(WITH PLATES VI-VIII) OUR knowledge of the development of the embryo of Cycadales is very meager, being largely included in Treub's (5) account of Circas circinalis and Ikeno's (8) of Cycas revoluta. These two accounts agree in every important particular, and deal chiefly with the development of the proembryo. The following facts were established: the egg nucleus divides, and successive simultaneous divisions give rise to a large number of free nuclei; the cytoplasm of the central region of the egg becomes vacuolate and then completely disorganizes; the remaining cytoplasm becomes parietal, massing somewhat toward the bottom of the egg; in this parietal layer the numerous free nuclei are imbedded, being equidistant from one another and forming usually a single layer except at the base of the egg, where simultaneous nuclear division begins again; later, walls appear, and the proembryo becomes a sac somewhat thickened at base, but with the wall composed of one or at most two layers of cells. This history differed so much from that of Ginkgoales and of Coniferales as known at that time that it seemed to stand somewhat stiffly apart among gymnosperms. A series of collections of Zamia floridana was obtained from southern Florida during the spring and summer of 1902. Some collections consisted only of the ovulate strobili, but in most cases the entire plant was secured. As stated by Webber (II), the ovulate strobili continue in their development for some time after removal from the plant, and we have observed mitotic figures in ovules after the strobilus had been lying in the laboratory for nearly two weeks. Entire plants sent from Florida in June and July were potted, and they continued to develop