[Illustration: GRASSLAND AND TREE VEGETATION. All over the world there is a contest between grassland and tree vegetation for dominance wherever both occur. In this thorn veld in Natal the struggle is particularly keen. (After Bews. Courtesy of Brooklyn Botanic Garden.)] Many other species now existing are to be viewed only as relics of a bygone, often much more continuous and more widespread dispersal. In the persimmons, of which over a hundred different kinds are now known, the original distribution covered all North America, all Europe and Asia, except the Scandinavian Peninsula, all of Africa, northern Australia and South America. To-day the genus is restricted mostly to tropical Asia, southern Africa, northern South America, and in North America it has dwindled to a handful of species confined to the region south of the Great Lakes and generally east of the Mississippi. In Mexico and south of it the genus is better represented. But with the persimmon, as in so many other types, our existing species are the remnants of preexisting ones, without a knowledge of which their present dispersal would be impossible to explain. It is easy to reason from the foregoing that widely separated but related species of plants are all either very ancient, or directly descended from ancient ancestors, and that other things being equal a widely dispersed species is older than one with a restricted distribution. It is most certainly true that relics are unquestionably very ancient, and do actually represent the last outposts of a preexisting condition. But many isolated plants are relatively very new, so far as their immediate origin is concerned, as witness the hosts of young species of the daisy family, some of which have spread scarcely at all from their obvious point of origin. In judging of the distribution of even the commonest tree or shrub of our woodlands there are these links with the past, as well as response to present conditions, to be weighed if we are to understand the story aright.