the so-called "germinal epithelium," but are the direct descendants of cells that have never lost their germ-cell characteristics.[1]
Artificial parthenogenesis is also easily produced among the low metazoa, and this has been known for a long time. 0. and E. Hertwig published in 1887 their experiments, which showed that various poisons might artificially induce segmentation in the eggs of echinoderms. Loeb carried the process further in 1899 and produced plutei from the unfertilized eggs of the sea urchins by raising the concentration of the sea water. "It was immaterial which substance was used to raise the concentration of the sea water, except for the fact that no substance could be used that injured the eggs too much."[2] Parthenogenesis may be initiated by such a variety of substances as chloroform, urea, sugar, salts and certain acids; but only in low forms of life. At least it is the legitimate inference that it is much easier to produce this effect among the simpler invertebrates. Experiments of this sort on the eggs of fishes are noticeable by their absence. Experiments on the eggs of frogs and Petromyzon have not, as far as I know, resulted in carrying the process beyond the segmentation stage. The experiments on insects which will be discussed under that heading have only a doubtful bearing on the present point, since parthenogenesis is among them more or less of a normal process.
Mollusks
Modification experiments on mollusks have been relatively rare in comparison with the lower invertebrates. This must be, to some extent, due to the practical difficulties which the experimenter would meet on attempting to modify an animal enclosed in a shell.
There is, however, much evidence to show that mollusks may be greatly influenced by their surroundings, at least as regards size. This especially concerns the question of the sizes of snails in relation to the extent of the media in which they are forced to live. The manner in which the volume of water affects the snail's growth is a matter of dispute and does not interest us here. The question for us is, how much are they affected? The answer is, that after about two months the snails grown under optimum environment were more than three times the length, or even five times the length, of those grown under the least favorable conditions.[3] Compare this with similar experiments on plants where the extreme variations artificially induced may be ten or twenty fold, and also with the experiments on the higher vertebrates where extreme variations resulting from any experimentally pre-
