FLORA OF THE GALAPAGOS ISLANDS. 257 



less difficult to account for on the emergence theory than it seemed 

 when the Baur plants were studied some years ago. 



Let us consider theoretically what would be likely to happen to a 

 plant casually introduced in a pelagic archipelago, where no plant of its 

 particular atfinity had previously found its way. Let us suppose, for 

 instance, that the ancestral form of Euphorbia viminea reached Chatham 

 Island from the continent, and that conditions of seed-transportation 

 were such that subsequent seedings from the maiidaud would not be 

 likely to happen oftener on the average than once in thirty, fifty, or 

 perhaps one hundred years, — no unreasonable assumption. Now, it is 

 known from observation that a plant introduced into a new region can 

 overrun considerable territory and increase to thousands of individuals, 

 even in a shorter space of time. It is further likely that a plant estab- 

 lished upon such an island would be at once exposed to modifying 

 influences and tend toward the formation of a new race particularly 

 suited to its altered environment. Whether this were effected by direct 

 influence or by natural selection is not significant in this case. In the 

 interval between the first and second seeding the change would probably 

 be very slight and taxonomically imperceptible, but that some modifica- 

 tion would have taken place seems likely. It is clear that the second 

 and subsequent seedings of the same island by the same species from 

 the continent would tend by the infusion of pure stock to reclaim the 

 incipient insular variation to the typical continental form of the species. 

 But to see how great or rather how slight the influence of these later 

 seedings would be, it is only necessary to consider the numerical relation 

 of both forms. The descendants of the first immigrant misrht well have 

 increased to many thousands of (slightly altered) individuals before the 

 second seed arrived from the mainland. Thus the reclaiming influence 

 of this second immigrant would not be as one to one but as one against 

 thousands, that is to say, virtually nil. Of course, it might be thought 

 that as the first seed increased in a few years to thousands of individuals, 

 the second might do so likewise, so that their influence would after all 

 be not very unequal. This, however, could scarcely happen for two 

 reasons. In the first place the original seed would have found in the 

 island (before uninhabited by any near relative) a fresh terrain, giving 

 it an opportunity to multiply rapidly. The second immigrant, however, 

 would find its proper habitat on the island no longer free for settlement, 

 but largely if not entirely occupied by countless individuals of a very 

 nearly related stock. It could scarcely fail to cross repeatedly with the 

 insular form and quickly merge into it, the more so because the descend- 



VOL. xxxviii. — 17 



