1897] BRIEFER ARTICLES 211 
the gyncecium, it would seem to be a clear case of axial prolongation 
and prolification. 
fig. 3 represents a very similar case of prolification. Here, how- 
ever, the main flower has four sepals and two stamens, without any 
sign of ecblastesis, while the gyncecium has undergone further devel- 
opment, the walls of the capsule being split down one suture, notched 
at the other, and thrown back as a sort of spathe-like envelope, from 
which emerges the here further developed axial prolification in the 
shape of a distinctly formed little trisepalous and diandrous flower (e). 
Its little ovary is borne upon a distinct stipe, which suggests the begin- 
ning of still further prolification. A careful examination has failed to 
reveal any traces of ovules or funiculi on the spathe-like organ (¢). 
However, its analogy to the similar and ovuliferous structure shown in 
Jig. 1 is too close to leave any doubt as to its gyncecial nature. Fig. ¢ 
shows another flower higher in the raceme and of similar but less 
developed structure. The bases of two stamens were found within 
the calyx (a), and the interior of the gyneecium (4) is shown in fg. 5 
(where it has been divided through the middle of the valves). The 
placenta, which, from the way the ovary was divided, here appears as 
a median line, bears a single ovule, while at the base is a rudimentary 
flower similar to that shown in fg. 2. 
A little higher upon the raceme occurred the proliferous flower 
shown in fig. 6. It has only two sepals, the dorsal and ventral (a), 
and one stamen (4) borne opposite the ventral sepal. The central axis 
was then prolonged and a second disepalous calyx (c) borne above. In 
the axils of each of these sepals were borne rudimentary flowers (d), 
and springing laterally from the axis at nearly the same height were 
three stamens. Only two of these can be shown in the figure, the third 
being concealed by the axis. The morphology of this flower is capable 
of various interpretations, and in its mature state the precise relation 
of the stamens to the sepals could not be satisfactorily made out; 
especially as the stamens were of an odd number. However, it seems. 
likely that here the lateral sepals have been developed, and the dorso- 
ventral pair suppressed, while the longer pairs of stamens of the theo- 
retical tetradynamous andrcecium are represented on one side by a 
single stamen, on the other by a pair. The structure of the minute 
flowers (a) was essentially as those described above, one of which is 
represented in fg. 7, e. The main axis was then produced by another 
distinct internode and bore another disepalous flower (e) which con- 
