The question whether caffein is cumulative in the rabbit, suggested in the preceding experiments, was the subject of further investigation in Series B. Caffein was given by mouth or subcutaneously. Carrots formed the exclusive diet, a measured amount being given. The rabbits were kept under observation for two weeks, except Nos. 370 and 373, records of which were made only for four days before the administration of caffein was begun. Caffein was given by mouth in experiments of Groups I and III. Rabbits 292, 293, and 295 received daily 20 cc water by mouth for four days previous to the administration of caffein, while in the rabbits of Group II the caffein treatment was preceded by the injection of 0.8 per cent salt solution subcutaneously. The object in both cases was to ascertain whether or not the method of the administration of caffein has any influence on the animal, but observation made from day to day failed to show any effect of such treatment. About 1 decigram of caffein per kilo was administered daily, with occasional intermissions. Later in the course of the experiment the doses were increased, 0.15 gram per kilo being the maximum dose given. Rabbit 293 died after the third dose with symptoms of typical caffein poisoning. The administration of the same dose of caffein was continued 10 days longer in Nos. 292 and 295. It was omitted on the seventh, fourteenth, and fifteenth days of the experiment. On the eighteenth day of the experiment the dose was increased to 150 mg per kilo and was repeated 2 days later. No. 295 was found dead the next day. No. 292 survived. Rabbits 313 and 315 may be considered together, as they were treated alike in every respect. The initial dose of 100 mg caffein per kilo was finally increased to 122 mg. After the twelfth dose the emaciation was well marked and the rabbits were very weak. No. 313 was found dead 2 days, and No. 315 3 days, after the last dose of caffein was given. It should be remarked in this connection that symptoms of caffein poisoning were never observed in these rabbits. Death was not due, therefore, primarily to caffein, but the rapid loss of flesh and strength observed during the last few days suggests that it was due to malnutrition apparently brought about by caffein.