1913] VYORK—DENDROPHTHORA 105 
further development of the gametophyte, and by the completion 
of the sac they have usually been entirely consumed. The sac 
toward the base of the inflorescence as a rule contains a greater 
amount of stored food substances than its mate, and is the one in 
which degeneration of the nuclei occurs. No examples were 
observed where the nuclei in both sacs had begun to disintegrate. 
Following this stage, the end of the sac grows almost straight 
downward in the floral axis until it extends below the level of the 
insertion of the placenta. It then curves outward into the tissue 
of the carpel, bends, and grows upward beneath the epidermis of 
the inner wall of the carpel until the micropylar end of the sac lies 
almost over the apex of the floral axis (figs. 29, 30). At first it 
forms a very narrow tube, having a diameter but slightly greater 
than the width of the cells adjacent to it. It forces its passage 
through the tissues of the floral axis and the carpel by digesting 
the cells in front of it. The behavior of the sac suggests strongly 
that of a pollen tube. GrirritH (11), in his observations on the 
Santalaceae, figured and described the branching of the chalazal 
portion of the embryo sac of Santalum and mentioned that it 
behaved much like a pollen tube. Lroyp (18) calls attention to 
the pollen-tubelike behavior of the embryo sac in his studies of 
the Rubiaceae. The disintegration of the cells indicates clearly 
the secretion of an enzyme by the tubelike sac. A portion of the 
protoplasm in this sac seems to be specialized for this purpose, as 
will be shown later. As the sac advances, it evidently derives its 
nutriment from the cells immediately surrounding it. In a sense 
it is a parasite within a parasite, feeding on the tissue through 
which it moves. The path of growth of the sac brings it into a 
most advantageous position for obtaining food. It also provides 
a line of transit through which food is readily transferred to the 
developing endosperm and embryo. When the sac grows down- 
ward, it comes into direct contact with the vessels at the base of the 
placenta (text fig. 20), As it turns out into the carpels and grows 
upward, it is in close proximity to vascular traces from which it 
draws its nourishment. The cells of the region of the carpel through 
which the sac moves are richer in protoplasm than the surrounding 
tissue (fig. 31). No starch was found in these cells, yet their general 
