Sex of the fetus is genetically determined such that an equal number of sons and daughters are born in large populations. However, the ratio of female to male births across human populations varies significantly. Many factors have been implicated in this. The theory that natural selection should favour female offspring under suboptimal environmental conditions implies that pathogens may affect secondary sex ratio (ratio of male to female births). Using regression models containing 13 potential confounding factors, we have found that variation of the secondary sex ratio can be predicted by seroprevalence of Toxoplasma across 94 populations distributed across African, American, Asian and European continents. Toxoplasma seroprevalence was the third strongest predictor of secondary sex ratio, β = −0·097, P < 0·01, after son preference, β = 0·261, P < 0·05, and fertility, β = −0·145, P < 0·001. Our preliminary results suggest that Toxoplasma gondii infection could be one of the most important environmental factors influencing the global variation of offspring sex ratio in humans. The effect of latent toxoplasmosis on public health could be much more serious than it is usually supposed to be.

Toxoplasma gondii, the coccidian parasite, is known to induce changes in the behaviour of its intermediate hosts. The high prevalence of this parasite in the human population (20–80%) offers the opportunity of studying the influence of the parasite on human behaviour by screening of a normal population. A total of 224 men and 170 women were tested for toxoplasmosis and their personality profiles were measured by Cattell's questionnaire. Highly significant differences between Toxoplasma-infected and uninfected subjects were observed (P < 0·01). For men the factors G (low superego strength), L (protension), O (guilt proneness), and Q2 (group dependency) were positively influenced in infected subjects. For women the prevailing factors were A (affectothymia, P < 0·01), L (alaxia), O (untroubled adequacy) and Q2 (self-sufficiency). To reveal whether toxoplasmosis induces personality factor-shifts or whether certain combinations of personality factors influence the probability of acquiring Toxoplasma infection, we examined the personality profiles of 164 male patients diagnosed with acute toxoplasmosis during the past 13 years. The existence of a positive correlation between the duration of latent toxoplasmosis and the intensity of superego strength decrease (P < 0·02) suggested that the decrease of superego strength (the willingness to accept group moral standards) was induced by T. gondii infection.

The sex ratio may be influenced by many factors, such as stress and immunosuppression, age of parents, parity and sex of preceding siblings. In animal systems, parasitism often changes the sex ratio of infected hosts, which can increase the probability of their transmission. The most common human protozoan parasite in developed countries, Toxoplasma gondii (prevalence 20%−80%), is known to change the behaviour of its intermediate hosts, thereby increasing the probability of transmission to its definitive host (the cat) by predation. The intermediate hosts, which under natural conditions are rodents, serve as the vector for Toxoplasma. Therefore, we speculate that Toxoplasma can alter the secondary sex ratio (i.e. male to female ratio in the offspring) of infected females to increase the proportion of (congenitally infected) male offspring, which are the more migratory sex in most rodent species. Here we studied the sex ratio of experimentally infected laboratory mice, expressed here as the proportion of males in the litter. In accordance with our hypothesis and results of previous retrospective cohort studies on human subjects, mice with toxoplasmosis produced a higher sex ratio than controls, in the early phase of latent infection. In the later phase of infection, mice with congenital toxoplasmosis had a lower sex ratio than controls, which is in accord with the Trivers-Willard hypothesis of sex ratio manipulation, suggesting that females in poor physical condition give birth to more female offspring.

The heteroxenous protozoan parasite Toxoplasma gondii is transmitted from the intermediate host (any warm-blooded animal) to the definitive host (members of the felidae) by carnivory. The infected intermediate hosts develop several specific behavioural changes that are usually considered products of manipulative activity of the parasite aimed to increase the probability of its transmission to the definitive host. Among other changes, the infected rodents were shown to have impaired learning capability. All previous studies were done 2–6 weeks after the infection. Therefore, it was difficult to resolve whether the observed impairment of learning processes was a result of acute or latent toxoplasmosis, i.e. whether it was a side-effect of the disease or a product of manipulation activity. Here we studied the learning capability of Toxoplasma-infected mice in the static rod test and 8-arm radial maze test and their spontaneous activity in the wheel running test 10 weeks after the infection. The infected mice achieved worse scores in the learning tests but showed higher spontaneous activity in the wheel running test. However, a detailed study of the obtained results as well as of the data reported by other authors suggested that the differences between infected and control mice were a result of impaired ability to recognize novel stimuli rather than of impaired learning capacity in animals with latent toxoplasmosis.

Toxoplasma gondii is known to induce specific behavioural changes in its intermediate hosts. This is usually considered to be an evolutionary adaptation aimed to increase the probability of transmission of the parasite into its definitive host, the cat, by predation. In rodents an increase of reaction time as well as many other specific behavioural patterns have been observed. Here we report the results of our double blind study showing the significantly longer reaction times of 60 subjects with latent toxoplasmosis in comparison with those of 56 controls. Moreover, the existence of a positive correlation between length of infection and mean reaction time suggested that slow and cumulative effects of latent toxoplasmosis rather than a one-step (and possibly transient) effect of acute toxoplasmosis disease are responsible for the decrease of psychomotor performance of infected subjects. To our knowledge, this is the first study confirming the existence of such parasite-induced changes in human behaviour that could be considered in evolutionary history of the human species as adaptive from the point of view of parasite transmission.

Previous laboratory studies have indicated that tenebrionid beetles infected with Hymenolepis diminuta display behavioural alterations. These are assumed to increase the likelihood of this intermediate host being predated by the Rattus sp. definitive host, and hence provide support for the Manipulation Hypothesis. We tested this hypothesis by both assessing beetle behaviour and predation rates in semi-naturalistic environments. Behavioural assays showed that infected beetles were more often exposed (not concealed under boxes) than uninfected beetles. However, there were no differences in predation rates between infected and uninfected beetles. We discuss this in terms of the historical and evolutionary contexts under which such altered behaviours could evolve and their implications for the Manipulation Hypothesis.

We have investigated the influence of heteroxenous coccidia (Apicomplexa: Eimeriorina) on the predation risk of intermediate hosts. Voles infected with Frenkelia spp. were found more frequently in buzzards' (Buteo buteo) prey than among snap-trapped rodents. To eliminate the possibility of traps selecting for uninfected rodents, a laboratory experiment was performed. Mice experimentally infected with Sarcocystis dispersa seemed to be more likely caught by the final host, the long-eared owl (Asio otus); this result was confirmed by a mathematical model. Field data confirmed the adaptive value of parasite-induced changes. The increase of predation is directed towards the specific final host only or is non-specific. In the populations studied the probability of predation of parasitized individuals by the specific predator was increased.