In order to determine whether host availability limits triatomine population growth, 5th-stage Panstrongylus megistus were maintained in feeding chambers containing 0, 1, 2, or 3 mice. During the 5-day feeding period, triatomines exposed to two or three mice gained significantly more weight than did bugs exposed to one mouse. In addition, half of the bugs exposed to two or three mice molted, as compared to one-fifth of the P. megistus exposed to one mouse. Thus, weight gain and molting were related to host density. In contrast, bug mortality was related to the triatomine-mouse ratio, being greatest among bugs exposed to one mouse. Twenty-nine nonplastered mud-stick houses in a Chagas' disease endemic area were censused and examined for triatomines. About 70% of houses with greater than or equal to 4 persons contained dense bug populations, while only 20% of houses with 1-3 persons were densely infested. Moreover, blood-meal identifications demonstrated that two-thirds of the P. megistus collected from these houses fed on man. The density of triatomines present in infested houses is related to the number of persons available as hosts.