76 EIGHTEENTH REPORT. 



varieties of the same species the sporophyte arises apogamously 

 from the gametophyte, and the latter aposporously from the spor- 

 angial sorus of the sporophyte, both generations retaining the 

 chromosome number equal to tlie (lii)loi(l number in the closely 

 related normal plant. 



In view of these facts elucidated from the apoganious terns and 

 flowering jilants and of the extreme variability of the position of 

 the main points of the sexual cycle in different plants it seems far 

 more reasonable to me to distinguish sporophyte and gametophyte 

 on morphological grounds and to be walling to homologize struc- 

 tures even Avlien the chromosomes are diploid in number in the one 

 and haploid in the other. This would permit the structure aris- 

 ing from the zygote in Coleochaete to be homologized with the 

 sporophyte of the liverwort, a much needed homology in phylo- 

 genetic speculation. 



One thing remains clear, however. The retention of the reduction 

 division seems to demand a sexual union somewhere. If this can 

 be in the normal way, well and good, if this is prevented there must 

 be a substitution union elsewhere. Thus in Aspidium falcatum 

 when the gametophyte buds off into a sporophyte, thus eliminating 

 the normal union, the sexual union is replaced by the union of 

 spore mother cells in the sporangium. In one species of Pyronema 

 studied by Brown in wliich the antherid and trichogyne do not fuse 

 the i)lace of the mah' nuclei is taken by other feniale nuclei already 

 present in the iiogone. In tlie rusts the pycniospores are almost 

 certainly s])erm cells, but they never have a chance to function so 

 that their place is taken by cells adjacent to the oogones, probably 

 modified oogones themselves. 



One further point, too, is clear. Just as the point of cell and 

 nuclear union is not absolutely tixed so the point of reduction 

 division is movable. In Nemalion it is shortly after nuclear union 

 and before carpospore formation, in Polysiphonia it is after carpo- 

 spore formation at tlie close of the tetrasi)ore stage. In Fucus it is 

 at the close of (lie vegetative growth at gamete production (as 

 in animals) in Tlotlirix probably at tlie germination of the zygo- 

 spore. 



I shall not undertake to show why this is so or what is the pur- 

 pose of sexual reproduction. I have merely attempted to show 

 some of tlie features connected with this subject. 



Department of Botany, Michigan Agricultural College, East 

 Lansing. 



