These essays are profoundly interesting, and will supply food for thought for many years, but it must be admitted that Weismann has gradually been led away to speculations of the most elaborate kind, built upon the most flimsy substratum of fact. There is evidence of this tendency even in his early writings, but in his later essays, especially his recent work on “The Germ Plasm: a Theory of Heredity,”[10] the speculative part quite overpowers the rest. As those who are interested in heredity will probably read his works with the greatest attention, I have ventured in an appendix to make clear what in his works may, in my opinion, safely be looked upon as speculative rather than legitimate, or even provisional, generalisation, and what, therefore, may be altogether omitted from the study of a practical problem such as that with which we are concerned. _Are Acquired Characters transmitted?_ The practical issues which both Galton and Weismann have raised cannot, however, be underestimated, and, in respect to the non-inheritance of acquired characters, the mass of modern thinkers may already be said to have given their allegiance to the views of those two thinkers. But we are living in times when mere authority is at a discount, and we may well demand the facts for ourselves. The point which we are inclined to question is one as to which a doubt was often present in Darwin’s mind. Granted that selection is a factor, we are inclined to believe that the transmission of acquired characters must also take place, at any rate, to some extent. In attempting to decide this question upon the facts themselves we may take two lines of research. In the first place, we may examine every case of racial change, the production of new or different parts, the development of a new instinct, or the degeneration or loss of parts or instincts present at some past epoch. If in every case we are not compelled to exclude natural selection, and if in every case that we can directly and experimentally follow, selection is the outstanding factor, then there is strong presumptive evidence that racial change is caused by selection and not by the inheritance of acquired characters. In the second place, we may artificially induce the acquisition of some character, and notice whether this is transmitted; if it is not, then the general operation of this kind of transmission is rendered very doubtful. _Many Cases of Supposed Transmission to be explained by Selection._