Dr David Livingstone died on May 1st 1873. He was 60 years old and had spent much of the previous 30 years walking across large stretches of Southern Africa, exploring the terrain he hoped could provide new environments in which Europeans and Africans could cohabit on equal terms and bring prosperity to a part of the world he saw ravaged by the slave trade. Just days before he died, he wrote in his journal about the permanent stream of blood that he was emitting related to haemorrhoids and the acute intestinal pain that had left him incapable of walking. What actually killed Livingstone is unknown, yet the years spent exploring sub-Saharan Africa undoubtedly exposed him to a gamut of parasitic and other infectious diseases. Some of these we can be certain of. He wrote prolifically and described his encounters with malaria, relapsing fevers, parasitic helminths and more. His graphic writing allows us to explore his own encounters with tropical diseases and how European visitors to Africa considered them at this time. This paper outlines Livingstone's life and his contributions to understanding parasitic diseases.

Trypanosoma congolense, T. brucei and T. simiae isolated from wild-caught Glossina pallidipes were fed to laboratory-reared G. morsitans centralis and G. m. morsitans to determine the effect of host blood at the time of the infective feed on infection rates. Bloodstream forms of trypanosomes were membrane-fed to flies either neat, or mixed with blood from cows, goats, pigs, buffalo, eland, waterbuck and oryx. The use of different bloods for the infective feed resulted in differences in infection rates that were repeatable for both tsetse subspecies and most parasite stocks. Goat, and to a lesser extent, pig blood facilitated infection, producing high infection rates at low parasitaemias. Blood from cows and the wildlife species produced low infection rates, with eland blood producing the lowest. Addition of D(+)-glucosamine (an inhibitor of tsetse midgut lectin) increased infection rates in most cases. These results indicate the presence of species-specific factors in blood that affect trypanosome survival in tsetse. In certain hosts, factors actually appear to promote infection. The nature of these factors and how they might interact with midgut lectins and proteases are discussed.

The epidemiology of trypanosome infections of Glossina pallidipes was studied at a riverine site in the Zambezi Valley, Zimbabwe for a period of 13 months. Over 9000 flies were captured using a single trap. These flies were dissected, screened for trypanosome infection, sexed, and aged using both wing fray and (for females) ovarian category indices. Midgut infections were identified to species using recently developed DNA probes. The overall prevalence of mature infections was 5·5%, comprising 3·1% Trypanosoma vivax-type and 2·4% T. congolense-type (which included very low prevalences of T. brucei, T. simiae and another Nannomonas species). The prevalence of infection increased with age. For T. vivax-type infections in flies aged by ovarian category this relationship could be described by a simple ‘catalytic’ model assuming a constant per capita rate of infection. For T. congolense-type infections this model tended to over-estimate prevalence in older age classes, implying that the rate of infection decreases with age, and/or that infected flies have higher mortality rates, and/or that a significant fraction of the population is resistant to infection. Prevalences of infection also varied between months. This variation was more marked for T. vivax-type infections and was negatively correlated with both temperature and rainfall. The shape of the age-prevalence relationship, however, did not vary significantly between months. These observations are not fully explained by variation in the age-structure of the tsetse population and are consistent with temporal variation in the rate of infection (rather than in the trypanosome developmental period or in effects of infection on fly mortality). Possible causes of this variation are discussed.

If species-specific male genitalia are courtship devices under sexual selection by cryptic female choice, then species-specific aspects of the morphology and behaviour of male genitalia should often function to stimulate the female during copulation. The morphology and behaviour of the complex, species-specific male genitalia of the tsetse fly, Glossina pallidipes Austen, were determined from both direct observations and dissections of flash-frozen copulating pairs; we found that some male genitalic traits probably function to stimulate the female, while others function to restrain her. The male clamps the ventral surface of the female's abdomen tightly with his powerful cerci. Clamping does not always result in intromission. Clamping bends the female's body wall and her internal reproductive tract sharply, posteriorly and dorsally, and pinches them tightly. The male performed sustained, complex, stereotyped, rhythmic squeezing movements with his cerci that were not necessary to mechanically restrain the female and appeared instead to have a stimulatory function. Six different groups of modified setae on and near the male's genitalia rub directly against particular sites on the female during squeezing. The designs of these setae correlate with the force with which they press on the female and the probable sensitivity of the female surfaces that they contact. As expected under the hypothesis that these structures are under sexual selection by female choice, several traits suspected to have stimulatory functions have diverged in G. pallidipes and its close relative, G. longipalpis. Additional male non-genitalic behaviour during copulation, redescribed more precisely than in previous publications, is also likely to have a courtship function. The elaborate copulatory courtship behaviour and male genitalia may provide the stimuli that previous studies showed to induce female ovulation and resistance to remating.

