\ 



s 



104 THE SCOTTISH BOTANICAL REVIEW 



near the base of the nucellus. This second embryo-sac (?) 

 had two nuclei, and some of the cells in it also looked as if 

 they might be of sporogenous nature. 



It will thus be seen that the genus Aglaonema exhibits a 

 remarkable degree of variation in the early history of the 

 embryo-sac, and this is also true of the later stages. 



Development of the Embryo-sac in 

 A. SIMPLEX and a. modestum. 



A. simplex resembles closely in appearance A. commutatum 

 and the species collected in Jamaica under the name Dieffen- 

 bachia Aglaonema. As the latter very closely resembles A. 

 simplex in the development of the embryo-sac, it may perhaps 

 have been this species rather than A. commutatum. 



The young embryo-sac in A. simplex grows rapidly in size, 

 and the primary nucleus divides into two, which may move 

 ' to the two ends of the sac, or may remain close together at 



almost any point in the sac. At this time (PI. I. fig. 7) the 

 embryo-sac has already almost obliterated the lateral tissue 

 of the nucellus, which undergoes no further growth but is 

 gradually destroyed by the growing embryo-sac. The speci- 

 men shown was badly shrunken and the nuclei did not 

 exhibit their normal form. A cross-section of a sac of about 

 the same age is shown in fig. 8, where the two nuclei occupy 

 a position near the middle of the sac. Two sections of a sac 

 of about the same age of A. commutatum are shown in fig. 9. 

 The nuclei are decidedly larger in this species, and in the 

 case figured the two nuclei were at opposite ends of the sac. 

 Fig. 10 shows a section of an embryo-sac of A. simplex in 

 which the two nuclei, still close together, were at the base of 

 the sac. 



No specimens with four resting nuclei were found in A. 

 simplex, but one specimen was found where the two nuclei 

 were dividing. These were near the centre of the sac and 

 were in the early metaphases of division. The number of 

 chromosomes was very large but could not be satisfactorily 

 counted. To judge from the next stage it is quite clear that 

 the division results in four nuclei, probably in most cases 

 arranged in pairs at opposite ends of the sac. Figs. 15 to 17 



