Chapter 7 Chapter 4 Natural Selection: Survival of the Fittest 2

Extinct by natural selection This subject will be dealt with at length in a chapter of the Geology; but because it is closely connected with natural selection, it must be dealt with here.The operation of natural selection consists entirely in preserving variations which are advantageous in some respects, and thereby causing their continuation.As all living beings multiplied at a high rate in geometric proportions, every country was filled with living creatures; thus, the advantageous forms increased in number, so that the more unfavorable forms often decreased in number and became rare.Geology tells us that scarcity is a herald of extinction.We know that any form, of which only a few individuals remain, is very likely to be completely extinct upon great changes in the nature of the seasons, or a temporary increase in the number of its predators.We may further say that, as new forms are produced, many of the old forms must become extinct, unless we allow an infinite multiplication of species-like forms.Geology plainly tells us that the number of species endowed with species has not increased infinitely; we now try to explain why the number of species in the world has not increased infinitely.

We have seen that the species with the greatest number of individuals has the best chance of producing beneficial variations within any given period.We have already shown this, in the second chapter, by the facts showing that the common, widespread, or dominant species possess the greatest number of recorded varieties.Species with few individuals are therefore slow to vary or improve at any given period; and consequently, in the struggle for existence, they are met with blows by the modified and improved descendants of the common species. From these arguments, I think, it must necessarily follow that, as new species are formed by natural selection in the lapse of time, other species will become rarer and eventually become extinct.Those who struggle most fiercely against forms which are undergoing modification and improvement, certainly sacrifice the most.We have seen in the "Struggle for Existence" that closely allied forms--that is, some varieties of the same species, and some species of the same or closely related genera--by having nearly the same structure, constitution, and habits, generally They also fight each other most violently; and consequently, each new variety or species, in the process of forming, generally oppresses most closely its nearest kin, and tends to destroy them.We see the same process of extinction in our domestic productions, through man's selection of improved forms.We could give many wonderful instances, showing how rapidly new breeds of cattle, sheep, and other animals, and varieties of flowers, replaced the old and inferior ones.In Yorkshire we know from history that the ancient black cattle were replaced by longhorns, who were "swept away by the shorthorns, as if by some cruel plague" (I quote a Agricultural author's words).

trait divergence The principle which I mean by this term is of the utmost importance, and I believe it can explain several important facts.First, that varieties, even those most marked, though more or less species-like,--as in many cases it is often a difficult question as to how they should be classified--certainly relate to each other. The differences between species are far less than those between pure and definite species.Varieties, in my view, are species in the process of formation, what I have called incipient species.How are the smaller differences between varieties extended to larger differences between species?This process occurs frequently, and we may infer it from the fact that, in nature, countless species exhibit marked differences, while varieties--the supposed prototypes and parents of the future marked species--show subtle differences. and unclear differences.Mere chance (as we may call it) may cause a variety to differ from its parent in some character, and the offspring of this variety to differ from its parent in a still greater degree in the same character; but only This by no means explains why the differences shown between different species of the same genus are so common and great.

It has been my practice to seek an account of the matter from domesticated creatures.Here we see a similar situation.It must be admitted that races so distinct, as short-horn and Hereford cattle, race and carriage horses, and several breeds of pigeons, &c., were by no means merely the accidental accumulation of like variations in many successive generations. And produced.In practice, for example, a pigeon with a slightly shorter beak attracts the attention of one fancier; another pigeon with a slightly longer beak attracts the attention of another fancier; Under the well-known principle of liking the middle standard and only the extreme types", they all select and breed pigeons with increasingly longer beaks, or shorter and shorter beaks (this is actually how the tumbler sub-breeds are produced) .Also, we can imagine that, early in history, people in one nation or region wanted fast horses, while people elsewhere wanted strong and clumsy horses.The first differences may be very slight; but as time goes on, with the successive selection of quick horses on the one hand, and strong horses on the other, the differences will increase, so that two sub-breeds will form.Eventually, over centuries, these subbreeds became stable and distinct breeds.When the differences were great, the inferior horses of intermediate character, that is, horses which were neither too swift nor too strong, would not be used for breeding, and were gradually exterminated thereafter.Thus we see in the products of man the operation of the so-called principle of divergence, which gives rise to differences, at first only slight, then gradually increasing, so that there are differences in character between the breeds and with the common parent. Disagreement.

