Summary: Studies of the pollen antigens have resulted in the following conclusions: Timothy and ragweed pollens have definite antigenic properties and may induce in the uinea-pig a condition of hyper-sensitiveness; intravenous injection of the specific antigen in such animals can result in death by acute or delayed anaphylactic shock.With the Dale and intravenous testing methods as criteria, comparison of the states of sensitivity induced by injections of the pollen extracts and by “border zone” doses of an immune egg-white serum indicates that: The pollen antigen sensitizes a much smaller number of animals.The pollen antigen induces a state of maximal or acute sensitivity in a smaller percentage of animals.The contraction of the pollen-sensitive uterus tends to be slower in starting, and it is more irregular than is the contraction of the egg-white sensitive uterus.Pollen sensitivity is specific, except against related pollens of the same family.Passive transfer of sensitivity was accomplished in only one animal with only partial success, the sensitivity induced being detected by the Dale method, but not by the intravenous test.The pollen antigen-antibody reaction is much more time consuming than that of egg-white, both in vivo and in vitro.The presence of precipitins has not been demonstrated in the sera of any of the pollen sensitive animals.The rules of Coca and Kosakai for precipitin neutralization are not applicable to the pollen antigen-antibody reaction.Digestion with trypsin and dialysis does not remove the active antigenic principle of ragweed pollen.The non-digestible, non-dialyzable factor is common to both high and low ragweed pollens, and is almost as strongly antigenic as the whole pollen. (This factor seems, thus, to be the same principle which Coca and Grove found to be the atopen in ragweed hay fever.)Beside the non-digestible and non-dialyzable factor, there are one or more less important antigens in ragweed, probably protein in nature.