720 
DUNCAN S. JOHNSON 
is from nearer the base of the anther, the latter from nearer 
the tip. That is, at the base of the theca in question, there seem 
to be two distinct sporangia, while a little higher up there is a 
continuous sporogenous mass across the whole width of the theca. 
Each of the two sporogenous masses shown in figs. 24 also seems 
the equivalent of the two usually present in each theca. So also 
the sections from different stamens shown in figs. 23, 27, 28, ap- 
pear to indicate different stages of a progressive fusion of two 
sporogenous masses like those which have remained distinct in 
the stamen shown in fig. 26. 
That no actual fusion of primarily distinct sporogenous masses 
really occurs during development is clearly seen from a study of 
sections of younger stamens. For example, I see no reason for 
doubting that the stamen shown in fig. 9 would have matured into 
one of the sort shown in fig. 26, or that the one shown in fig. 10 
would have matured into one such as is shown in figs. 16 and 20. 
There is no evidence, from the development of Piper betel, that 
originally distinct sporogenous masses may become continuous, 
either by the degeneration of the sterile partition separating them, 
or by the cells of the partition becoming sporogenous. 
A third possible explanation of the origin of this variability 
in the number of pollen masses present in the mature stamen of 
Piper betel is that discovered for Lemna, in which Caldwell ('99), 
has shown that certain portions of the primarily sporogenous 
tissue fail to form spores, and, instead, give rise to three sterile 
septa, dividing the sporogenous mass into four. In a stamen like 
that shown in fig. 8, for example, the belated appearance of a 
septum of sterilized archesporial cells in the upper theca would 
give rise to a stamen like that in figs. 9 or 26. But here also the 
study of development shows that nothing of the sort occurs. 
In brief, all available evidence goes to show that the number of 
pollen masses present in the mature stamen is the same as the 
number of archesporial groups first laid down in the young anther. 
There is no fusion of fertile tissues, nor any secondary steriliza- 
tion, to form septa, of sporogenous tissues once delimited. Hence 
it is evident that, whatever the terms confluence of pollen sacs'' 
and '^fusion of sporangia" may be intended to mean, they cannot 
