The nature of parasitism at the population level is defined in terms of the parasite's influence on the natural intrinsic growth rate of its host population. It is suggested that the influence on this rate is related to the average parasite burden/host and hence to the statistical distribution of parasites within the host population.

Theoretical models of host–parasite associations are used to assess the regulatory influence of parasitic species on host population growth. Model predictions suggest that three specific groups of population processes are of particular importance: over-dispersion of parasite numbers/host, density dependence in parasite mortality or reproduction and parasite-induced host mortality that increases faster than linearly with the parasite burden. Other population mechanisms are shown to have a destabilizing influence, namely: parasite-induced reduction in host reproductive potential, direct parasite reproduction within the host and time delays in the development of transmission stages of the parasite.

These regulatory and destabilizing processes are shown to be commonly observed features of natural host-parasite associations. It is argued that interactions in the real world are characterized by a degree of tension between these regulatory and destabilizing forces and that population rate parameter values in parasite life-cycles are very far from being a haphazard selection of all numerically possible values. It is suggested that evolutionary pressures in observed associations will tend to counteract a strong destabilizing force by an equally strong regulatory influence. Empirical evidence is shown to support this suggestion in, for example, associations between larval digeneans and molluscan hosts (parasite-induced reduction in host reproductive potential counteracted by tight density-dependent constraints on parasite population growth), and interactions between protozoan parasites and mammalian hosts (direct parasite reproduction counteracted by a well-developed immunological response by the host).

The type of laboratory and field data required to improve our understanding of the dynamical properties of host–parasite population associations is discussed and it is suggested that quantitative measurement of rates of parasite-induced host mortality, degrees of over-dispersion, transmission rates and reproductive and mortality rates of both host and parasite would provide an important first step. The value of laboratory work in this area is demonstrated by reference to studies which highlight the regulatory influence of parasitic species on host population growth.

Isolation of blood and intracellular forms of Trypanosoma cruzi was made mainly from rats (90–110 g) which had received 580 rad of whole-body γ-irradiation not more than 24 h before subcutaneous inoculation with 107 trypomastigotes of the Sonya strain of T. cruzi. Unirradiated chinchillas (250–350 g) were, however, used for some experiments. Blood forms were isolated using a technique involving differential centrifugation to remove most of the erythrocytes and DEAE–cellulose chromatography to remove the remaining blood cells. Overall recoveries were usually in the range 30–70%. Parasites were mainly (approximately 98%) broad forms and were motile, metabolically active (as judged by respiratory and radio-tracer incorporation studies) and had lost none of their infectivity for mice. Intracellular forms were isolated from hind-limb muscle tissue. This was disrupted in an MSE tissue homogenizer and the homogenate incubated with DNase, collagenase and trypsin. Parasites, contaminated only by a few blood cells, were then obtained by differential centrifugation. For purer preparations, a terminal sucrose gradient step was used. Recoveries ranged between 40 and 70%. About 1–3% of the parasites isolated were epimastigotes and trypomastigotes; the remainder are probably best collectively termed ‘amastigotes’, though they were pointed and most had a short, free flagellum. They were undamaged as judged by light and electron microscopy and metabolically active as judged by respiratory and radio-tracer incorporation studies. However, the infectivity for mice of both these purified preparations and the initial cell homogenates could be accounted for by the epimastigotes and trypomastigotes present in them. Preliminary biochemical studies with isolated parasites have shown that blood, intracellular and culture forms of T. cruzi have a respiratory system which is in part sensitive to CN- and that all forms synthesize nucleic acids and proteins when incubated in vitro. There appears, however, to be a lack of DNA synthesis in blood stages, and thus it is not surprising that these forms do not divide.

The development of resistance by the Houghton strain of Eimeria tenella to the anticoccidial drugs amprolium, clopidol and methyl benzoquate has been studied. Resistance to amprolium and clopidol developed more readily in experiments where a large number of coccidia were exposed to the drug, either by increasing the number of oocysts in the inoculum or by increasing the number of birds in the group. When 45 birds were given 2·0 × 106 oocysts, resistance to amprolium and clopidol appeared after 6 and 7 passages respectively. In previous experiments, under similar conditions, resistance to robenidine developed after 6 passages, suggesting little difference between these three drugs. Resistance to amprolium and clopidol arose gradually as the concentration of drug was increased, but resistance to methyl benzoquate appeared in a single step from sensitivity to high-level resistance. Both amprolium and clopidol-resistant lines showed an 8-fold reduction in drug sensitivity. Attempts to measure the degree of resistance by calculation of the ED50 were unsuccessful.

