in healthy, starved and parasitized animals.

Two distinct effects were apparent: (a) The germinal sacs compress the walls of the digestive gland tubules and block the lumen. Since food cannot pass this block, digestive gland cells behind undergo starvation autolysis. (b) In digestive gland tubules not cut off from their food supply, glucose, glycogen, glycoproteins and lipid food storage globules are reduced and there is a compensatory increase in the number of food vacuoles in the digestive cells; an increase in glucose but a decrease in glycogen in the visceral haemocoel; and an increase in the number of secretory cells, perhaps in order to eliminate the parasite's excretory products from the visceral haemocoel.

While some variation of detail may be found in different species of the hermaphroditic Digenea, the process of spermatogenesis conforms to a common plan throughout the group. There is divergence of opinion concerning the interpretation of the structure of the ripe spermatozoon.

In the testis of a young animal, primary, secondary and tertiary spermatogonia are present. The tertiary spermatogonia divide to give a group of eight primary spermatocytes that undergo the first division of the meiotic phase to form sixteen secondary spermatocytes. As a result of the second division of the meiotic phase, the sixteen secondary spermatocytes give a group of thirty-two spermatids that through a complicated series of changes are transformed into spermatozoa. During spermateleosis much of the cytoplasm of the late spermatid is sloughed off.

Recent research, including the application of the Feulgen technique, the examination of living material, and electron microscopy, supports the view that the ripe spermatozoon is composed of an elongate head, and a flagellum. The head, with the exception of its external sheaths, consists of nuclear material.

The flagellum of the sperm of Fasciola hepatica is Feulgen-negative and is composed of a proximal and a distal region. Its proximal part is provided with two axial filaments, a middle fibril and an external sheath. The axial filaments and the middle fibril are continued into the distal region where the external sheath appears to be absent. The external sheath of the flagellum of the sperm of Haematoloechus ( = Ostiolum) medioplexus is said to be composed of thirty-six fibrils. Claims that the flagellum of some digeneans contains a centriole, a middle-piece, and mitochondria have not so far been substantiated by means of electron microscopy.

Polycercus lumbrici, previously described only by Metchnikov (1868) in Russia, has been found in Scotland in the earthworm Allolobophora terrestris.

Experimental infestation of earthworms with Polycercus lumbrici as a result of feeding them with the eggs of Paricterotaenia paradoxa shows that the proliferating cysts are the larvae of the latter as suggested by Joyeux and Baer (1939).

There is evidence that Polycercus may induce parasitic castration of the earthworm in which it occurs.

Metchnikov's description of the development of the larva is confirmed for the most part but is modified to include a retroversion of the larva in its early stages of development. The development of the scolex is shown to follow a pattern similar to that of Cysticercus fasciolaris in that the hooks are borne on a ‘prebulb’ which later invests a ‘bulb’, the former giving rise to the muscles of the walls of the double sac of the rostellum, the latter to the hook-elevator muscles.

The rostellar hooks of P. paradoxa are elevated by the action of a double layer of muscles which radiate from a conical invagination of the bulb. These muscles, in association with the muscles of the wall of the rostellum, elevate the hooks by modifying the shape of the rostellum rather than by direct action on the hooks.

The nervous system and the excretory system in the scolex of P. paradoxa are briefly described and the nature of the ‘glandular’ structure of the rostellum is discussed.

I wish to record my gratitude to Professor H. G. Callan for affording me the facilities for my work and to the many persons, too numerous to mention by name, who helped me to obtain my material. I am particularly indebted to Mr D. R. R. Burt for guidance and constructive criticism at all stages of my work and to Mr I. M. Sandeman for much helpful advice.

The work was carried out during the tenure of Research Grants from Madras College, St Andrews, and from the D.S.I.R., to whom I would express my appreciation for their aid.

A large number of nematodes from Protemnodon rufogrisea, P. bicolor, P. dorsalis, and Macropus canguru, from southern Queensland, M. robustus from Western Australia, and Megaleia rufa (? locality) have been examined. The species described in this paper are: Pharyngostrongylus macropodis Yorke & Maplestone, P. kappa, sp.nov. (syn. P. macropodis in part), P. eta Johnston & Mawson, P. lambda, sp.nov., Rugopharynx epsilon (J. & M.), R. delta (J. & M.), R. zeta (J. & M.) (syn. R. longibursaris (Kung)), R. theta (J. & M.), Coronostrongylus coronatus J. & M.; a new genus is proposed, Oesophagonastes, with type O. kartana syn. Spirostrongylus kartana Mawson to include O. parma syn. Pharyngostrongylus parma J. & M., O. gallardi syn. P. gallardi J. & M., and O. leptos, sp.nov.

