An account is given of the bacterial flora isolated from the blood-sucking Reduviid bugs, Triatoma infestans and T. protracta, and the predatory Reduviid, Sinea diadema. The micro-organisms which have been found in Triatoma spp. differ from that found in Rhodnius prolixus (Actinomyces rhodnii), in both specific identity and in their relation to the cells of the mid-gut. The symbiont occurring most frequently in T. infestans has been identified as a diphtheroid organism resembling Coryne-bacterium hofmannii, and sections show this organism inside vacuoles within some of the endothelial cells of the insect's intestine. The organism occurring in R. prolixus, on the other hand, cannot be found in this situation, but remains in the lumen of the gut. While the occurrence of the bacteria found in T. infestans in so intimate a connexion with the cells of its host seems to suggest a more highly organized state of affairs, T. infestans does not show the fixed association with one species of micro-organism which is shown by R. prolixus. Other bacteria occur commonly in T. infestans, although they do not occur in cell-vacuoles; and the range of bacterial types is so great that it seems unlikely that the species recorded in this paper are the only ones that may occur. The presence of these other bacteria may be due either to the anti-biotic action of the Corynebacterium sp. being weaker than that of A. rhodnii, or to masking of the anti-biotic action by the cell wall that encloses the organisms. The situation in R. prolixus shows clearly that a constant symbiotic association with a micro-organism can be maintained without the presence of specialized organs of the mycetome type.

With regard to the predatory species, no general conclusion can be reached on the evidence so far available. Compared with the blood-sucking Reduviids, insects of this kind usually have powerful proteolytic enzymes secreted in the anterior mid-gut, while the same region of the blood-suckers is a simple storage and concentration sac, digestion taking place lower down the intestine. Thus the region of the gut where bacteria are found in blood-suckers might not be a favourable situation for them in predatory bugs. Hence it is perhaps significant that most of the few successful isolations from a predatory type were from the rectum of the insect. It is the usual opinion that insects with a predatory diet do not need bacterial symbionts, but the aim in examining the predatory Reduviidae has been to find out whether they show a condition which might have evolved into that found in their blood-sucking relatives. The search is still in progress for a rather unspecialized type of predatory Reduviid in which the salivary and anterior mid-gut digestion is less strongly developed than in those species which have been available for study up to the present.