INVEBTEBRATA, CRYPTOGAMIA, MICROSCOPY, ETC. 907 



The processes in tlie embryo-sac of Gymnosperms also commence with 

 a division of the nucleus ; after which appears the first difference, 

 namely, that the septation continues, instead of ceasing when four 

 nuclei have been formed in each end of the embryo-sac. But after 

 impregnation the formation of endosperm begins afresh by cell- 

 division or division of the nucleus. Strasburger is inclined to regard 

 the cells of the germinal apparatus and the antijiodal cells as 

 endosperm-cells ; the germinal cell appears to him to be a very 

 reduced archegonium. 



With regard to the relationship of Gymnosperms to Vascular 

 Cryptogams, Strasburger is disposed to look on the former as directly 

 descended from the Lycopodiaceae. The genetic connection of Angio- 

 sperms with Gymnosperms he considers much more doubtful ; but is 

 unable to regard the Gnetacete as the ancestors of the former. 



Investigations on the process of fertilization, especially in Juni- 

 perus virginiana, resulted in the detection of a canal-cell, previously 

 overlooked in the Juniperaceae, originating only very late, at the com- 

 mencement of impregnation, and very soon becoming disorganized. 



The formation of the embryo in Ceplialotaxus and Araiicaria 

 presents several remarkable peculiarities. Among others, the apex 

 of the rudiment of the embryo consists of large cells containing but 

 little j)rotoplasm, which form a kind of protective apparatus, and are 

 afterwards thrown off. The apex of the stem is therefore formed 

 from inner cells. 



Nucleus of the Embryo-sac* — In his ' Cell-formation and Cell- 

 division,' Strasburger states, as a universal rule, that no fresh forma- 

 tion of nuclei takes place in the embryo-sac ; all the nuclei proceed- 

 ing from one another by division. According to Darapsky, an 

 exception to this rule occurs in the case of Hi/acintl us cuiatus, where 

 the large size and great clearness of the nucleus in the embryo-sac 

 when in a state for fertilization present favoui-able conditions for 

 observation. During the development of the endosperm in this 

 l)lant, the fresh formation of nuclei for the cells of which it is 

 ultimately composed can be readily observed. 



Gynmospermy of Conifers, f — A contribution to the literature of 

 this subject is furnished by Celakovsky, in a description of a pro- 

 liferous fir-cone carefully examined by him. 



The rachis of the cone was produced into a terminal shoot, the 

 interuodes of which were moderately elongated, and the scales much 

 smaller than in the ordinary cones, and exhibiting transitions to 

 axillary buds. Near the apex was a dense rosette of large, leafy 

 buds arranged in several spiral rows. Above this were several 

 whorls of sterile acicular leaves, and finally a leafy spiral terminal 

 bud consisting of ciliated bracts. With respect to the position of the 

 ovules on the fertile scale or on the two bracts of the bud, the 

 abnormal structures show beyond a doubt that they spring from the 

 back or under side of the bracts; and a careful examination proved 

 that the origin of the ovule was clearly in the channel-like depression 



* 'Bot. Zeit.," xxxvii. (1S79) p. 553. f ' Flora,' Ixii. (1879) p. 257. 



