38 



were free, while infection was rare in larv'ae undergoing 

 the last moults, becoming very rare or completely absent in young 

 lan-ae. Obser\^ations made on young larvae taken in the cottages 

 showed that, in nature, they suck blood from a very early stage ; 

 cannibalism, such as was observed in the laboratory in the case of 

 Triatoma sordida, seems to be exceptional in nature. These facts 

 offer a simple explanation of the seasonal appearance of the acute 

 infections of the disease. As Chagas noted early in the study of this 

 disease, newly-born children presenting signs of recent infection by 

 Trypanosoma cruzi are more common in the hot months of the year ; 

 the acute cases first appear in September, are relatively very abundant 

 in December and January, and are either very scarce or altogether 

 absent in the cold months beginning in May. The species of Triatoma 

 attain their final stages in the hot months, the individuals then met 

 with in the cottages being nearly all nymphs or adults. No significance 

 attaches to the fact that the cycle of Ttypanosonia cruzi does not depend 

 on the stages of the insect, careful experiments having shown that even 

 when undergoing the first and second laival moults, Triatoma can 

 infect a vertebrate by its bite. The means by which Triatoma becomes 

 infected indicates the most important phase in the developmental cycle 

 of Trypanosoma cruzi. The infection of the Triatoma larvae by contact 

 with the faeces of infected Hemiptera must be absolutely excluded. 

 The coprophagism noted by Brumpt in the case of Rhodnius jyrolixus, 

 does not occur in Triatoma megista or T. sordida. Experiments also 

 showed that in nature cannibalism only occurs between young larvae 

 of the same age. It was established that Triatoma spp. are only infected 

 by sucking infected vertebrates and that man and cats are the chief 

 reservoirs in the cottages. Details are given of an experiment showing 

 that of ten vertebrates, including five men and women, two cats and 

 three dogs — three only, a man, a woman and a cat, were capable of 

 infecting the insects after a single meal ; the man infected 40 per cent, 

 of the Triatoma fed on him once only, the woman 14 per cent., and the 

 cat not less than 66 per cent. One woman, though presenting all the 

 clinical characters of the disease, was incapable of infecting Triatoma. 

 According to Machado practically all the Triatoma in one cottage 

 might contain flagellates, while in another cottage they might be free, 

 although the inhabitants might present the same symptoms of the 

 disease. This anomal}^ is now expHcable if it is recollected that 

 vertebrates constitute the sole reservoir of the virus. Assuming that 

 in a cottage containing Triatoma harbouring flagellates and vertebrates 

 harbouring T. cruzi, the vertebrates may temporarily lose their 

 power to infect, then the Triatoma will become absolutely free from 

 flagellate parasites at the end of approximately 386 days, which is 

 the duration of their life. In some benign forms of the disease, such as 

 that in which affection of the thyroid gland is the most marked 

 symptom, the capacity for infecting Triatoma must be very much 

 reduced and the role of reservoir quite unimportant. 



Chagas recorded that Triatoma geniculata, Latr., living in the 

 burrows of armadillos, contained flagellates of Trypanosoma cruzi. 

 It has been found that not only Taius novemcinctus, L., but also 

 Dasypus sexcinctus and D. unicinctus, L., harbour Trypanosoma 

 cruzi, and in parts of the South American continent where the 

 disease has never been reported in man and even in uninhabited 