The fine structure and development of the endogenous stages of an embryo-adapted strain of Eimeria tenella in chicken embryos is described. First-generation merozoites were formed by endodyogeny in uninucleate developmental stages as well as schizogony and the endoplasmic reticulum of merozoites of the second-generation schizonts appeared as a circular structure, often enclosing other organelles. In contrast to the parent strain, development of the second-generation schizonts was restricted to cells of epithelial origin. These observations confirm that the complete endogenous cycle of the embryo-adapted strain of E. tenella is restricted to epithelial cells of the chorio-allantoic membrane (CAM) and that no major ultrastructural changes have occurred as a result of repeated embryo passage.

Rediae of Cryptocotyle lingua were studied using transmission and stereoscan electron microscopy. The mouth is surrounded by numerous ciliated sensory structures. Posteriorally, the body wall is concentrically folded and its surface area is greatly increased by the closely ridged configuration of the outer membrane of the tegument. Attached to the surface, especially at the anterior end, are numerous, small spherical vesicles. At the posterior end there is a retractable papilliform process. The appearance of the surface of the birth papilla is variable. The birth pore is lined by an epithelium which is specialized to withstand extreme distortion. A single cilium is frequently visible in the birth pore lumen. The release of each cercaria is preceded by the emission of a tiny drop of fluid from the birth pore. It is suggested that this fluid acts as a lubricant to the passage of the cercaria through the pore.

Two lines of guinea-pigs, whose susceptibilities to infection with the parasitic nematode Trichostrongylus colubriformis differ significantly, have been established. This difference is probably based on genetically determined differences between the ability of members of each line to bring about immune expulsion. of the parasite.

Infection of rats with 2000 infective juveniles of Nippostrongylus brasiliensis and of lambs with 60000 infective juveniles of Nematodirus battus results in a well-marked immunity to these nematodes in their respective hosts. There is a fall in the adenylate energy charge value of these nematodes during the course of these infections, reaching values of 0·37 in males and 0·27 in females of N. brasiliensis, and 0·31 in males and 0·23 in females of N. battus towards the end of the infections. In hosts given relatively small numbers of infective juveniles, the values for the nematodes removed from the hosts late in the infection remain at a relatively high level. These results indicate that the immune response of the host may affect the energy status of these nematodes, and this could help to explain their subsequent expulsion from the immune host.

There were no differences in mean intestinal worm burdens 8 days after subcutaneous injection of 4000 infective larvae of Strongyloides ratti into rats in dioestrus, pro-oestrus, oestrus and metoestrus. Thus, changes in the hormonal environment of the migrating larvae dependent on the oestrous cycle did not alter the worms' destination or affect their potential for development. In particular, the results are prima facie evidence that prolactin is not, on its own, responsible for the re-orientation of larvae in the tissues of nursing mothers. Other sources of variability in experimental S. ratti infections are analysed and the ‘exact dose’ technique offered as a corrective for some procedural errors.

The size and form (length, width, and length: width ratio) of eggs of Diphyllobothrium dendriticum, D. ditremum and D. laturn vary considerably among individual worms within each species. The size of eggs varies with host species and a decrease in egg size with increasing intensity of infestation is indicated. The egg size of D. laturn increases during the first 10–12 days of egg production. For single worm infections in golden hamsters the mean egg length and width of D. ditremum are significantly smaller than the corresponding means of D. dendriticum and D. latum, while D. latum has significantly wider eggs than D. dendriticum. As taxonomic characteristics, egg size and form may contribute to species delimitation at the population level. For identification at the individual level the best possible accuracy is about 80%. This accuracy is considerably reduced when variation in host species and intensities of infestations are introduced. Scanning electron microscope studies did not reveal any differences among eggs of the three species.