It has been shown by references to the writings of other workers, and some personal observations in the early part of this paper, that destruction of certain species of bacteria takes place in the alimentary canal of flies—especially in the early developmental stages of the insect—and ticks. In addition, emphasis has been laid on the remarkable sterility of the gut-contents and faeces of some parasitic and semi-parasitic blood-feeders.

That these conditions result from the operation of several factors is possible, and the bactericidal principle described in this paper may reasonably be accepted as one of these factors.

The investigation has been limited to the adult stage of the insects Stomoxys calcitrans, Musca domestica, Anopheles bifurcatus, Aëdes cinereus, Cimex lectularius and Rhodnius prolixus, and the arachnids Argas persicus and Ornithodorus moubata. In all of these a bactericidal principle has been demonstrated in the gut-contents, and, in the case of the insects, also in the faeces.

The bacteria most affected were the common sporing aerobes and the staphylococci, but B. anthracis was destroyed by gut-contents of Stomoxys, and M. catarrhalis and S. haemolyticus by those of Argas.

It may be inferred, from the results of the bactericidal and absorption tests, that the active principle in any one species of arthropod is single, but that different groups of bacteria exhibit differences in susceptibility to its action; some, such as the subtilis group, being very susceptible, are destroyed by concentrations too weak to affect other species, while others, as the staphylococci, require relatively high concentrations for their destruction, and many bacteria appear to be wholly resistant.

The bactericidal principle occurring in different arthropod species showed some differences in the groups of bacteria susceptible to its action, and in the degree of susceptibility of particular groups. Thus, the widest range of action was shown by gut-contents of Argas and Stomoxys, and the least by those of the bugs. The sporing bacilli seemed to be affected more by material from Stomoxys than by that from Argas, but, on the other hand, the staphylococci were apparently more susceptible to the action of Argas material than to that of Stomoxys. The gut-contents of the ticks had a feebly destructive action on B. pestis, while those of some insects actually enhanced its growth. This last observation is of some interest in connection with the case of the plague flea.

With regard to the properties of the active principle, bactericidal action is greater and more rapid at 37° C. than at room temperature. This action is not accompanied by any visible bacteriolysis. The bactericidal principle retains its activity unimpaired for at least six months when kept in the dry state. It is very thermostable, resisting temperatures as high as 120° C. It is not destroyed by tryptic digestion. It is precipitated from solution with proteins by alcohol or acetone, but is not itself affected by these reagents. It is not soluble in the common fat-solvents, ether, chloroform, alcohol or acetone. By allowing it to act upon repeated small doses of bacteria, it rapidly becomes exhausted, and it can be inactivated, possibly through adsorption, by large doses of killed bacteria; even those species which are not destroyed by it. It may also be adsorbed in small amount by bibulous paper.

It exhibits none of the properties of the bacteriophage, and it differs from lysozyme.

In the bactericidal action it is clear that no vital process is involved, in fact it very closely resembles the action of an antiseptic.

Regarding the source of the active principle, there is no doubt that it is formed in the stomach, but whether as a secretion of the gastric cells or as a result of the processes of digestion is not clear. (Nuttall (1908) showed that the destruction of Spirochaeta duttoni in the gut of the bed bug was definitely related to digestion.) An examination of material collected from ticks, kept at 25° C. at various stages of digestion seemed to point to an increase in bactericidal activity during the first two weeks after feeding, and a decline after the sixth or eighth week, but the activity of material from individual ticks, taken at similar intervals after feeding, often varied greatly. Control tests with the blood of the hosts, on which the arthropods were fed, showed that this was not the source of the active principle.