=17. Physiology and pathology.= The effect of ecology in emphasizing the intrinsically close connection between physiology and morphology has already been mentioned. Its influence in normalizing the former by forcing it into the field as the place for experiment, and by directing the chief attention to the plant as an organism rather than a complex of organs, is also rapidly coming to be felt. Ecology will doubtless exert a corrective influence upon pathology in the near future. This is inevitable as the latter ceases to be the merely formal study of specific pathogenic organisms, and turns its attention to the cause of all abnormality, which is to be found in the habitat, whether this be physical, as when the water-content is low, or biotic, when a parasitic fungus is present. The relative ease with which specific diseases can be studied has helped to obscure the essential fact that the approach to pathology must be through physiology. Much indeed of the observational physiology of the laboratories has been pathology, and it will be impossible to draw a clear line between them until precise experiment in the habitat has come into vogue. =18. Experimental evolution.= As a result of the extremely fragmentary character of the geological record, nothing is more absolute than that there can be no positive knowledge of the exact origin of a form or species, except in those rare cases of the present day, where the whole process has taken place under the eye of a trained observer. The origin of the plant forms known at present must forever lie without the domain of direct knowledge. If it were possible, by a marvel of ingenuity and patience to develop experimentally _Myosurus_ from _Selaginella_, this would not be absolutely conclusive proof that _Myosurus_ was first derived in this way. When all is said, however, this would be the very best of presumptive evidence. It must also be recognized that this is the nearest to complete proof that we shall ever attain, and with this in mind it becomes apparent at once that evidence from experiment is of paramount importance in the study of evolution (the origin of species).