TIIE EMBRYO-SAC IN ANGIOSPEBMS. 521 



stages. Now, too, for the first time is evident a difference in the 

 appearance of one cell- of the axial row : the terminal cell of 

 this series has become slightly larger and more granular j while 

 its nucleus is sharply rounded, with an exceedingly bright 

 nucleolus (PL XVII. fig. 4). This cell is clearly becoming 

 distinct from its neighbours ; and we shall have no great difficulty 

 in demonstrating that it produces the embryo-sac as a daughter 

 cell or cells : we may accept the term " embryo-sac mother cell " 

 for it*. 



As a rule, this upward direction of the whole ovule also indi- 

 cates the first bending on itself of the normally anatropous ovule; 

 but there occur ovules which attain maturity, and have otherwise 

 a normal structure, which, however, never curve at all, but the 

 nucellusf with its integument sits in an orthotropous manner on 

 the end of a long funiculus projecting perpendicularly from the 



placenta. 



From the point where the curvature may be said to begin, and 

 which marks junction of funiculus and nucellus, the " axial row " 

 soon ceases to be distinct ; and the cells, by elongation and divi- 

 sion, later help to form the feeble fibro- vascular bundle of the funi- 

 culus ; above this point, however, they persist to the end as a series 

 of a few cells, and may be recognized even in fertilized ovules. 

 The importance attached to this will be shown hereafter. 



Soon after the mother cell of the embryo-sac becomes recogni- 

 zable by its nucleus and contents, active division is set up in the 

 cells around the axial row, so that, by walls parallel to the periphery 

 of the whole organ, there is soon cut out, as it were, a second sub- 

 epidermal layer or hollow cylinder (PI. XVII. figs. 4-8). The 

 embryo-sac mother cell has now become much longer, and in the 

 shape of an inverted truncated cone, resting on two or three cells 

 of the axial row ; while the inner integument, growing up around 

 the nucellus, is soon followed by a second integument, which arises 

 at its base by divisions similar to those with which this started 

 (PI. XVII. figs. 5 & 7, and PI. XVIII. fig. 8). We have now 

 the typical anatropous ovule, since the funiculus becomes closely 

 applied to one side from the first sharp bend. 



See Warming, " De rOvule," and Vesque, " Developpement du sac embry- 



onnaire," Ann. des Sc. Nat. 1878. 



t The word 7incellus is employed throughout this paper to mean the nucleus 

 of the ovide, in the same sense as that phrase is used by English writers. 



