442 BOTANICAL GAZETTE [December 



embryo in development (figs. 6, 7). These cases of budding of 

 embryos represent a process which, according to Buchholz (2), does 

 not occur in the Abietineae. Also, although the secondary sus- 

 pensors are often closely associated, no case was observed in which 

 they fused to form a single embryo, a process described by Buch- 

 holz as normal in the higher Abietineae, but absent in the pine. 

 From the material examined, it is concluded that in G. sp. 15 and 

 G. sp. 59, as in other species of the genus as well as in Ephedra, 

 (Land 5), although a number of embryos may begin to develop, 



only one reaches maturity. 



Discussion 



In 



more 



such species as G. Gnetnon, in which the suspensors are widely sepa- 



m. Furthermore 



in G. Gnemon the embryo develops directly on the end of the 

 primary suspensor, in the other species studied a long multicellular 

 secondary suspensor is produced, corresponding, although different 



form 



In all 



rvelopment of the external form, a 

 mbryo proper, is similar and resembl 



for Welwitschia (4). 



In certain features of the gametophyte and endosperm 

 Thompson has shown that G. Gnemon is quite different from the 



losperms 



It is of 



course also distinct in its arboreal habit, for the others are all vines. 



from 



species in the development of its embryo. No loi 

 suspensor is formed, and the primary suspensors ramify widely 

 through the endosperm. The reduction in the suspensors appears 

 to bring it nearer the angiosperms. The great reduction in the 

 amount of free nuclear division is a character which separates 

 Gnetum widely from the lower gymnosperms. Coulter and Bower 

 differ as to whether there is any free nuclear division in the forma- 

 tion of the embryo proper a* the end of the primary suspensor. 



