regions in a greater degree than proportionally to the difference of climates on each side. Thus great chains of mountains, spaces of sea between islands and continents, even great rivers and deserts. In fact the amount (of) difference in the organisms bears a certain, but not invariable relation to the amount of physical difficulties to transit1.

There are some curious exceptions, namely, similarity of fauna of mountains of Europe and N. America and Lapland. Other cases just (the) reverse, mountains of eastern S. America, Altai (?), S. India (?)2: mountain summits of islands often eminently peculiar. Fauna generally of some islands, even when close, very dissimilar, in others very similar. [I am here led to observe one or more centres of creation3.]

The simple geologist can explain many of the foregoing cases of distribution. Subsidence of a continent in which free means of dispersal, would drive the lowland plants up to the mountains, now converted into islands, and the semi-alpine plants would take place of alpine, and alpine be destroyed, if mountains originally were not of great height. So we may see, during gradual changes of climate on a continent, the propagation of species would vary and adapt themselves to small changes

1 Note in the original, "Would it be more striking if we took animals, take Rhinoceros, and study their habitats?"

Note by Mr A. R. Wallace. "The want of similarity referred to, is, between the mountains of Brazil and Guiana and those of the Andes. Also those of the Indian peninsula as compared with the Himalayas. In both cases there is continuous intervening land.

"The islands referred to were, no doubt, the Galapagos for dissimilarity from S. America; our own Islands as compared with Europe, and perhaps Java, for similarity with continental Asia."

The arguments against multiple centres of creation are given in the Origin, Ed. i. p. 352, vi. p. 499.

In the Origin, Ed. i. p. 366, vi. p. 516, the author does not give his views on the distribution of alpine plants as original but refers to Edward Forbes' work (Geolog. Survey Memoirs, 1846). In his autobiography, Darwin refers to this. "I was forestalled" he says, "in only one important point, which my vanity has always made me regret." (Life and Letters, i. p. 88.)

causing much extermination1. The mountains of Europe were quite lately covered with ice, and the lowlands probably partaking of the Arctic climate and Fauna. Then as climate changed, arctic fauna would take place of ice, and an inundation of plants from different temperate countries (would) seize the lowlands, leaving islands of arctic forms. But if this had happened on an island, whence could the new forms have come,-here the geologist calls in creationists. If island formed, the geologist will suggest (that) many of the forms might have been borne from nearest land, but if peculiar, he calls in creationist,— as such island rises in height &c., he still more calls in creation. The creationist tells one, on a (illegible) spot the American spirit of creation makes Orpheus and Tyrannus and American doves, and in accordance with past and extinct forms, but no persistent relation between areas and distribution, GeologicoGeograph.-Distribution.

1 (The following is written on the back of a page of the Ms.) Discuss one or more centres of creation: allude strongly to facilities of dispersal and amount of geological change: allude to mountain-summits afterwards to be referred to. The distribution varies, as everyone knows, according to adaptation, explain going from N. to S. how we come to fresh groups of species in the same general region, but besides this we find difference, according to greatness of barriers, in greater proportion than can be well accounted for by adaptation. (On representive species see Origin, Ed. i. p. 349, vi. p. 496.) This very striking when we think of cattle of Pampas, plants (?) &c. &c. Then go into discussion; this holds with 3 or 4 main divisions as well as the endless minor ones in each of these 4 great ones: in these I chiefly refer to mammalia &c. &c. The similarity of type, but not in species, in same continent has been much less insisted on than the dissimilarity of different great regions generically: it is more striking.

(I have here omitted an incomprehensible sentence.) Galapagos Islands, Tristan d'Acunha, volcanic islands covered with craters we know lately did not support any organisms. How unlike these islands in nature to neighbouring lands. These facts perhaps more striking than almost any others. [Geology apt to affect geography therefore we ought to expect to find the above.] Geological-geographical distribution. In looking to past times we find Australia equally distinct. S. America was distinct, though with more forms in common. N. America its nearest neighbour more in common, -in some respects more, in some less allied to Europe. Europe we find (?) equally European. For Europe is now part of Asia though not (illegible). Africa unknown,-examples, Elephant, Rhinoceros, Hippopotamus, Hyaena. As geology destroys geography we cannot be surprised in going far back we find Marsupials and Edentata in Europe: but geology destroys geography.

Isolation

Now according to analogy of domesticated animals let us see what would result. Let us take case of farmer on Pampas, where everything approaches nearer to state of nature. He works on organisms having strong tendency to vary: and he knows (that the) only way to make a distinct breed is to select and separate. It would be useless to separate the best bulls and pair with best cows if their offspring run loose and bred with the other herds, and tendency to reversion not counteracted; he would endeavour therefore to get his cows on islands and then commence his work of selection. If several farmers in different rincons1 were to set to work, especially if with different objects, several breeds would soon be produced. So would it be with horticulturist and so history of every plant shows; the number of varieties2 increase in proportion to care bestowed on their selection and, with crossing plants, separation. Now, according to this analogy, change of external conditions, and isolation either by chance landing (of) a form on an island, or subsidence dividing a continent, or great chain of mountains, and the number of individuals not being numerous will best favour variation and selection3. No doubt change could be effected in same country without any barrier by long continued selection on one species: even in case of a plant not capable of crossing would easier get possession and solely

1 Rincon in Spanish means a nook or corner, it is here probably used to mean a small farm.

2 The following is written across the page: "No one would expect a set of similar varieties to be produced in the different countries, so species different."

