io6 BOTANICAL GAZETTE. 



4. The variations which I have observed in the different types of 

 Angiosperms depend on the arrest of development more or less early 

 which seizes the special mother cells. 



5. The first cell always forms the sexual preparation. It 

 blends itself with the second cell to thus constitute the greater part of 

 the embryo-sac. When the second cell produces a ^'tdrade'' the eight 

 nuclei freed from the embryo sac act as M. Strasburger describes it in 

 Orchis and in Monot'Opa. This fact is observed in certain Monocoty- 

 ledons and dialypetalous Dicotyledons. 



6. The other special mother cells (3, 4, 5) may produce some 

 '•'■tetradesy Each one of the vesicles is homologous with a grain of 

 pollen, and it is tempting to give to it the name of antipodal. When 

 these mother cells persist in their primitive conditi'-n without produc- 

 incr '■'■tdradcsr they themselves simulate antipodal vesicles superposed, 

 not juxtaposed. They differ from them from a morphological point 

 of view and 1 have given to them the name of anticlinal cells. 



This condition has been observed in many Monocotyledons, certain 

 dialypetalous Dicotyledons, and in almost all the (iamopetalte. 



7. The 2d cell appears to underg > at first an arrest of develop- 

 ment. In this case, its nucleus becomes directly the nucleus proper 

 of the embryo-sac, and this cell does not produce any antipodal vesi- 

 cle. This fact, observed in some Monocotyledons and Dialypetaht, 

 becomes the rule in Gamopetalffi, which are, from this point of view, 

 the plants most removed from Cryptogams. 



8. In the Camopetalae (with very few exceptions), cell one alone 

 produces a *'tei?'ade" complete or incomplete, which is no oiher 

 thing than the sexuai preparation composed of two or three or four 



vesicles. 



The second cell seems to perform the vegetative part of the em- 

 bryo-sac. Its undivided nucleus becomes the nucleus of the embryo- 

 sac. 



The cells 3, 4, 5 (or 3, or 3 and 4, according to the number of 

 the special mother cells) are some anticlinal, or produce the antipodal 

 vesicles by dividing their nucleus. 



0. In the greater part of Gamoj^etaUt, the formation of the en- 

 d isperm is deferred to subsequent development, by division, of 

 one or several of the special mother cells. These last being homo 

 logons with the mother ct\U of spores, it is legitimate to consider the 

 endosperm of these plants as a sterile female prothallus. 



Botanical Contributions, by Asa Gray. Issued September 



J 1880. These annual contributions to Noith American Botany are 



always greeted with the greatest interest by systematic botanists, and 

 the p>iges can hardly be cut and run through hastily enough to satisfy 

 tiieir eager curiosity. What new species and genera have come into 

 the world and what have departed this life, are questions that first oc- 

 cur. This paper is largely devoted to recording some of the results 

 of Dr Gray's elaboration of the vast order Compositse for his Synop- 

 tical Flora. As this portion of the Flora cannot be published at once, 

 botanists are under very great obligations for some of the "ad- 



