i 9 o8] BURLINGAME—PODOCARPUS 171 



remains large and the cell itself is often much larger than when 

 two are cut off {text fig. 1). Sometimes the three resulting nuclei 

 then lie in a row across the spore {text fig. 1) and differ but little from 

 one another in appearance. The walls separating them then run 

 nearly straight across, and it is only by courtesy that the tube nucleus 

 can be said to be free in the spore cytoplasm. To all intents and 

 purposes it is as much walled-off into a cell of its own as the pro- 

 thallial nucleus, for both eventually are freed in a common cytoplasm 

 {text fig. 2). Further, it is to be noted that in this case the prothallial 

 cell does not have a fixed position with respect to the wings, as in the 

 ordinary course of events. 



After the tube nucleus has been cut off, the primary sperma- 

 togenous cell divides transversely, giving rise to two cells {fig. 25). 

 Usually these cells differ markedly in size, the smaller lying to one 

 side and the larger lying nearly in the center of the pollen grain {fig. 

 26). In this case one may safely use the usual terms for them and 

 speak of the larger as the body cell and the smaller as the stalk cell. 

 It not infrequently happens that the two derivatives of the generative 

 cell are about equal in size {figs. 24, 27; text figs. 5, 6), though even 

 then one is centrally placed and the other laterally. Before the 

 division of the generative cell, the second primary prothallial cell has 

 usually divided at least once. In this case the generative cell often 

 sinks down among its derivatives {figs. 23, 27) . The stalk cell is some- 

 times placed transversely to the body cell {text figs. 5, 6). Whether 

 these large stalk cells will produce male cells or not can only be con- 

 jectured, though the fact that some of them seem to retain their 

 cytoplasmic envelope, almost as distinct from the cytoplasm as the 

 body cell itself, up to a certain early stage, would lead one to suppose 

 that in some cases they might do so. Text fig. 7 may show three deriv- 

 atives of the generative cell and two prothallial cells in addition to 

 the tube nucleus. One of the three derivatives of the generative 

 cell has no cytoplasmic sheath and is probably a stalk nucleus; the 

 two lying close together are ensheathed, though I have been unable 

 to ascertain definitely whether each has a distinct sheath or whether 

 both lie in a common sheath, or whether perhaps one of them is 

 not merely beneath the cytoplasmic sheath of the other. If either the 

 first or second supposition is true, there still remains the question 



