164 INTRODUCTION TO EMBRYOLOGY OF ANGIOSPERMS 



There is also considerable evidence to indicate that even in those 

 plants in which the pollen grains are shed in the two-celled condi- 

 tion, the generative nucleus is already in the prophase stage and the 

 process of division is merely continued in the pollen tube. Some- 

 times the nucleus may even show a pro-metaphase stage which is 

 distinguishable from a typical metaphase only by the delay in the 

 dissolution of the nuclear membrane and the organization of the 

 spindle. This has been demonstrated very clearly in Impatiens 

 (Wulff, 1934; Heitz and Resende, 1936), Bulbine (Geitler, 1942) 

 and other plants. 



Occasionally both two- and three-celled pollen grains have been 

 reported in the same plant, as in the cleistogamous flowers of Viola 

 (West, 1930), in Dionaea (Smith, 1929), Circaeaster (Junell, 1931), 

 Nicotiana (Poddubnaja-Arnoldi, 1936), Epimedium, and Iris 

 (Schnarf, 1937), but this is probably due to environmental influ- 

 ences. Poddubnaja-Arnoldi (1936) found that, in several kinds 

 of pollen grains which are normally two-celled, the generative 

 nucleus divided before germination if the grains were kept for some 

 time on a sugar-agar substrate. Eigsti (1941) was similarly able to 

 induce a precocious division of the generative cell in the pollen 

 grains of Polygonatum canaliculatum. In Holoptelea integrifolia 

 (Capoor, 19376) the pollen grains are shed at the two-celled stage, 

 but the generative cell divides on the surface of the stigma before 

 the pollen tube has started to grow. 6 



Details of the division of the generative cell vary depending on 

 whether it takes place in the pollen grain or in the pollen tube. In 

 the former case, spindle fibers and a normal metaphase plate have 

 been regularly observed, and the process does not seem to differ in 

 any essential way from a normal mitosis. Cytokinesis, resulting 

 in a bipartitioning of the cell, may take place either by a process of 

 furrowing as in Juncus (Wulff, 1939a) (Fig. 98//"), or by the laying 

 down of a cell plate as in Asclepias (Finn, 1925) and Portulaca 

 (D. C. Cooper, 1935). Witmer (1937), who observed both cell 

 plates and constriction furrows in Vallisneria, states that in his 

 material these two factors varied in importance. 6 In some pollen 



6 In Euphorbia terracina (D'Amato, 1947), which is at the other extreme, the 

 division occurs only after the pollen tube has entered the embryo sac and its tip 

 has come to lie by the side of the egg. 



6 See also Kausik and Rao (1942). 



