212 Campbell: The embryo-sac of Pandanus 



were invariably present. The history of the chalazal nuclei is dif- 

 ferent. Figure id shows a stage with four chalazal nuclei, of 

 which three appear in the section shown. The cytoplasm has in- 



F 



creased noticeably in amount, and this becomes still more marked 

 in the later stages. Owing to the failure to secure division stages, 

 the sequence of nuclear divisions by which the number of the 

 chalazal nuclei is increased cannot be determined. Several cases, 

 one of which is shown in figure i i, show six chalazal nuclei, but 

 whether there is any rule as to which of the four nuclei of the pre- 

 ceding stage divide and which remain undivided must for the pres- 

 ent remain in doubt, and the same is true of the subsequent division 

 by which the number of chalazal nuclei increases to eight and 

 finally to twelve. There seem to be regularly twelve chalazal 

 nuclei in the older stages, although it is possible that in some 

 cases there may be only ten. 



There is more or less variation in the form of the lower end of 

 the embryo-sac. It is sometimes pointed (figs. 9, 12), or it may 

 be broad and rounded. Not infrequently one or more of the cells 

 of the nucellus push up into it, so that the lower end of the sac is 

 divided into short, sac-like extensions (fig. ii, a). 



In the most advanced stages that were secured, the embryo-sac 

 had increased materially in size. At the micropylar end, which is 

 narrower than the chalazal end, there are two nuclei, while at the 

 chalazal end there are twelve nuclei embedded in a large mass of 

 densely granular cytoplasm containing several conspicuous vacuoles 

 (figs. 14-16). A similar but smaller mass of cytoplasmic material 

 occupies the micropylar end, and the large central vacuole is 

 bounded laterally by a rather thick layer of cytoplasm, which, 

 however, contains no nuclei. In most cases observed there was 

 no apparent differentiation of the micropylar cytoplasm, but in a 

 very few instances, the most conspicuous of which is shown in 



b ■ 



FIGURE 14, there was a sHght indication of what looked like the 

 separation of an egg-cell and synergid, but this was very vague, 

 and the hmits of the egg, if such it was, were very poorly defined. 



this stage, the two cells lying above the embryo-sac are still 

 visible, but are evidently disintegrating. 



Corresponding to the enlargement of the embryo-sac is a 

 marked increase in the size of the nuclei, which at the same time 



At 



