Allen—Spermatogenesis and Apogamy in Ferns. 43 
mutter Zellen war, die man in den Makrosporangium angelegt 
sail.’ 7 The observation that there were fewer than sixteen 
spore mother cells at the time when the reduction division be¬ 
gins, is at least suggestive. 
Against the possibility of such a fusion in Marsilia is the 
fact that the smaller spore mother cells which were poor in chro¬ 
matin content also undergo reduction divisions, according to 
Strasburger. It would certainly be hard to believe that cells 
already possessing the reduced number of chromosomes could 
carry out regular reduction divisions. Possibly these cells cor¬ 
respond to the cells remaining unfused in the sporange of Aspid¬ 
ium, but they seem to have been of more frequent occurrence in 
Marsilia than I have found them in Aspidium. 
On the second hypothesis suggested as to the origin of the 
condition in Aspidium falcatum, i. e., that the fusion in the 
spore sac arose before the fern had become apogamous, the re¬ 
duction divisions of unfused cells might of course be expected. 
In this case the difficulty Strasburger feels in the assumption 
of a sporophyte with the haploid number of chromosomes would 
also be obviated. I am, however, of the opinion as argued 
above, that in Aspidium falcatum it is the haploid number 
which appears in both generations. 
The finding of a fusion of vegetative nuclei at the beginning 
of the sporophyte in the prothallium of Lastraea as described 
by Farmer and Digby, combined with the above described dis¬ 
covery of a fusion of vegetative nuclei in so unexpected a place 
as the young sporange in Aspidium falcatum emphasizes still 
more strongly the fundamental significance of cell and nuclear 
fusion for the maintenance of the life cycle. In these cases, at 
least, when the normal sexual fusion has disappeared it is re¬ 
placed by a substitute fusion either at the same or some other 
stage in the ontogeny. The facts already ascertained show the 
need of careful search for similar fusions in such plants as are 
now supposed to be continually apogamous. 
This work was done under the supervision of Dr. H. A. 
Harper, to whom I am indebted for many valuable suggestions 
and criticisms. 
