There is no doubt that the influences of the environment must always have a powerful effect upon the soma of the individual, but we have seen, in the case of Alpine plants and of galls, how very far this effect may go without leaving any trace in the germ-plasm. LECTURE XXXII INFLUENCE OF ISOLATION ON THE FORMATION OF SPECIES Introduction--Isolated regions are rich in endemic species--Is isolation a condition in the origin of species?--Moriz Wagner, Romanes--'Amiktic' local forms, the butterflies of Sardinia, of the Alps, and of the Arctic zone--Periods of constancy and periods of variation in species--Amixia furthered by germinal selection--The thrushes of the Galapagos Islands--The intervention of sexual selection--Humming-birds--Central American thrushes--Weaver-birds of South Africa--Papilionidæ of the Malay Archipelago--Natural selection and isolation--Snails of the Sandwich Islands--Influences of variational periods--Comparison with the edible snail and with the snail fauna of Ireland and England--Changed conditions do not always give rise to variation--Summary. In an earlier lecture I endeavoured to show, by means of Darwinian arguments and examples, how important for every species, in relation to its transmutation, is the companionship of the other species which live with it in the same area. We saw that the 'conditions of life' operated as a determining factor in the composition of an animal and plant association quite as momentously as any climatic conditions whatsoever, and, indeed, Darwin rated the influences of vital association even more highly, and attributed to them an even greater power of evoking adaptation than he granted to the physical conditions of life. We are, therefore, prepared to recognize that even the transference of a species to a different fauna or flora may cause it to vary, and this occurs when the species gradually extends the area of its distribution, so that it penetrates into regions which contain a materially different association of forms of life. But these migrations are not necessarily only gradual, that is, due to the slow extension of the original area of distribution in the course of generations as the species increases in numbers; they may also occur suddenly, when isolated individuals or small companies of a species transcend in some unusual manner the natural boundaries of the old area, and reach some distant new region in which they are able to thrive.