1.The working hypothesis that under special conditions dengue might provoke a certain immunity to yellow fever is discussed. This hypothesis is considered to be as yet unproved. The known facts, in any event, testify against a humoral immunity.
2.A more sensitive adjustment was devised in Theiler's as well as in Sawyer's mouse-protection test in order to avoid the risk of missing moderate and lower degrees of immunity. The modifications applied to the technique are described (in Theiler's test a 1 in 10,000 filtered brain-emulsion was used; in Sawyer's a centrifuged emulsion of about 2.5 per cent concentration).
3.As a control for the technique, 5 sera of former yellow fever patients were examined. By Theiler's method a distinct protection could be demonstrated in all cases; by Sawyer's method protection was evident in 3 cases (in 1 case the test was not applied, in the other it failed, repetition not being possible).
4.The examination of the sera of 5 assistants working more or less regularly in the yellow fever laboratory revealed a weak protective action in 2 of them, both by Theiler's method (emulsion, 1 in 10,000) and by the modification of Sawyer's test. Considering that these two individuals were the persons who had come into the most intimate contact with the yellow fever material, it is believed that they must have had mild (abortive) attacks of yellow fever at some time in the past.
5.The sera of 20 volunteers, who had been experimentally inoculated with virus of Sumatran and Javanese dengue, were tested by both methods; in several instances the tests were repeated. Despite the sensitive adjustment of the tests, a protective action against yellow fever virus could not be demonstrated.
6.The immunity in dengue and in yellow fever are compared, and the remarkable differences pointed out. In dengue the immunity shows considerable individual variation with regard to intensity and duration, and specific antibodies cannot be demonstrated in the serum. In yellow fever the immunity is almost absolute and antibodies are almost always present in the serum in high concentration.