1908] BROWN—PEPEROMIA 451 
closely with that just described for P. Sintensii. There is a single 
archesporial cell which cuts off a tapetal cell and then forms the 
embryo sac. The tapetum divides as in P. Sintensii. The embryo 
sac nucleus divides to two, four, eight, and finally sixteen nuclei. 
The first division is heterotypic and shows the usual synapsis. 
At this division a cell wall is formed across the sac. This wall is 
variable in position and generally disappears before the next division, 
but may persist as a remnant after it. At the second division plates. 
are formed on both spindles. One soon disappears, while the other 
forms a wall separating one nucleus from the other three. This wall 
is variable in position, and all signs of it are usually lost before the 
next division. The next two divisions leave the mature sac with 
sixteen nuclei. 
Peperomia ottoniana ~ 
The material of this species was collected in Mexico by Dr. CF: 
CHAMBERLAIN. It is very much like P. Sintensii, but shows some 
constant differences. It was identified by M. Castmir DE CANDOLLE. 
Owing to lack of young material the investigation of this species 
had to be begun with the four-nucleate stage. There were only three 
ovules showing a four-nucleate sac, but they are worth recording, as 
they appear perfectly normal, and seem to throw some light on the 
problem under discussion. 
The youngest sac (fig. 31) shows the micropylar nucleus com- 
pletely cut off from the other three by a very distinct and well-devel- 
oped wall. This nucleus is much larger than the other three, which 
are all about the same size. The cell occupied by this large nucleus 
contains much denser protoplasm than the one containing the other 
three. The stage just described resembles rather closely the four 
megaspores of the ordinary angiosperm, where one megaspore is to 
form the sac while the other three degenerate. |The further develop- 
ment, however, seems to be different, and agrees with the two species 
of Peperomia just described. 
The two other four-nucleate sacs which were seen are larger and 
apparently older than the one just described, and show no sign of a 
wall separating the nuclei, which are all about the same size (fig. 32). 
The eight nuclei formed from these four show no appreciable differ- 
ence in size. 
