Page:Natural History Review (1861).djvu/464

452 was published in the Memoirs of the Academy of St. Petersburg, (VII. Ser. Tome I. No. 2.) He considers that Spallanzani's experiments are the only ones which have been well conducted, and admits that if he (Regel) could have procured perfect seeds by operating as Spallanzani did with the Hemp and the Water-melon, the doctrine of parthenogenesis might be considered established. Regel attempts to get rid of Spallanzani's authority, by suggesting either that his observations were not carefully conducted, or that Spallanzani did not speak the truth. The latter accusation appears to have no sort of foundation; at least the grounds for it given by Regel are of the weakest description.

In the course of a series of experiments on hybridization, Regel observed that the anthers of many plants are fully formed and contain perfect pollen some time before the opening of the flowers; from which he concludes that it is necessary, in all experiments on parthenogenesis, to cut off the anthers at a very early period, or otherwise impregnation may have taken place without the observer having any suspicion of it. He considers monæcious and diæcious plants to be especially likely to have led to deceptions, because sufficient care was not taken to examine each individual flower— a precaution which is necessary on account of the frequent occurrence of accidental anthers in the female flowers, and because monæcious and diæcious plants produce a greater quantity of pollen than hermaphrodites, and consequently accidental impregnation by wind or insects is highly probable. He considers the Hemp-plant, which has been so much employed for these experiments, to be especially ill-suited for them, on account of the abundance of its easily dispersible pollen, and of its universal cultivation. As therefore (he says) these precautions have not been taken by any trustworthy observer, and as his own observations have afforded only negative results, he considers it certain that the formation of a true embryo can only take place under the influence of impregnation.

Regel then proceeds to give an account of his observations upon the Cycadeæ, and especially upon Ceratozamia, and the result that he arrived at was, that no embryo can be formed without the influence of a pollen-tube, but that the growth of the embryo-sac and the production of endosperm, as also the formation of corpuscula in Gymnosperms, may take place independently of impregnation. The case of the Cycadcæ he considers very conclusive, as showing that the development in the embryo-sac can only proceed up to the point at which the stimulus to the commencement of the formation of an embryo must be given, and that where this stimulus is wanting, the germinal vesicle, even in vigorous seeds, exhibits no further development.

Regel subsequently proceeds to detail his experiments on Mercurialis annua, from which he concludes that the previous observers who have imagined that they have procured perfect seeds without impregnation, have overlooked the very frequent occurrence of anthers in the so-called female flowers; and with regard to the persistence