The more important points that have emerged in the course of our investigations may now be assembled.

(1) An adhesion phenomenon has been described in which, in certain circumstances, red blood corpuscles adhere to trypanosomes.

(2) We believe this red-cell adhesion to be a manifestation of the adhesion phenomenon already described by other workers both with trypanosomes and spirochaetes.

(3) The substance responsible for this phenomenon of adhesion is designated in this paper adhesin.

(4) The red-cell adhesion test has been studied in relation to several strains of polymorphic trypanosomes, all of which were either recently isolated from naturally infected man or animals, or had been recently passed cyclically through tsetse.

(5) It is shown that the adhesion phenomenon is irregular and uncertain in its appearance. A single or even a number of negative observations do not exclude trypanosomiasis. A positive reaction, on the other hand, indicates recent or actual infection with a trypanosome of the same group, but not necessarily specifically identical with that employed as antigen in the test. Thus T. gambiense adhesin bloods not uncommonly react with T. rhodesiense trypanosomes and also on occasion with the Damba strain: the converse is apparently of less common occurrence.

(6) Adhesins appear in an animal's blood in the course of its infection with trypanosomes, whether the animal be treated with trypanocidal drugs or not. When sterilising doses of such drugs are employed the adhesins gradually disappear in the course of a few months; but in untreated infections of T. gambiense they have been shown to persist in an active form for more than two years.

(7) With our material it was found impossible to distinguish between “passage” and “relapse” strains by means of the adhesion test. On the contrary it was found, at all events with T. gambiense, that the best way to call forth the reaction in the blood of animals long infected is to employ trypanosomes from a very early infection.

(8) Guinea-pigs have been found to be unsatisfactory animals in which to demonstrate adhesins; they are, however, useful as a source of sensitive trypanosomes. When only guinea-pig elements are present in a test a positive reaction is indicated by platelet not by red-cell adhesion.

(9) The presence of adhesins in the blood has no apparent relation to immunity against the homologous trypanosome.

(10) The test has been applied to over 200 human cases of trypanosomiasis in the Uganda Protectorate, and with six of these repeated observations were made. Some of these human cases reacted both to T. gambiense and T. rhodesiense, but none to T. rhodesiense alone.

(11) A partial analysis has been made of the mechanism of red-cell adhesion, which has been shown to be in some way associated with the presence of the red corpuscles of primates. This fact probably explains why other workers appear not to have noticed red-cell adhesion in the extreme form we have described in this paper.