It may be of some interest to point out that we do not need to make any definite assumption concerning the mechanism of evolution and that we may yet be able to account for the fact that the surviving organisms are to all appearances harmonious. The writer pointed out that of all the 100,000,000 conceivable crosses of teleost fish (many of which are possible) not many more than 10,000, _i. e._, about one-hundredth of one per cent., are able to live and propagate. Those that live and develop are free from the grosser type of disharmonies, the rest are doomed on account of a gross lack of harmony of the parts. These latter we never see and this gives us the erroneous conception that harmony or “design” is a general character of living matter. If anybody wishes to call the non-viability of 99-99/100 per cent. of possible teleosts a process of weeding out by “natural selection” we shall raise no objection, but only wish to point out that our way of explaining the lack of design in living nature would be valid even if there were no theory of evolution or if there had never been any evolution. 3. v. Uexküll is perfectly right in connecting the problem of design in an organism with Mendelian heredity. The work on Mendelian heredity has shown that an extremely large number of independently transmissible Mendelian factors help to shape the individual. It is not yet proven that the organism is nothing but a mosaic of Mendelian factors, but no writer can be blamed for considering such a possibility. If we assume that the organism is nothing but a mosaic of Mendelian characters it is difficult indeed to understand how they can force each other into a harmonious whole[6]; even if we make ample allowance for the law of chance and the corresponding wastefulness in the world of the living. But it is doubtful whether this idea of the rôle of Mendelian factors is correct. The facts of experimental embryology strongly indicate the possibility that the cytoplasm of the egg is the future embryo (in the rough) and that the Mendelian factors only impress the individual (and variety) characters upon this rough block. This idea is supported by the fact that the first development--in the sea urchin to the gastrula stage inclusive--is independent of the nucleus, which is the bearer of the Mendelian factors. Not before the skeleton or mesenchyme is formed in the sea urchin egg is the influence of the nucleus noticeable. This has been shown in the experiments of Boveri in which an enucleated fragment of an egg was fertilized with a spermatozoön of a foreign species. If this is generally true, it is conceivable that the generic and possibly also the species characters of organisms are determined by the cytoplasm of the egg and not by the Mendelian factors.