428 The American Naturalist. [May, 



spheres of the female nucleus which they quickly join and so form two 

 new couplets of spheres, each couplet containing a sphere of two dif- 

 ferent elements. This conjunction of the spheres takes place kefore 

 the nuclei unite, but they do not fuse until the division of the nucleus 

 begins. While still as couplets they take their position above and 

 below, marking out the axis of the coming spindle, which is that of 

 the long axis of the oospore. The second male nucleus sometimes 

 escapes from and sometimes remains in the pollen tube and ultimately 

 disappears in the protoplasm. The reason for the existence of two 

 male nuclei is in doubt. That the division of the single primary one 

 into the two smaller masses is so that the male and female nuclei may 

 be nearly in equilibrum as to size, is not a satisfactory suggestion; 

 and since the two male nuclei are equally active and effective in fer- 



matter of differentiation. After a Ion- contact of the male and female 

 nuclei, the nuclear membranes at the surface of contact disappear ami 

 the contents form a common mass; that is. the nuclear fluids mingle, 

 but there is not a true fusion of the chromatic elements. ( 'ontraction 

 and orientation of these elements follow with the formation of a spindle 

 and nuclear disk, and a return to tweim-fbiir rods. Twelve of these 

 may have been furnished by each nucleus but they are not to be dis- 

 tinguished from one another. The two groups of secondary segments 

 collect at the poles, a wall of cellulose is formed at the equator and 

 two embryonic cells come into existence. Thus fecundation has resulted 

 in doubling the number of rods in the first segmenting nucleus. 



It happens with curious constancy that the secondary nucleus of the 

 embryo-sac begins its division for the formation of endosperm, just at 

 the time that the male nucleus penetrates the oosphere. This second- 

 ary nucleus formed by the union of the two polar nuclei, which were 

 larger and rich in chromatin, contains a -renter number of rods, as do 

 its derivatives, than are to be found in the eizg nucleus. So likewise 

 do all the nuclei ,,f the .-ml«..-p.-rm contain a -reater number of rods 

 than do the nuclei of the rot of tin- ovule and m m r . l-am of ^ 

 plant. This variation in the number of rods opposes the idea of the 

 chromatic segments retainin- tlmi- individuality .hiring the resting 

 stage of the nucleus. In L>:nc»niirt)> and in (!<ihuitht)», as the embryo- 

 sac becomes somewhat crowded with nuclei and their spheres, and as 

 cell walls arise, there mav be inrbul-d a. manv a- ten or twelve nuclei 

 in a single cell. Under" t lies,- conditions the Spheres a-- 



