244 
Ethel M. Poulton 
flower, with the margins closing over and forming small ovules. The 
remaining parts of the same flower (after the removal of calyx and 
corolla.) are seen in Fig. 6. The central stigma, csg, is 2-lobed, a-e are 
“staminate carpels” all of which are joined at the base; a and h tend 
to curl over at the tip, and produce minute ovules in various stages 
of development. Each also has a distinct stigmatic surface. Fig. 7 is 
an enlarged view of b, showing the ovules, and the stigmatic surface sg. 
In Figs. 8 and 9 are seen “staminate carpels” from other flowers. 
Fig. 10 represents four barren “staminate carpels” (i.e. without 
ovules) a-d, and two better developed ones e, /, which unite with 
each other and with the base of the main gynaecium. These (e,f) had 
rudimentary ovules on their margins. In Fig. 11 the union is almost 
complete, a solid column being formed of the central gynaecium and 
the six “staminate carpels” the stigmas (sg) being separately recog¬ 
nisable. The normal 2-lobed stigma, csg, is seen projecting from the 
surrounding stigmatic surfaces (sg). Fig. 12 shows the same flower 
rotated through 90°. 
A cross-section of this composite gynaecium is shown somewhat 
diagrammaticallyin Fig. 13. The main gynaecium is seen in the centre 
and is of typical structure. The six “staminate carpels” have com¬ 
pletely fused with it. The plane of the section has passed through 
some of the ovules formed by these. 
Thus, the suppression of maleness, as regards its ordinary mani¬ 
festation is complete throughout the whole plant, and the tendency 
to femaleness is strongly marked. It is possible that even the corolla 
shows this tendency in the curling over of the edges of some of the 
petals (Fig. 4). Very significant is the preservation of the character¬ 
istic number and arrangement of the stamens, as though the pri- 
mordia of these had been formed as usual. It is difficult to resist 
speculation. Perhaps one might hazard the suggestion that an absence 
of the factor (or factors) making for maleness, or a preponderance of 
that (or those) associated with femaleness, might account for these 
curious divergencies from the familiar structure. 
Worsdell(i) describes a similar abnormal flower of Cheiranthus 
cheiri and states that it is an example of “carpellody” which “is a 
frequent phenomenon revealing to us the fact that the stamens and 
carpels are very closely allied organs, and the facility with which the 
one may change into the other, doubtless due to the fact that both 
are derived from a common ancestor, the asexual sporophyll which 
exists to-day in some of the more primitive types of plants such as 
ferns, horse-tails, and some lycopods.” (2) 
