140 MINNESOTA BOTANICAL STUDIES. 



cell nucleus in dividing, so that it was seen dividing much more 

 frequently than the equally large and apparently as well-fed 

 nucleus of the macrospore. 



As stated above the two cells resulting from the division of 

 the macrospore mother cell were not seen in the resting condi- 

 tion, though the four potential macrospores resulting from their 

 division were frequently found. The two cells were seen once 

 with their nuclei dividing, but the nuclei were indistinct because 

 of improper staining. The two-celled condition, which was 

 not seen, is doubtless a very transient phase, the upper of the 

 two cells almost immediately cutting off a small cell from its 

 lower end, and the lower of the two likewise dividing at once 

 into two cells, a large one below and a small one above ( Fig. 7 ). 

 This four-celled stage seems to be constant and of quite long 

 duration as it was frequently seen as represented in Fig. 7 or as 

 in Fig. 10. In the latter the lower more successful cell which 

 is to become the fertile macrospore has begun to absorb the 

 other three potential macrospores for its own nourishment. 

 This may be seen by observing the decrease in amount of cyto- 

 plasm contained in each of the three cells as compared with the 

 same three in Fig. 7 and the swelling of the softened periclinal 

 cell walls between any two of them and between the lowest one 

 and the absorbing cell below. The next observed develop- 

 mental condition was that in which the three cells were all ab- 

 sorbed except possibly a refractive cytoplasmic cap at the sum- 

 mit of the absorbing cell, which more probably represents a 

 nearly absorbed tapetal cell (Fig. 8). This brings us to the 

 macrospore as shown in the figure. 



Germination of the Macrospore. — When first formed the 

 macrospore has very nearl}^ the shape of the four cells replaced 

 and shows the nucleus at the center surrounded by cytoplasm, 

 while the upper and lower ends are each occupied by a large 

 vacuole. Though I cannot account for this apparent poverty 

 in cytoplasm at this time, the condition seems to be typical. As 

 the macrospore increases in size by the absorption of tapetal 

 cells and those cells of the macrosporangium which surround its 

 upper lateral wall, cytoplasm increases in amount (Fig. 9). In 

 the figure the nucleus is lying in its usual longitudinal direction 

 about the middle of the macrospore while dividing. Before the 

 nucleus divides, the macrospore increases considerably in size 

 by the absorption and pressing upon surrounding tissues (comp. 



