356 TRANSACTIONS AND PROCEEDINGS OF [Sess. lvi. 



and such a division may also have occurred in the ovule 

 (fig. 12), in which the cell x shows a sharply defined and 

 deeply stained portion beneath the dermatogen. 



The division of the archesporium into four cells seems to 

 be the rule in the ovules of a young bud ; if, however, ovules 

 be examined which develop at or after the time a bud has 

 fully expanded into a flower, the archesporium is found to 

 give rise to only three cells (figs. 10 a-d, 11-13 a), as no 

 division of the cell x (figs. G and 7) takes place. The 

 explanation of this difference does not seem to be difficult : — 

 The basal ovules of a. young flower will be better supplied with 

 nourishment than the apical ovules of a mature flower, and 

 hence the greater amount of nutritive material at the disposal 

 of a basal ovule will give each individual cell in that ovule 

 a "Teater chance of fulfillinfi its function. How different 

 if food is scarce and the competition between cells severe ! 

 The cell placed most advantageously for having its wants 

 supplied will thrive, while the remaining cells have to 

 struggle against odds, and will succumb in the end. This 

 bitter struggle is clearly shown in the ovule ; for what is the 

 cause of the development of the central periblem-cell into 

 the archesporium (fig. 2, &c.), of the unequal division of the 

 archesporium (fig. 6), of the larger cell thus derived preceding 

 its sister-cell in division (figs. 7, 8), or the smaller cell (figs. 

 11, 13 «) not undergoing division at all? — The reason has 

 to be looked for in the fact that the successful cell lies in 

 each case in close contact with the plerome-elements, and 

 that hence it is supplied directly with food-material, which 

 travels to the ovule along the plerome-tract. For the same 

 reason each one of the four cells derived from the physiolo- 

 gical archesporium may attain an exceptional size, if the 

 competition in the ovule has not been as great as usual, 

 due to the development of a smaller number of periblem- 

 cells than normal (fig. O*^). 



The next change we have to notice is a peculiar gelatinous 

 swelhiig of the walls of the three or four cells derived from 

 the archesporium (figs. 9'', 9'', 10 a-d). This change is in 

 most cases restricted to the two periclinal walls between 

 the cells x, y, and E.,S. (fig. 10 <■/) ; next in frequency the 

 condition is found in which again the two transverse walls 

 are swollen, but in addition also portions of the anticlinal 