But, it may be asked, how can a similar principle be applied to nature?I believe it can be applied and be applied very effectively (though I did not know how to apply until a long time ago), because in short, the more divergent the offspring of any species are in structure, constitution, and habits, the more they will be in the natural composition. The more various places can be occupied, the more they can be multiplied in number. We can see this clearly in animals with simple habits.Take, for example, the carnivorous quadrupeds, which, wherever they can sustain themselves, have long since reached the average of fullness.If allowed to increase naturally in its numbers (without any change in the conditions of the region), it can only succeed in increasing the numbers of other animals by virtue of modified offspring to take the place now occupied: for example, among them Some become able to eat new kinds of prey, dead or alive; some can live in new places, climb trees, ford water, and some can perhaps reduce their carnivorous habits.The more divergent the descendants of the carnivores become in habits and structure, the more ground they can occupy.What applies to one animal applies to all animals at all times—that is, if they vary—without which natural selection can have no effect.The same is true with plants.Experiments have shown that if a piece of land is sown with only one species of grass, and at the same time a similar piece of land is sown with several species of grasses of different genera, more plants can grow on the latter piece of land than on the former, Harvest heavier hay weights.The same thing would happen if two plots of equal size were sown on one with one variety of wheat, and on the other with several varieties mixed together.If, therefore, any one species of grass were continuing to vary, and if the varieties were successively selected for, they would differ from one another, though in a small degree, like grasses of different species and genera, and the majority of the species would The individual, including its mutated offspring, can successfully live on the same soil.We know that every species and every variety of grass annually disperses innumerable seeds; and that they all endeavor, so to speak, to increase their numbers.Consequently, after thousands of generations, the most marked varieties of any one grass species will have the best chance of succeeding and increasing in number, thus repelling the less marked varieties; species rank.

The truth of the principle that great divergence of structure can sustain the greatest number of biological life has been seen in many natural cases.In a very small area, especially when open to free immigration, the struggle between individuals must be extremely violent, and there we can always find a great divergence of organisms.For example, I saw a meadow, three feet by four feet in size, exposed to exactly the same conditions for many years, and growing on it twenty species of plants, belonging to eighteen genera and eight orders, It can be seen how much these plants differ from each other.The same is true of plants and insects on small islands where the situation is the same; and the same is true of freshwater ponds.Farmers know that more food can be harvested by rotating plants of very different "orders": what occurs in nature may be called simultaneous crop rotation.Most of the animals and plants that live densely on any small piece of land are able to live there (assuming the land is free of any particular character), and, so to speak, make a hundredfold effort to live there; but, it can be seen , where the struggle is most acute, the interest of divergence of structure, and concomitant difference of habits and constitution, determines, as a general rule, that the creatures which compete most fiercely with each other, are those which belong to what we call heterogeneous and Creatures of different "orders".

The same principle can be seen in the naturalization of plants through the action of man.It might be supposed, that the plants capable of becoming naturalized in any land will generally be those species which are closely allied with the native plants; for native plants are generally regarded as specially created and adapted to their native land.Or it may also be supposed that the naturalized plants probably belong to only a few groups, which are specially adapted to certain points of the new country.But the actual case is quite different; and de Candolle, in his admirable great work, expressly observes that naturalized plants, when compared with the number of native genera and species, have far fewer new genera than There are many new species.To give an example, in the last edition of Dr. Asa Gray's Flora of the Northern United States, 260 species of naturalized plants are cited, belonging to 162 genera.From this we can see that these naturalized plants are of a highly divergent nature.Moreover, they are quite different from the native plants, since, out of the 162 naturalized genera, no less than 100 are non-native, thus greatly increasing the genera now living in the United States.

