i9o8] YAMANOUCHI—APOGAMY IN NEPHRODIUM 307 



\ 



sion, as in normal cases; in rare cases the tetrad division with reduc- 

 tion is entirely cut out. The egg retaining the sporophytic number 

 of chromosomes then develops the embryo parthcnogcnetically. But 

 in the majority of cases he reports that while the normal development 

 of the embryo sac is proceeding, an embryo sac from a cell quite near 

 the tetrad (in H. atir a nti a cum), or in the integument; or in the chalazal 

 region commences to develop. The normal embryo sac is then 

 destroyed sooner or later by the encroaching embryo sac of vegetative 

 origin. This development of an embryo sac from the nucellus is a 

 new case, entirely different from those known in Funkia, Coelebogync, 

 Citrus, Opuntia, and Alchcmilla pasioralis, for in the latter cases the 

 embryo is produced directly from the nucellus, instead of through an 

 intercalation of embryo sac formation, and hence the embryo is a 

 vegetative bud from the sporophyte and entirely independent of 



gametophytic activities. 



The papers of Strasburger and of Farmer and Dibgy, which 

 are the latest contributions to the cytology of apogamy among pterido- 

 phytes, were briefly reviewed in the first part of this paper. 



Summarizing the cytological facts in connection with apogamy 

 among spermatophytes, as interpreted by different investigators, it 

 seems evident that apogamy is closely associated with the suppression 

 of sporogcnesis in the megaspore mother cell. This necessarily results 



^ I in the chromosome number in the nucleus of the sporo- 

 phytic generation, yet a structure is developed with the morphology 

 of the gametophytic generation. Thus the embryo sac will contain 

 the usual number of nuclei grouped in the typical manner, but these 

 nuclei have the sporophytic number of chromosomes. From the facts 

 of apospory, it seems probable then that the development of a game- 

 tophyte may result from an interference with the normal life-history 

 and under conditions favorable to the gametophyte, even though 

 the nuclei retain the sporophytic number of chromosomes. 



If the doubling of chromosomes is that result of fertilization 

 necessary to start the sporophyte generation, it is no surprise that either 

 an egg with the sporophytic number or a vegetative nucleus with the 

 same number may develop a sporophyte. In all the foregoing cases 

 of apogamy this seems to be the situation. 



The case of vars. bolydadyla studied by Farmer and Digbv is 



