1914] COULTER— REPRODUCTION IN PLANTS 349 



gest a working hypothesis. There are three features belonging to 

 the most primitive gametes that deserve attention: they are 

 motile, small, and pairing cells. 



It is evident that motility is not an essential feature of sexual 

 cells, for early in the evolution of plants, one of the pairing gametes 

 becomes passive, and finally both are non-motile. Motility, there- 

 fore, is a secondary feature common to both gametes at first, 

 retained with remarkable Dersistence bv the male eramete. but 



most 



small size of the primitive 



as compared with the spores of the same plant, is not an essential 

 feature of sexual cells. In other words, they are not gametes 

 simply because they are smaller than the spores. Later in the his- 

 tory of plants, one of the pairing gametes becomes much larger 

 than the spores, and still it is a gamete. The difference in size 

 is due chiefly to the varying bulk of the cytoplasm, and in some 

 seed plants the sperm is a naked nucleus. The conclusion is that 

 the amount of cytoplasm is also a secondary feature of sexual cells. 

 It is certainly true that the activity of the cytoplasm of the egg is 

 intimately related to the act of fertilization, not only as a source of 

 nutrition, but also as the source of an activating substance, which 

 Lillie has called " fertilizing which determines the physiological 

 moment of fertilization. It should be recognized, however, that 

 even this activating substance is not an essential feature of sexual- 

 ity, but belongs to the category of secondary features which aid in 



m 



gametes 



belongs to all gametes, and yet there are pairing and fusing cells 

 that are not gametes. If pairing and fusing are not peculiar to 

 gametes, they do not represent the essential features of sexuality. 



seems 



asm, and reoresents a mutual attraction that makes 



mo 



possible. 



It is certainly true, however, that the primitive gametes differ 

 from the spores with which they are clearly associated in pairing 

 and fusing, and this difference should be accounted for first. Since 



