THE FEMALE GAMETOPHYTE 111 



The protrusion of the sac bodily into or through the micro- 

 pyle may be regarded as only a more extensive development of 

 the vermiform mieropylar haustorium, but it deserves separate 

 mention. Torenia is the oldest and most conspicuous illustra- 

 tion of this phenomenon, the sac passing beyond the micropyle 

 and even reaching the wall of the ovary. The phenomenon 

 also occurs among the Rubiaceae, the sac entering the micro- 

 pyle in Diodia and the Galieae, while in Vaillantia the mega- 

 spore mother-cell passes into the micropyle and divides there. 



The projection of the synergid as an haustorium has been 

 observed by Billings 1 "" in Calendula lusitanica, in which the 

 synergid develops into the micropyle and enlarges greatly; and 

 in Trapella (Oliver 21 ), large, persistent synergids occur, which 

 are evidently haustorial. Other synergid haustoria have been 

 reported, as in Lobelia, but they prove to be merely haustoria 

 from the sac, containing endosperm. 



The antipodal cells are often very prominently associated 

 with the haustorial apparatus for obtaining nutritive supplies 

 from or through the chalazal region. The nutritive function of 

 the antipodals seems to have been claimed first by Wester- 

 maier 23 ' 30 in his studies of the prominent antipodals of the 

 Ranunculaceae. This was confirmed by Osterwalder 60 in his 

 study of Aconitum Napellus; and also by Mile. Goldflus 61 in 

 connection with the Compositae. The latest contribution to the 

 subject is that by Ikeda, 106 in connection with Tricyrtis hirta, 

 who claims that the antipodals in that species are nutritively 

 active from the full maturation of the sac to the formation of 

 endosperm, after which they change in structure and gradually 

 weaken; and that during that period they not only elaborate 

 food for endosperm-formation, but also for the growth of the 

 < ug-apparatus. The cutinization of the integument prevents 

 the passage of materials except by way of the chalaza, and 

 hence much of the nutrition must pass through the antipodals. 

 Ikeda describes and figures the position of starch, dextrine, 

 and cutinized membranes at various stages in the development 

 of the ovule and embryo (Fig. 52). From this point of view 

 antipodals are of two general types, that may be spoken of as 

 the passive and aggressive types. In the passive type the antip- 

 odal- remain active, often become very much enlarged (as 

 among Ranunculaceae), or even form a mass of tissue (as in 



