The association between Trypanosoma cruzi parasitaemia in dogs and cats and Tryp. cruzi infection rates in domestic Triatoma infestans was studied in a cross-sectional survey of 31 houses (89%) in the rural villages of Trinidad and Mercedes, north-west Argentina, where no spraying of insecticides had ever been done. Similar prevalence rates of parasitaemia, determined by xenodiagnosis, were recorded among 68 dogs (41.2%) and 28 cats (39.3%). Bug infection rates were significantly associated with the presence of infected cats (those with positive xenodiagnosis) stratified by the number of infected dogs (relative risk = RR = 1.90; 95% confidence interval = CI = 1.51-2.38), and with the number of infected dogs stratified by the presence of infected cats (RR = 2.71; CI = 1.81-4.07). The percentage of infected bugs in houses with and without children stratified by the presence of infected dogs or cats was not significantly different (RR = 0.69; CI = 0.45-1.05). The combined effect of infected dogs and infected cats on bug infection rates fitted closely with an additive transmission model. Bug infection rates were significantly higher when infected dogs shared the sleeping areas of people than when they did not (RR = 1.79; CI = 1.1-2.91). Our study showed that infected dogs and infected cats increase the risk of domestic transmission of Tryp. cruzi to T. infestans.