560 — PROCEEDINGS OF SECTION D. 
I have never seen one on the flowers. This may arise, however, 
from the fact that it is not an indigenous plant, and the particular 
insects affecting it not being found here.* The sepals are greenish, 
and the upper one is hood-shaped. The inner petals form a tube, 
the lower two being thickened and yellow, with a scale on the 
inner side, and some hairs. The remaining petals are white, larger, 
and spreading. There are two nectar-secreting glands at the base 
of the tube. The tube of two petals encloses the anthers in the 
first stage, and the style is short and not seen. An insect visiting 
the flower for nectar, when in this stage, must insert its proboscis 
between the anthers, and so get dusted with pollen. After a time, 
the trifid stigma protrudes between the anthers, while the hood- 
like sepal moves down so as to cut off all access to the honey 
except through the tube, A pollen-dusted insect must, in this 
second stage of the flower, smear the pollen on the stigma. But 
if no insects should visit it, the anthers still retain pollen, and the 
stigma touches them as it expands, and also receives it from the 
hairs within the lip as it forces its way out of the tube. Thus if 
no insects work at the flowers, they will be self-fertilised. This is 
apparently what occurs at Mount Kembla in my garden, for every 
flower sets seed, and the plant springs up self-sown every year. In 
pollen taken from flowers in the first stage, very many grains had 
emitted short tubes from one, two, or three of the corners (Fig. 16). 
THYMELE. 
Pimelea ligustrina, Labill., var. hypericina.—In a paper in the 
Journal of the Linnean Society (3) Mr. J. C. Willis describes the 
method of fertilisation of P. decussata, R. Br. (P. flava in “ Flora 
Aust.”). I find that in the species under notice fertilisation is 
effected in much the same way. The flowers grow in a dense 
globular head, those on the circumference opening first. The flower 
is 13 mm. long, with 4 perianth lobes 8 mm. across, the opening of 
the tube being 15 mm. The anthers, two in number, stand out 
at an angle from each other and shed pollen freely, the style and 
stigma being at this time hidden within the tube. Then the style 
grows out so that the stigma is about 2 mm. above the opening of 
the tube and about the same distance below the anthers. These 
now diverge slightly, but not to the degree observed in P. decussata. 
The flowers are sweet-scented in the daytime, but not at night. 
Insects visiting them are mostly those with long proboscides, and 
they probe the blossoms one after another, receiving pollen from 
the mature anthers and placing it on the mature stigma. They 
visit head after head on the same plant, and then fly to other 
plants, so that cross fertilisation is almost certain. From the posi- 
tion of the anthers when the stigma is mature, any flower is not 
* Jt however occurs in Queensland and North Australia.—(Ep.) 