Genetic differentiations in natural populations of the tsetse fly, Glossina pallidipes Austen (Diptera: Giossinidae) collected from three ecologically distinct regions of Kenya, namely Lambwe Valley, Shimba Hills and Nguruman were analysed based on two enzymes, phosphoglucomutase (PGM) and phosphoglucose isomerase (GPI). The PGM gene frequencies of allele (a) of male and female flies from Lambwe Valley showed significant differences. A dendrogram of populations collected from different localities showed two clusters: one comprising Shimba Hills, Majimboni and Muhaka Forest and the other comprising Gendo, Homa Hills and Nguruman. This suggests mat different subpopulations of G. pallidipes have adapted to different habitats.

The diel activity patter n of Glossina austeni was studied on the south coast of Kenya, and in the laboratory. In the field, flies were sampled hourly by continuous catch from a standing vehicle, or by traps. In the laboratory, observations were made of the flies' take-off-responses to host odour at various times of the diel cycle, in a flight chamber. The results showed that G. austeni is day-active with some activity throughut the daylight hours, but with significant peaks occurring at 09.00–10.00h and between 14.00 and 17.00h, i.e. a roughly V-shaped pattern.

Plotting the movement of air through typical tsetse habitats in Zimbabwe revealed that it is liable to change direction by over 90° within a few metres. In winds of < 1 m/sec, there was a negative correlation between windspeed and the wind's tendency to meander, whether in thick bush or out in the open. And at the mean windspeed of wet season mopane woodland (0.3 m/sec), the wind changed direction by c. 15°/sec. Accordingly, host odour does not move downwind in simple trajectories that tsetse flies could easily follow. Even 5 m from a source, odour (modelled with smoke) approached a notional tsetse fly for a quarter of the time from >90° away from the true source direction. Also, air turbulence at common tsetse resting sites generates much nonsense information about the “true” direction of any odour-bearing wind. The suggested answer to how tsetse manage to find distant, invisible hosts, in spite of this confusing information from the wind, is that they progress by a biased random walk which is the outcome of their upwind anemotactic responses to odour-bearing wind and the proportion of the time for which the wind does blow in the “correct” direction. Computer simulation shows that this might work in principle.

In West and Central Africa attention has recently been drawn to the role of commercial crop plantations in the creation of new tsetse habitat, and the problems this poses to trypanosomiasis control. Reported here is the colonization by the tsetse, Glossina pallidipes Austen, of a plantation of a very different kind—exotic coniferous forest. This has been observed in two localities in Kenya: the Lambwe Valley in the west, and the Shimba Hills near the coast. The features of this habitat, and the factors associated with its exploitation by tsetse, are described, with particular reference to the Lambwe Valley situation. While extension of tsetse fly-belts by afforestation practise is unlikely to be a problem of widespread significance in Kenya, afforestation policy ought, in particular circumstances, to take into account the tsetse and trypanosomiasis situation, the Lambwe Valley being a case in point.

The pattern of singing activity of Glossina morsitans morsitans was monitored in LD 12: 12. Both males and females showed a clear “U”-shaped diurnal pattern of singing activity typical of other responses in tsetse flies. The morning peak comprised more activity than the evening peak. The significance of sound production and the possible endogenous control are discussed.

Although mating is a necessary prerequisite for the induction of ovulation in the tsetse fly, Glossina morsitans morsitans, neither the accessory gland substance from the male nor any testicular component is responsible for causing ovulation. Experimental evidence suggest that two neural factors, one originating from the uterus during mating act as an afferent stimulus, and the other originating from the ovary during oocyte maturation regulate a cerebral hormonal factor which controls the event of ovulation. It is probable that one successful copulation removes an inhibition for the release of an ovulation stimulating hormone from the neurohemal organ into the circulation. The rate of synthesis and/or transport of the hormone from the neurosecretory cells to the neurohemal organ is probably regulated by the rate of oocyte development.

