Twenty old houses in a rural area of central Brazil were sampled monthly for Triatoma infestans during 16 months. The samples indicated peak adult emergence between January and May, preceded by peak emergence of stage V, IV and III nymphs respectively. Stage I and II nymphs were rarely taken in the samples. A much smaller peak of adult emergence occurred in September; thus the two peaks of adult emergence during the year were consistent with the approximately six months duration of egg-to-adult development usually recorded in laboratory studies. During January, 1977, when the study began, all houses were searched for dead bugs, exuviae, eggs and eggshells. These samples were used to calculate stage mortalities, from which a simple life-table was constructed. These data were combined with laboratory data concerning stage development times and female longevity and fecundity. Sensitivity analysis of the life-table showed it to be most sensitive to small changes in daily mortality rate and total egg-to-adult development time. The data were used to advance an hypothesis that the size of domestic populations of T. infestans is mainly controlled by changes in the egg-to-adult development time and, to a lesser extent, by changes in female reproductivity. Both these factors are in turn controlled by the nutritional status of the bug population, which, if the supply of food, i.e. number of hosts, is constant, is dependent upon the density of the bug population. It is argued that this mechanism of density regulation functions within limits set by the prevailing ambient temperature.