218 Nichols, A morphological study of Jumperus communis var. depressa. 



in fig. 1. The integument is well defined, and the nucellus is 

 composed of a mass of cells among which a slight differentiation 

 is already apparent. Those near the tip have ceased to divide 

 and have become vacuolate. Before long their growth stops entirely 

 and their walls become slightly thickened. In the basal portion 

 of the nucellus, in marked contrast to the apical region, the cells 

 continue to divide by periclinal walls, giving rise to longitudinal 

 rows of prismatic cells (fig. 59). At the lower end of these rows 

 several cells soon become prominent by reason of their large size, 

 big nuclei, and dense cytoplasm. This group constitutes the 

 archesporium, which it will thus be seen is organized over twelve 

 months before the formation of the prothallium. During the summer 

 the archesporium increases somewhat in size, yet at no time is 

 there a sharp line of demarcation between the sporogenous cells 

 and the other cells of the nucellus. Certain cells of the arche- 

 sporium take the lead in growth, but it is impossible to State with 



Fig. 1. 



Longitudinal section through young ovule immediately before pollination. 

 Micropyle open. X 66. 



certainty which of these will become the spore mother cell. The 

 vegetative cells surrounding the archesporium, as noted by Noren 

 (1907), become somewhat flattened and form inore or less concentric 

 layers. It is of interest to note that in Cephalotaxas (Coker 1907), 

 where pollination likewise takes place the year preceding fertiliza- 

 tion, the development of the female gametophyte proceeds at the 

 same rate as in J. communis, while in Pinus (Ferguson 1904), 

 where similar conditions are found, the embryo sac Starts to devel- 

 op the first season and passes the winter in the thirty-two nucleate 

 stage. 



Tetrad division. — The writer has made no attempt to 

 obtain a complete series of the various stages in spore formation, 

 but they are doubtless similar to those described by Noren (1907). 

 The megaspore mother cell can be positively identified for the first 

 time early in April, when its nucleus enters synapsis (fig. 60). Dur- 

 ing this period the nucleus presents an appearance very similar to 

 that described in the nuclei of the microspore mother cells, the 