It has become increasingly noticeable that common features of many forest and forest/ savanna foci of West African human trypanosomiasis (WAT) are peri-domestic tsetse populations closely associated with domestic pigs. Recent evidence from various sources almost completely proves that, in at least some WAT foci, domestic pigs play an active role as reservoirs of Trypanosoma brucei gambiense. On the basis of this assumption, various aspects of Glossina palpalis and G. tachinoides local distribution and ecology, in relation to domestic pig and human populations, are discussed.

In those WAT foci where domestic pigs are kept, tsetse are widely distributed in forest islands, forest/plantations, and riverine forests, but major concentrations are in villages containing pigs. Close pig/fly relationships at the village level, involving peri-domestic feeding, breeding and resting behaviour on the part of flies, could be important factors in maintaining T.b. gambiense transmission in pig populations.

Endemic transmission of WAT to man could occur, either in villages where low pig densities dictate that a significant proportion of the fly population must also feed on man, or away from the village as a result of contact with infected flies which have strayed from a village transmission cycle.

The development of epidemic conditions could be explained, on the basis of changes in pig density and resultant changes in fly feeding behaviour. The sudden reduction of a pig population could result in an increased number of infected flies feeding on man, until such time as the fly population became stabilized on a reduced pig population (or on other hosts if pigs disappeared totally). During the period of increased feeding of flies on man, T.b. gambiense transmission could be intense.

Minor flare-up of WAT could occur if small numbers of infected pigs were introduced into a village area which had no previous history of pig husbandry. If a loose man/fly association persisted in such a village, endemic WAT could become a characteristic feature of that village.

Because of the fluidity of pig distribution and density, infected pigs could be instrumental in dispersing T.b. gambiense from village to village within a WAT focus and contribute to the persistent character of the focus. Pig movements involving much greater distances could also be important to the transfer of T.b. gambiense strains from one focus to another.

A method of marking tsetse flies is described based on the transfer of fluorescent pigment from marked males to females during copulation. Powder applied between the coxae of the male lodges in the scutellar groove of the female during mating. Powder persists well on males before mating and on females afterwards. Individual males can mark several females, and the marks of different males can be detected on females after multiple mating. Aspects of behaviour which may possibly be investigated using this method are described.

Ecological surveys were carried out to assess the population characteristics and epidemiological importance of Glossina palpalis palpalis (Rob-Desv.), 1830 in the human trypanosomiasis focus of Vavoua in the forest zone of Ivory Coast. G. p. palpalis was widespread in all components of the ecosystem: forest, interstitial savannahs, coffee plantations, tracks and villages. Most flies were caught at the forest edge. Sex ratio and age composition of populations varied in relation to trapping sites. Open areas were dispersal places. Plantations provided resting, feeding and breeding sites giving rise to close man–fly contact. Here, teneral flies could become infected and later on transmit trypanosomes to plantation workers and their families. The role of man and domestic pigs together with that of fly behaviour and genetic factors in the transmission of the disease are discussed. Methods of tsetse control are suggested in the light of the findings.

Studies of some vector-borne diseases have shown genetic factors to be involved in the vector–parasite relationships and these results have often influenced measures taken to control such diseases. Novel methods of vector control have emerged from genetical studies, such methods having the dual advantages of being species specific and of causing little environmental damage. However, genetics has made little impact on trypanosomiasis research or control when compared with other vector-borne diseases, for reasons which are related to the biology of the tsetse fly and the trypanosome. Despite these natural limitations, genetical studies may yet provide useful information for those engaged in trypanosomiasis research. For example, chromosome and isozyme studies have shown that there is a large amount of genetic variation within tsetse populations, but it is not known whether such variation is related to the distribution of the disease. The infection rates of flies are known to vary between species and are dependent on fly-trypanosome interactions but it is not known whether these interactions have a genetic component. It is suggested that studies of the genetics of natural populations of tsetse, combined with selection experiments in the laboratory, could provide answers to these questions which might be of importance to control programmes and to studies of the epidemiology of trypanosomiasis.