246 BOTANICAL GAZETTE [OCTOBER 
of times) to such a condition as Habenaria where the archesporial 
cell without dividing functions as a spore mother cell. 
The only way in which we can claim that megaspore nuclei must 
accompany chromosome reduction is by defining a spore mother cell 
as that cell in which reduction is initiated and spore nuclei as the nuclei 
resulting from the reducing divisions. The logical conclusion of this 
would be to make chromosome behavior the sole criterion for distin- 
guishing spores, sporophytes, and gametophytes; but since the four 
megaspores and embryo sac of Alchemilla (MuRBECK ’o01) may have 
the diploid number of chromosomes, or the apogamous embryos of 
Nephrodium (YAMANOUCHI ’08) the haploid number of chromosomes, 
we cannot regard chromosome behavior as the sole, if indeed the most 
important, criterion for distinguishing any of these structures. 
A distinction between the first division of the megaspore and a 
division giving rise to megaspores is that while in the first case no cell 
plate is formed on the spindle, in the latter case either a wall or a cell 
plate is formed on the spindle. This wall may be formed in the 
embryo sac when this is derived from more than one megaspore, 4S 
is apparently the case in Peperomia (BRowN ’08). The first division 
of the embryo sac mother cell nucleus of Cypripedium (PACE ’0 
agrees with that of those derived from one megaspore in having 0 
cell plate formed on the spindle. ‘ 
The evidence for either view is inconclusive, but seems to the writer 
to be in favor of the idea that in Cypripedium the second division 1) 
megaspore formation has been left out, and the place where “ reduc- 
tion is completed” changed to the first division of the nucleus of the 
embryo sac mother cell. 
Lroyp (’02) and Coutrer (’08) have advanced the idea that when - 
megaspores are not formed the first four nuclei of the embryo BAC ATE 
megaspore nuclei. This is probably true in such cases as Lilium 
(CouLTER and CHAMBERLAIN ’03) and Peperomia (BRowN 708), and 
as COULTER (’08) suggests in some other sixteen-nuclea bryo sacss 
but if the ideas brought forward here concerning Cypripedium 4t° 
correct, it need not be of universal application. CAMPBELL ('09)» 
however, seems to think that even the embryo sacs of Lilium and 
Peperomia are not derived from more than one megaspore. 
In discussing the view of CouLrer (’08) that when the megaspore 
+a 
c 