Most of the parasites were sent for examination by Dr M. J. Mackerras, then of the Queensland Institute of Medical Research. Those from Western Australia were from Dr T. Ealey of the Zoology Department, Monash University, Victoria, who was working on Macropus robustus in the north of that state. I am most grateful to both of these helpers.

The present studies were designed to characterize the response of sheep infected with N. spathiger to challenge infection with the same species.

The results showed that: (1) the level of worm counts attained on secondary infection with the parasite was much reduced compared with the counts recorded following initial infection; (2) the relative fecundities of parasite populations which developed in individual hosts following challenge infection, ranked in magnitude in an order which was similar to that established following initial infection; (3) as fecundity was highly correlated with the number of egg-laying females and the number of males and females was nearly equal in individual specimens, it was concluded that worm burdens were also ranked for individual animals on primary infection and following challenge; and (4) the animals of the control group remained as susceptible as the animals of groups 1 and 3 at the time of their primary infection; consequently, in the absence of previous immunological experience due to infection, there was no apparent change in susceptibility of the animals due to an effect of age during the course of the 17 weeks of the experiment.

We are greatly indebted to Dr Michael Tallis, Division of Mathematical Statistics, C.S.I.R.O., for advice and assistance with the statistical analyses, and to Dr D. F. Stewart for his interest and constructive criticism during the course of these studies. Our thanks are also due to Mr E. Teleki and Misses Helen Giller and Lindy Stothart for their valuable technical assistance.

The cuticle of adults of Nippostrongylus brasiliensis has been described using histological, histochemical and ultrastructural techniques.

The cuticle has the following layers: an outer triple-layered membrane; a single cortical layer; a fluid-filled layer which is traversed by numerous collagen fibrils; struts which support the fourteen longitudinal ridges of the cuticle and which are suspended by collagen fibrils in the fluid-filled layer; two fibre layers, each layer apparently containing three layers of fibres; and a basement lamella.

The fluid-filled layer contains haemoglobin and esterase.

The muscles of the body wall are attached to either the basement lamella or to the fibre layers of the cuticle.

The mitochondria of the hypodermis are of normal appearance.

The longitudinal ridges of the cuticle appear to abrade the microvilli of the intestinal cells of the host.

Possible functions of the cuticle are discussed.

I wish to thank Dr P. Tate, in whose department this work was done, for helpful suggestions and criticism at all stages of this work, and Mr A. Page for technical assistance. I also wish to thank Professor Boyd for permission to use the electron microscope in the Department of Anatomy.

Female mice survived longer than male mice when infected with 400 eggs but not when infected with 100 eggs of Amplicaecum robertsi.

The larvae grew to greater lengths in young than in older mice; smaller worm burdens were also found in the older animals.

The growth of the larva was associated with the total worm burden of the host. As the worm burdens, of 40-day-old female mice, increased up to 281 individuals the length of the larvae attained after 14 days decreased; when the larval population increased above this number the lengths of the worms also increased.

These crowding effects are discussed in relation to competition for nutrients, immunological suppression of larval growth and to the ability of large larval populations to swamp the resistance mechanisms of the host. It is suggested that the increased growth of the larvae in large populations is due to their ability to break down the liver of the infected host more efficiently than small populations. However, this eventually leads to the deaths of the host through total liver breakdown.

This work was done during the tenure of a Senior Research Fellowship from the Wool Research Board of Australia. My best thanks are due to Mr J. James for his assistance with the statistics, to Professor J. F. A. Sprent for his help and advice and to Mr C. Wilkinson for his assistance.

Rhabdostyla elongata sp.nov. is described and figured. This peritrich was found as an epibiont on Lernaea barnimiana, a copepod parasitic on the Nile Perch, Lates niloticus in Lake Albert.

Epistylis articulata Fromentel is recorded as an epibiont of Dolops and Argulus, which parasitize a wide range of fish in Lake Albert.

My thanks are due to Mr M. J. Holden who identified the fish as they were removed from the nets, and to Dr G. Fryer, who identified Lernaea barnimiana and has helped with the literature. I am also grateful for a grant from the Leverhulme Trust which made possible the collection of this material.