Hill , — The Floral Morphology of the Genus Sebaea . 485 
and the lower part of the style bearing the two secondary stigmatic 
patches. 
After the removal of the apical stigma, the secondary stigmas 
increase very markedly in size, and in aurea and 5 . amhigua in 
particular they become so much enlarged that they almost block the throat 
of the corolla tube. 
These organs in flowers so mutilated were then pollinated, when in 
a receptive condition, with pollen from other flowers. The subsequent 
increase in size was very marked, and in vS. aurea they measured 
075 mm. in length when fully developed after fertilization had taken place, 
while in 5 . ambigua the development of papillae down the style may take 
place to such an extent that in several cases the secondary stigmas were 
found to be as much as 2-5 mm. long (Fig. 16). In almost every instance 
where the apical stigma was removed and the secondary ones were 
pollinated, ovules were formed and seeds were eventually ripened. These 
seeds were found to be capable of germination, and gave rise to a fresh 
crop of vigorous plants. As the portion of style above the secondary 
stigmas, left after the removal of the apical stigma, withered up before the 
secondary stigmas were pollinated, it seems clear that the development 
of ovules can only be attributed to the pollination of these lateral patches 
of stigmatic papillae. Good sections of the secondary stigmas are very 
difficult to obtain, owing to their somewhat spongy nature when well 
developed, and also to the very slender nature of the style, but in several 
instances pollen-tubes have been seen not only growing out of the pollen- 
grain adhering to the papillae, but also penetrating between them into the 
tissues of the style (Text-fig. 2). 
The first experiments in removing the apical stigma were performed 
with Sebaea aurea in July, 1908, and further experiments were made on the 
other species in the two following years. Control experiments consisted 
in removing the unburst anthers from flower-buds, to prevent self-fertiliza- 
tion, and in pollinating the upper stigma only, care being taken that no 
pollen reached the lower secondary stigmas. In these cases the amount 
of seed produced was small and poor, and was in marked contrast to the 
quality and amount of that yielded by the ovaries of the flowers mutilated 
by decapitation of the style. No seed was formed in flowers from which 
pollen was excluded. It was very noticeable that in flowers in which the 
anthers were not removed the lower stigmas were covered in due course 
with their pollen, and the amount is probably augmented, owing to the 
twisting of the corolla as the flower withers. The function of these, secondary 
stigmatic patches appears, therefore, to be a direct aid to self-fertilization. 
It is also evident that from their position at the throat of the corolla tube 
they could quite easily be cross-pollinated by insects who might visit the 
flower charged with pollen from another source. 
L 1 2 
