G YMNOSPERMAE 



Eocene they had shrunk to their present size. They begin about the 

 Permian period, about which time the great class of the Pterido- 

 spermae, or seed- bearing ferns (the bulk of what used to be regarded 

 as true ferns in the coal-measures, &c. ), was disappearing, and show 

 considerable relationship to the latter. Some of the fossil forms have 

 actual ? fls. with numerous sta., and it is quite possible that these are 

 on the direct line of ascent to the Angiosperms. From what the 

 Coniferae and Ginkgoales, which are first found about the beginning 

 of the Permian, took their rise, is as yet quite uncertain, while we do 

 not know enough about the Gnetaceae to dra\v many conclusions about 

 them. 



The fls. in most Cycads and Conifers take the form of cones, and 

 whether each cone represents a fl. or an infl. is still a disputed point. 

 The sta. is of simple structure ; in the Cycads there are several 

 pollen-sacs, looking like the sporangia of Marattiaceae, on the lower 

 side of a leaf-like organ; in the Conifers the sta. has usually fewer 

 pollen-sacs and is more leaf-like, while in the Gnetaceae the anthers 

 are sessile. The ovules are always naked in the sense of not being 

 enclosed in an ovary formed of one or more hollow cpls., but they are 

 usually protected in some way from the weather. Wind-pollination 



occurs. 



In the Cycads a considerable mass of sporogenous tissue is formed 

 in the ovule (mega-sporangium); one of the cells of this tissue gives 

 rise to the embryo-sac (mega-spore). This behaviour is closely com- 

 parable to that of the higher Pteridophyta. In the embryo-sac the 

 ? prothallus (endosperm) forms by cell-division, and archegonia in 

 which the ova are contained develope at the micropylar end. The 

 ovule is now mature and consists of an integument, nucellus, and 

 embryo-sac and its contents. 



In most Conifers the sporogenous tissue consists only of the cell 

 which goes to form the embryo-sac. In the sac the same process goes 

 on as in Cycads. 



In Ephedra the phenomena are very similar to those in Conifers ; 

 in Gnetum several embryo-sacs are frequently formed, and the division 

 of the nucleus of the sac gives rise, not to a prothallus as in the cases 

 above described, but to a number of free nuclei lying on the wall of 

 the sac. Those in the upper part remain free, and one or more 

 of them being fertilised, produce pro-embryos. The lower part of 

 the sac becomes septate into multinuclear compartments (in some sp.), 

 which become uninucleate cells of the primary endosp. as a result of 

 nuclear fusion ; in these sp. the primary endosp. is constituted before 

 fert. In Welwitschia it is constituted in a similar way. 



The whole question of the relationships of the G. is one of much 

 difficulty. (See Angiospermae, Chalazogamae, Pteridophyta, Cyca- 

 daceae, Coniferae, Ginkgo, and refer also to Coulter and Chamberlain, 

 Morphology of Gymnosperms, Campbell's Mosses and Ferns, Nawaschin 

 in Mem. de I'Acad. des sc. de St-Pttersbourg, XLII. 1894 (reviewed in 

 Bot. Centr. 62, p. 324), and other papers referred to in the articles 

 quoted.) 



Gymnosporia Benth. et Hook. f. Celastraceae. 80 trop. and subtrop., 

 esp. Afr. Many have branches modified into thorns. 