3 (The following passage seems to have been meant to follow here.) The parent of an organism, we may generally suppose to be in less favourable condition than the selected offspring and therefore generally in fewer numbers. (This is not borne out by horticulture, mere hypothesis; as an organism in favourable conditions might by selection be adapted to still more favourable conditions.)

Barrier would further act in preventing species formed in one part migrating to another part.

occupy an island'. Now we can at once see that (if) two parts of a continent isolated, new species thus generated in them, would have closest affinities, like cattle in counties of England: if barrier afterwards destroyed one species might destroy the other or both keep their ground. So if island formed near continent, let it be ever so different, that continent would supply inhabitants, and new species (like the old) would be allied with that continent. An island generally very different soil and climate, and number and order of inhabitants supplied by chance, no point so favourable for generation of new species?,especially the mountains, hence, so it is. As isolated mountainsformed in a plain country (if such happens) is an island. As other islands formed, the old species would spread and thus extend and the fauna of distant island might ultimately meet and a continent formed between them. No one doubts continents formed by repeated elevations and depressions3. In looking backwards, but not so far that all geographical boundaries are destroyed, we can thus at once see why existing forms are related to the extinct in the same manner as existing ones are in some part of existing continent. By chance we might even have one or two absolute parent fossils.

The detection of transitional forms would be rendered more difficult on rising point of land.

The distribution therefore in the above enumer

1 (The following notes occur on the back of the page.) Number of species not related to capabilities of the country: furthermore not always those best adapted, perhaps explained by creationists by changes and progress. (See p. 34, note 1.)

Although creationists can, by help of geology, explain much, how can he explain the marked relation of past and present in same area, the varying relation in other cases, between past and present, the relation o different parts of same great area. If island, to adjoining continent, if quite different, on mountain summits,-the number of individuals not being related to capabilities, or how &c.—our theory, I believe, can throw much light and all facts accord.

2 See Origin, Ed. i. p. 390, vi. p. 543.

On oscillation see Origin, Ed. i. p. 291, vi. p. 426.

ated points, even the trivial ones, which on any other (theory?) can be viewed as so many ultimate facts, all follow (in) a simple manner on the theory of the occurrence of species by (illegible) and being adapted by selection to (illegible), conjoined with their power of dispersal, and the steady geographicogeological changes which are now in progress and which undoubtedly have taken place. Ought to state the opinion of the immutability of species and the creation by so many separate acts of will of the Creator1.

(From the back of Ms.) Effect of climate on stationary island and on continent, but continent once island. Moreover repeated oscillations fresh diffusion when non-united, then isolation, when rising again immigra tion prevented, new habitats formed, new species, when united free immigration, hence uniform characters. Hence more forms (on ?) the island. Mountain summits. Why not true species. First let us recall in Part I, conditions of variation: change of conditions during several generations, and if frequently altered so much better [perhaps excess of food] Secondly, continued selection [while in wild state]. Thirdly, isolation in all or nearly all, as well to recall advantages of.

[In continent, if we look to terrestrial animal, long continued change might go on, which would only cause change in numerical number (proportions): if continued long enough might ultimately affect all, though to most continents (there is) chance of immigration. Some few of whole body of species must be long affected and entire selection working same way. But here isolation absent, without barrier, cut off such (illegible). We can see advantage of isolation. But let us take case of island thrown up by volcanic agency at some distances, here we should have occasional visitants, only in few numbers and exposed to new conditions and (illegible) more important,-a quite new grouping of organic beings, which would open out new sources of subsistence, or (would) control (?) old ones. The number would be few, can old have the very best opportunity. (The conquest of the indigenes by introduced organisms shows that the indigenes were not perfectly adapted, see Origin, Ed. i. p. 390.) Moreover as the island continued changing,-continued slow changes, river, marshes, lakes, mountains &c. &c., new races as successively formed and a fresh occasional visitant.

If island formed continent, some species would emerge and immigrate. Everyone admits continents. We can see why Galapagos and C. Verde differ (see Origin, Ed. i. p. 398)], depressed and raised. We can see from this repeated action and the time required for a continent, why many more forms than in New Zealand (see Origin, Ed. i. p. 389 for a comparison between New Zealand and the Cape) no mammals or other classes (see however, Origin, Ed. i. p. 393 for the case of the frog). We can at once see how it comes when there has been an old channel of migration,-Cordilleras; we can see why Indian Asiatic Flora,-[why species] having a wide range gives better chance of some arriving at new points and being selected, and adapted to new ends. I need hardly remark no necessity for change.