Page:Popular Science Monthly Volume 63.djvu/20

16 once by variation and by cross-fertilization. To what peculiarities of substance or structure symbasis is due we have as yet no intimation, but the same might have been said of gravitation and many other properties of matter for which names have proved useful, as well as of growth, irritability, and similarly unexplained attributes of protoplasm.

Variations do not appear and are not selected or accumulated merely because of their usefulness or desirability with reference to environment, but useless or even injurious characters may be adopted as a means of evolutionary movement. Specialization in the sense of extreme accentuation of characters is often harmful and therefore not to be ascribed to adaptation. The influence of natural selection increases with the nicety of adjustment already attained, or as the range of permissible variation is narrowed. Adaptive specializations also commonly imply a narrow dependence on external conditions, and thus give no assurance of permanence for the type; they are more common on the side-twigs of life than on the main branches. Evolution is both accelerated and retarded by narrow selection or segregation; accelerated if the motion be estimated on the basis of a single character; retarded if the organism be viewed as a whole. Normal evolutionary progress does not go forward on the line of a single character, but requires the accumulation of many variations to maintain the structural coordination and functional cooperation of parts. External modifications require less coordination than internal, and are often exaggerated far beyond the requirements of use, and beyond the limits of developmental welfare.

Organic change and diversity inside the species are necessary and universal, but species and higher organic groups decline and become extinct if their variations become limited to non functional parts and do not provide, as it were, the facilities by which adjustment to changing environment may be maintained. Nevertheless, fitness for the environment is only one aspect of the evolutionary problem; adaptation is an incident and not a cause of evolutionary progress. Results commonly ascribed to selection are due to the normal motion of organic groups. Environment, including natural selection, segregation,