Page:History of botany (Sachs; Garnsey).djvu/416

396 flower with those of Tournefort, who was a decided opponent of the doctrine of sexuality in plants. The parts of the flower are hastily described, figures are given of some forms of pollen-grains, and the notion that the style is a tube receives some apparent confirmation from the experiment of drawing water through the style of a lily. The view that the pollen is not an excrement, as Tournefort and Malpighi had maintained, is defended partly by arguments which prove nothing, for instance, by the erroneous assertion that the anthers are always so disposed that the extremity of the pistil must necessarily receive their dust. The only proof offered for the fact that seeds are infertile if deprived of the cooperation of the pollen, is a very hasty account of some experiments with maize and Mercurialis. The result of these experiments, as well as some other remarks of Geoffrey, remind us of the text of Camerarius' letter to an extent which mere accident will hardly account for. If Geoffroy really made these experiments, which is open to some doubt, yet they were made fifteen years later than those of Camerarius, who did make the same experiments among others and has described them better. Geoffroy next endeavours to show how the pollen effects the fertilisation, and offers two views on the subject; first, that the dust contains much sulphur and is decomposed on the pistil, the more subtle parts forcing their way into the ovary, where they set up a fermentation and cause the formation of the embryo ; the second view is, that the pollen-grains already contain the embryos, which find their way into the seeds and are there hatched. This is Morland's notion, who however is not mentioned. Geoffroy considers the latter to be the more probable hypothesis, chiefly because no embryo is found in the ovule before fertilisation, and also because the seed of the bean has an orifice (the micropyle); it does not occur to him that these facts speak as much for the first as for the second view.

Enough has been produced to show that Morland and