On the other hand, not only is there no experimental evidence of a germinal change by positive acquisition, that is, by the addition of genes, but it is hard to conceive how such a change could come about. “At first,” admits Bateson, “it may seem rank absurdity to suppose that the primordial form or forms of protoplasm could have contained complexity enough to produce the divers types of life.” “But,” he asks, “is it easier to imagine that these powers could have been conveyed by extrinsic addition? Of what nature could these additions be? Additions of material can not surely be in question. We are told that salts of iron in the soil may turn a pink hydrangea blue. The iron cannot be passed on to the next generation. How can iron multiply itself? The power to assimilate iron is all that can be transmitted. A disease-producing organism like the pebrine of silkworms can in a very few cases be passed on through the germ cells. But it does not become part of the invaded host, and we can not conceive it taking part in the geometrically ordered processes of segregation. These illustrations may seem too gross; but what refinement will meet the requirements of the problem, that the thing introduced must be, as the living organism itself is, capable of multiplication and of subordinating itself in a definite system of segregation?” (_Heredity_, Smithson. Inst. Rpt. for 1915, p. 373.) Nor can we agree with Prof. T. H. Morgan’s contention that the foregoing difficulty of Bateson has been solved by the discovery of the chromosomal mutation. All unbalanced chromosomal mutants are subnormal in their viability and vitality, not to speak of their marked sterility. Haploidy represents a regressive, rather than a progressive, step. The triploid mutant is sterile. The tetraploid race of Daturas is inferior in fertility to the normal diploid plant. The origin of balanced tetraploidy from diploidy must be presumed, since it has never been observed. Moreover, tetraploidy represents only quantitative, and not qualitative, progress. The increased mass of the nucleus produces an enlargement of the cytoplasm, the result of which is giantism. This effect, however, is not specific; for giant and normal races possessing each the same number of chromosomes are known to exist in nature. Hence giantism may be due to other causes besides chromosomal duplication. The only effect of this doubling is a reinforcement and intensification of the former effect of the genetic factors, their specificity remaining unchanged. Double doses are substituted for single doses of the factors, but nothing really new is added. Morgan himself recognizes that this mere repetition of identical genes is insufficient, and that their multiplication must be qualitative as well as numerical, to answer the specifications of a progressive step in evolution. Hence he suggests that the chromosomal mutation is subsequently supplemented by appropriate factorial mutation.