1915] HUTCHINSON— MALE GAMETOPHYTE OF PICEA 295 



Discussion 



1. Double pollen grains 



There are accounts of double pollen grains occuring in a number 

 of species. Probably first was Chamberlain's (10) description 

 of Lilium ligrinum. "In about 20 cases there was a distinct wall 

 dividing the microspore into two nearly equal parts." Both cells 

 contained starch. His rig. 20 shows one of the cells containing 

 two nuclei "which seem to represent generative and tube nuclei." 

 One of the cells was regarded as prothallial, the other as antheridial. 

 Schaffner (n) found compound grains where two or more of the 

 spores of a tetrad clung together (Typha latifolia). Guignard (12) 

 and Miss Pace (13) figure four microspores of an orchid within a 

 common wall dividing to form tube nuclei and generative cells. 

 Coker (13a) describes double grains in Larix euro pea. His fig. 6 is 

 similar to my fig. 13; his fig. 8 corresponds to my fig. 12. He sug- 

 gests that "the mother cell had only divided once, so that only 

 two instead of four pollen grains were formed." In some of these 

 grains "division proceeded as usual except that only one prothallial 

 cell is evident" (cf. fig. 32). Pollock (3) has described a number 

 of variations in the pollen grain of Picea excelsa. "In the material 

 examined, the proportion of double pollen grains was found to be 

 2.4 per cent in a count of n 20. The three or four cells lying along 

 the dorsal side of the pollen grain of this type do not constitute a 

 prothallium or gametophyte of unusual size. They constitute 

 the smaller portion of a pollen grain separated by a division wall 

 into two nearly equal portions, each of which may form a typical 

 antheridium." Double pollen grains have been variously inter- 

 preted. In Picea canadensis a study of the stages of development 

 has shown that the two cells from which the double grain arises 

 are the result of a primary division of the microspore (figs. 11, 13), 

 and that one of these cells corresponds in origin to the more usual 

 evanescent cell. All gradations between an equal division and 

 one which cuts off a lenticular evanescent cell have been found 

 (figs. 6, 7, 10, n, 13). In the double pollen grain of Picea one of 

 the antheridial groups is homologous with the usual evanescent 

 polar cell. 



