﻿MIC HELL— RICH A RDIA A FRICA NA 



333 



1913, without reference to the later paper, confirmed Campbell's 



In 191 2 Campbell (4) published the results of his researches 

 on A. simplex and A. modestum. He believes that there are only 

 5 nuclei in the embryo sac in these species, this stage arising from 

 the further division of one of the micropylar nuclei in the 4-nucleate 

 stage. The obstacle in the way of accepting Campbell's view is 

 that he has never been able to demonstrate the supposed nuclear 

 division. My investigations on the embryo sac of Richardia, 

 which about the time of maturity has every appearance of being 

 5-nucleate, have led me to doubt the division of one only of the 

 4 nuclei. 



In many cases the antipodals degenerate almost as soon as 

 formed, and are indistinguishable from the nucellar cells, which 

 also show signs of degeneration. However, it has been possible 

 to show that in Richardia, at least, the embryo sac is at one stage 

 8-nucleate; and in the absence of definite proof to the contrary 

 one cannot but feel that the same may be true for Aglaonema. In 

 the fourth species A. pictum, Campbell records the presence of 

 cells which look like antipodals, although in the light of his work 

 on A. modestum and A. simplex, he is inclined to believe that they 

 are of nucellar origin. 



In a comparison of Richardia with the other genera of the 

 Araceae, mention must be made of Spathicarpa, the development 

 of which has been studied by Campbell (2). At the time when the 

 embryo sac is filled with endosperm, it bears a striking resemblance 

 to that of Richardia, but in Spathicarpa Campbell derives the large 

 cells at the base of the endosperm from the antipodals, which up 

 to the time of fertilization are inconspicuous. In Richardia this 

 is not the case, the antipodals being evanescent in character. 



1. The ovary of Richardia africana is usually trilocular and has 

 axile placentation. Four ovules are borne in each loculus. 



2. The ovule is not very decidedly anatropous and has two 

 integuments. 



