210 Campbell: The embryo-sac of Pandanus 



Development of the embryo-sac in Pandanus Artocarpus 



I 



The earliest stages of the embryo -sac found in P. Artocarpus (figs. 



% m 



4—6) were somewhat more advanced than that described for P. 

 affinis. The mother-cell is more nearly cylindrical and the base 

 is truncate and sometimes almost as broad as the apex, and in 

 such cases it forms the lowermost of a single series of cells, or at 

 least such is the appearance in a longitudinal section, (See fig. 

 6,) More frequently the upper end of the mother-cell is broader 

 and two series of parietal cells appear in longitudinal sections. 

 These two series probably arise from a median division of the 

 tapetal or ''parietal" cell, and by subsequent periclinal divisions 

 give rise to about three layers of cells between the mother-cell 

 and the epidermal layer at the apex of the nucellus. The sporog- 

 enous cell divides by transverse division into two cells, of which 

 the lower is the larger and becomes the embryo-sac. The upper 

 cell, x^ divides again into two by a vertical wall and these two 

 small cells persist with little change for a long time, but finally 

 disintegrate and are visible only as two small, darkly stained, 

 shrunken bodies lying above the apex of the embryo-sac. In 

 these early stages Pandanus conforms closely to the ordinary 

 angiosperms, except for the vertical division in the upper sporog- 

 enous cell, and differs from Peperomia and Gunnera in the for- 

 mation of three cells which may be interpreted as megaspores, 

 instead of having the sporogenous cell develop directly into the 

 embryo-sac without preliminary division. 



4 



Unfortunately no nuclear divisions were found, although these 

 were carefully looked for, and this was true also for the later stag^ 

 in the development of the sac. This was perhaps due to the fact 

 that the material was all collected at about the same time of the 

 day. The collectors brought the material in usually at about 

 eight o'clock in the morning, and apparently the nuclear divisions 

 occur at some other period of the day. 



It is probable, as in other forms that have been critically 

 studied, that the first reduction division occurs when the spore 

 mother-cell first divides. 



The youngest stage met with in P, Artocarpus is showai in 

 figure 6. The young embryo-sac is easily recognizable, the 

 cytoplasm being noticeably more densely granular than that of the 



