Insufficient use has been made of ecological data concerning potential hosts in studies to determine the life cycles of zoonotic parasites and pathogens. Factors such as the geographical distribution of hosts, the altitudes at which they live, their affinities for specific habitats, their vertical distribution within the habitat, and the periodicity of their activities have bearing on the hosts' predisposition to involvement in disease cycles. Diets and feeding habits may determine the likelihood of acquiring infection. Reproductive characteristics determine whether a species is suitable as a reservoir or as an amplifying host. Behavioral factors, such as selection of a particular kind of nest site, may also predispose the involvement of the host with parasites and pathogens. Behavior patterns may determine the maximum population densities of hosts. Estimates of population sizes, of relative abundances of species, and of the involvement of species in disease cycles may be strongly influenced by the collecting and sampling methods that are used and also by the behavioral response of the mammals toward collecting devices, such as traps.
No focalization of rats (Rattus tiomanicus and R. argentiventer) infected with Rickettsia tsutsugamushi could be discerned over a 500 m trapping transect at the border between a forest and lalang grass (Imperata cylindrica). R. tiomanicus appeared to occupy 250 m of the transect on the average and had periods during which infections were observed which averaged 97 days. Calulations indicated that more than 50% of individuals become infected over their life-time. The high rate of infection in this and other areas described in earlier publications and the habits of the rats suggest that infected mites are densely and widely dispersed in the areas studied in Malaysia.
The overall comparisons of habitats are given in (Table III). The habitats are arranged in order of extent of alterations by man, with the least disturbed at the top. The highest average blood isolation rates came from the least disturbed areas. The highest monthly maximal rickettsial isolation rates from blood and maximal prevalence rates of antibody per month were also obtained at Bukit Lanjan, the habitat least altered by activities of man. The lowest average blood isolation rate (6%) and the lowest monthly maximal rickettsial isolation and antibody prevalence rates were obtained at Bukit Mandol, the habitat most extensively and intensively altered by man. The intermediate habitats had intermediate rates. We caution anyone interpreting these observations, however, in terms of human disease, which seem to be associated with hyperendemic foci. Here we are not dealing with hyperendemicity from the standpoint of human disease, but present evidence of widespread endemicity from which hyperendemic foci may derive. Also, we have not yet identified the prevalent strains and do not know their infectivity to man.
Rickettsia tsutsugamushi isolations were attempted from blood samples obtained from rats captured in four adjacent habitats near Kuala Lumpur, Malaysia. Antibody surveys were also made. Rickettsial infections were most frequent in rats captured in the forest and in lalang grass (Imperata cylindrica) and least frequent in the most extensively disturbed habitat, an Orang Asli (aborigine) village. Forest rats such as Rattus sabanus (31%), as well as rats in the subgenus R. (Rattus), i.e. R. tiomanicus (26%) and R. argentiventer (35%) had frequent active infections. The house rat R. exulans had less frequent infections (15%). Frequency of antibody occurrence followed a similar pattern. No marked seasonal differences in the frequency of infections could be detected during the 18-month study.
Numbers of L. (L.) deliense larvae were determined in adjacent habitats over a 16 month period. Both R. argentiventer and R. tiomanicus were highly efficient hosts for L. (L.) deliense. R. argentiventer was host to significantly greater numbers of chiggers per rat than was R. tiomanicus. The 2 habitats were similar in numbers of chiggers collected. No consistent correlation was apparent between numbers of chiggers and any single weather factor, but the chigger population seemed to be adversely affected by a 2 month period during which total evaporation greatly exceeded total rainfall. Direct fluorescent antibody examination of tissues from unfed L. (L.) deliense showed that 2 of 420 larvae (0.5%) contained organisms morphologically resembling R. tsutsugamushi. Considering the vector load and numbers of chiggers being returned to the ground by a given host, a rate of 0.5% appeared adequate to account for the prevalence rate of R. tsutsugamushi observed in the 2 host species.