﻿192 BOTANICAL GAZETTE [march 



single free nuclear division, resulting in two free nuclei; and in 

 Gnetum and Tumboa no free nuclear division occurs at all. These 

 last-named forms have reached the condition of the embryogeny 

 of angiosperms, in which the first division of the nucleus of 

 the ^^^ is accompanied by the formation of a wall. The gap 

 between Zamia with its 256 free nuclei and T_axus with its 32 

 may seem great at first sight ; but it should be noted that it 

 represents the omission of only two successive divisions, the 

 sixth and seventh. It is entirely possible, therefore, that these 

 missing numbers (64 and 128) may be found among the numer- 

 ous Cycadales and Coniferales that await investigation. 



In the second series shown in the diagram a later stage is 

 presented. In this series Ginkgo is clearly at one extremity 

 with its numerous free nuclei equally distributed, and its wall 

 formation resulting in a proembryo that completely fills the ^g%^ 

 In Cycas the free nuclei do not remain evenly distributed, but 

 there is a massing toward the base of the ^^%, and most of the 

 remaining nuclei through the formation of a central vacuole pass 

 to the periphery and enter into the formation of a parietal layer. 

 In Zamia wall-formation is still further restricted, appearing only 

 in connection with the basal nuclei. There is also no develop- 

 ment of a central vacuole, and hence no parietal placing of free 

 nuclei. In Zamia, therefore, a tissue that fills the sac, as in 

 Ginkgo, or that lines the egg, as in Cycas. is reduced, and its 

 formation is restricted to the basal region, as in Coniferales. In 

 passing from this condition, represented by Zamia, to the con- 

 dition displayed by the Coniferales, there is a gradual reduction 

 in the number of nuclei that enter into the formation of this 

 basal tissue, and as a result a gradually more definite setting 

 apart of the function of each particular nucleus. For example, 

 in Zamia the suspensor cells are not recognizable until they 

 begin to elongate; in Taxus with its 32 nuclei, and in Cephalo- 



r 



taxus, Ppdocarpus, and T^xodiun with their 16, there is no set- 

 ting apart of a distinct suspensor-forming layer; while in Thuja 

 with its 8 nuclei a distinct suspensor-forming layer is recognized, 



and this definiteness is continued into the 4-nucleate forms. 



It would seem to follow from these considerations that the 



