For instance, in a plant-embryo roots are produced at the lower end under the influence of the soil and of gravity. But it is upon the specific nature of the protoplasm of different kinds of plants that the special shape of the whole root system depends: whether, for instance, the root system ramifies superficially or strikes deep into the soil; whether the rootlets grow quickly or slowly; in what fashion they fork, and whether or no they form special structures like bulbs. Thus, even from my point of view, explanation of the process of development requires the assumption of the existence of different kinds of germinal material in different kinds of organisms. These germinal substances must be possessed of an extraordinarily complex organisation, and must be able to react in specific fashion--that is to say, in a fashion different in each species--to all the slightest internal and external stimuli encountered from time to time as the organisation becomes formed by cell division. In this sense I agree with what Naegeli says: 'The egg-cells contain all actual specific characters as truly as the adult organisms; when they exist in the condition of eggs, organisms are as distinct from each other as in the adult condition. The species is present as truly in the fowl's egg as in the fowl, and the egg of a fowl differs as much from the egg of a frog as the fowl differs from the frog. Men, rodents, ruminants, invertebrates display more or less important and outwardly visible differences in constitution; so also the sexual cells to which they give rise, since they represent the rudiments of the future adults, must be different from each other in the constitution of the rudiments, although we are not yet able to prove these differences by observation.' In this assumption of a specific and highly-organized germinal substance with which a development begins, I agree with evolutionists; but in its details my conception is quite different from their conception. For I can ascribe to the germinal substance only such characters as are appropriate to the true nature of a cell, but I cannot ascribe to it the numerous characters that can come into existence only by the interrelations of many cells and the action of the environment. Haacke, in his recently-published book (_Gestaltung und Vererbung)_, has expressed a doubt that my conception of development is, after all, a preformational theory. 'For preformation,' he says, 'it is not necessary to imagine that the egg contains a miniature of the adult. If only, like Hertwig, one assumes to be present in the germinal material a prearrangement of qualitatively different idioblasts, one has steered into the harbour of preformation with all sails set.'