I 



Campbell: The embryo-sac of Pandanus 219 



to conjecture until the further history is known. For the present 

 the embryo-sac of Pandanus must be assumed to represent a new 

 type with its nearest analogue in Pepero7nia^ from which it differs 

 in its marked polarity in its early stages, and the apparent absence 

 of any nuclear fusions, as well as in the fact that there are only 

 fourteen instead of sixteen free nuclei. It probably represents a 

 less specialized and more ancient type of sac than the typical eight- 

 nucleate OWQ. 



The recent work of Porsch on the embryo-sac has not been seen 

 by the writer, and so it will not be possible to discuss here the 

 views of that writer as to the homologies of the parts oi the em- 

 bryo-sac. From the references to his work by other investigators, 

 it seems that he regards the egg-apparatus and antipodal cells of 

 the ordinary embryo-sac as representing two archegonia. The 

 condition of things in Pandanus would certainly lend no support 

 to such a view. The accumulating discoveries of plants in which 

 the double, or approximately double number of nuclei is normally 

 present m the embryo-sac, makes it more and more likely that, as 



Ernst maintains, this really is a distinct type of sac not derived from 

 the eight-nucleate one, but an independent and presumably more 

 primitive type. It is to be expected that further researches will 

 add to the number of these sixteen-nucleate sacs and may throw 

 more light upon the characteristic eight-nucleate sac which distin- 

 guishes the great majority of angiosperms. 



Stanford University, 

 California- 



Kjxplanation of plates 16 and 17 



Plate i6 



.# 



Griflf. Except Rg^. I, 2, the drawings were all made with a Leitz ^ oil immersion, 

 oc. I. The drawings are reduced about one fourth in the reproduction. 



Fig. I. A single pistillate flower o{ Pandanus Artocarpus^ X4/^ \ ^^ the ovule. 



Fig. 2. Section of a young ovule, X^^^^^ 4^ i ^» ^^ embryo-sac. 



afL 



the sporogenous cell is still undivided ; the nuclei of the parietal cells are shown. 



Fig. 4. Young embryo-sac, w, and the sister-cell, x. 



Fig. 5. Upper part of the nucellus showing the young embryo-sac, w, the sister- 

 cells, x^ and the parietal cells lying above. 



Fig. 6. A somewhat younger stage than the last ; only a single row of parietal 



cells. 



