May, 1909,] Nuclear Divisions in Convallaria 501 



('05) found in certain Orthoptera. Such a pronounced irreg- 

 ularity in appearance as shown in Fig. 19 is of comparatively 

 rare occurrence. 



In this division the spindle fibres are not nearly as prominent 

 as in the first division. No centrosomes are present. The 

 fragmented nucleolus, which was in the niucleus during the 

 prophase, is again in the cytoplasm. 



That the plane of division is longitudinal is evident (Figs. 

 19a, 20a). The line of cleavage seems to be that occasionally 

 indicated earlier in the process (Figs. 14a, 16a, 19a, 20a). The 

 homologous daughter chromosomes, as expected, show a marked 

 similarity in shape and size (Figs. 19a, 20a). Although a great 

 variety of shapes were seen the U and I shaped chromosomes 

 predominated. (Figs. 20, 21). 



The nuclear membrane disappears "during late prophase 

 (Fig. 14), is absent during metaphase and early telophase (Figs. 

 15, 16, 26), and again appears in late telophase (Fig. 17). It 

 remains present during the subsequent stages (Figs. IS, 22, 23, 

 24, 25). The breaking down of the membrane is co-incident 

 with the migration of the nucleolus into the cytoplasm as 

 micronticleoli. 



THE MICROSPORES. 



After the late telophase of the second division (Fig. 21), the 

 chromosomes of the daughter nuclei develop into irregular net- 

 works, and, with the appearance of a transverse wall, a cell is 

 formed out of each of the two hemispheres, and thus the spore- 

 tetrad is completed. The formation of the wall is, in Conval- 

 laria, as in most of the Monocotyledons, "successive," i. e., 

 begun after the first division, and completed after the second. 

 The young microspores are arranged bilaterally in the spore- 

 tetrad (Fig. 23). Occasionally, as already said, a cell is found in 

 which one of the cells has not yet divided, and then only three 

 cells are enclosed by the wall (Fig. 22). Whether the undivided 

 cell will ultimately divide, could not be determined. 



Sometime after the common wall is completed, each micro- 

 spore develops a delicate wall which seems to be independent of 

 the former. This thin wall grows in thickness and later becomes 

 differentiated into the intine and exine layers (Figs. 23, 24, 25). 

 The tetrad then breaks, and the four microspores become free and 

 separated. All this occurs long before the flower opens. 



The next change that the nucleus undergoes is a division, 

 whereby the generative cell is formed. The nucleus of this cell 

 is at first chromatic and so deeply stained that its structure can- 

 not be determined (Fig. 24). After a short time, however, the 

 cell becomes differentiated by a distinct cell wall. Although the 

 generative nucleus is at all times more chromatic than the 

 nucleus from which it arises, its structure becomes discernable 

 sometime before the flower opens. The generative nucleus usu- 