By examining the nature of the plants or animals which have been victorious in any country over the natives, and have become there naturalized, we may have a general idea of ​​how some of the natives must have been modified in order to outwit them. Inhabitants; we may at least deduce that divergence of structure to new genera differences is to their advantage. In fact, the divergence of the constitution of the organisms of the same place has the same interest as that of the physiological division of the organs of an individual--a subject which has been discussed at length by Milne Edwards.No physiologist doubts that a stomach devoted to the digestion of vegetable matter, or that of meat, can absorb the most nourishment from these substances.Therefore, in the general system of any piece of land, the more extensive and perfect the divergence of animals and plants into different habits of life, the greater the number of individuals who can live there.A group of animals with little divergence in structure can hardly compete with a group of animals with more complete divergence in structure.For example, the various species of marsupials in Australia may be divided into several groups which differ little from each other, and which, as Mr. Waterhouse and others have pointed out, vaguely represent carnivorous, ruminant, rodent-mammalian species, but it is doubtful whether they would be able to successfully compete with these well-developed orders.In the Australian mammals we see the divergence process at an early and incomplete stage of development.

The possible action of natural selection on the progeny of a common ancestor by divergence and extinction of characters From the above very compressed discussion, we may suppose that the more divergent in structure the descendants of any one species are, the more successful they are, and the more able they are to invade places occupied by other organisms.Let us now see how the principle of deriving this advantage from divergence of characters can work in connection with the principles of natural selection and extinction. A chart attached to this book can help us understand this more complicated issue.The species of one large genus in this country are represented by A to L; they are not supposed to be equally similar, as is generally the case in nature, and as indicated in the diagram by letters at different distances.I speak of a large genus, because it has been said in the second chapter that on average more species vary in large genera than in small genera; and the species which vary in large genera have a greater number of varieties.We also see that the most common and widest-ranging species are more varied than the rare and narrow-ranging species.Suppose A is a common, widespread, varied species, and that this species belongs to a large local genus.The divergent dashed lines of unequal length emanating from A represent its mutant offspring.These variations are supposed to be very slight, but very divergent in nature; they are supposed not to occur simultaneously, but often at long intervals; and they are supposed to persist for unequal lengths of time after their occurrence.Only those mutations of some interest are preserved, or naturally selected for.Here arises the importance of the principle that advantage can be gained by divergence of characters; for this will generally lead to the preservation and accumulation of the most divergent or divergent variations (indicated by the outer dashed lines) by natural selection.When a dotted line meets a horizontal line, a small number is indicated there, and it is assumed that a sufficient amount of variation has been accumulated to form a variety well marked and considered worthy of record in the taxonomic work.

The distance between the horizontal lines in the graph, representing a thousand or more generations.After a thousand generations, the hypothetical species (A) has produced two distinct varieties, named a1 and m1.The two varieties have generally remained under the same conditions under which their parents had varied, and the variability itself has been inherited; consequently they have the same tendency to vary, and generally have occurred nearly as well as their parents. Mutations.Again, these two varieties, being only slightly modified forms, tend to inherit the advantages of the parent (A), which made it more numerous than the native organism; The more general advantages of a genus which make it a large genus in its own region.All these conditions are favorable for the generation of new varieties.

At this point, if the two varieties are still capable of varying, the greatest divergence in their variation will generally be preserved during the next thousand generations.After this period, suppose that the variety a1 in the diagram has produced the variety a2, the difference between a2 and (A) is greater than the difference between a1 and (A), on the principle of divergence.Suppose m1 produces two varieties, m2 and s2, differing from each other and still more from their common parent (A).By the same procedure we may extend the process to any remote period; some varieties, after every thousand generations, produce but one, but others, under conditions of greater and greater variation, produce two or three. variants, and some cannot produce variants.Varieties, therefore, the modified descendants of the common parent (A), will generally continue to increase their numbers and diverge in character. In the form, it ends at 14,000 generations. But I must say here: I am not assuming that this process will proceed as regularly as in the diagram (although the diagram itself is somewhat irregular), it is not very regular, and it is not continuous, but more It is probable that each form remains the same for a long period of time, and then changes again.Nor do I suppose that the most divergent varieties will necessarily be preserved: an intermediate form may be able to persist for a long time, or may or may not produce more than one modified offspring: The nature of the position which is not fully occupied; and this is determined by infinitely complex relations.But, as a general rule, the more divergent in structure the descendants of any one species are, the more places they occupy, and the more their modified descendants multiply.In our diagrams the systematic lines are broken at regular intervals, where they are marked with lower-case numerals, which mark successive forms which have become sufficiently different to be classed as varieties.But such breaks are imaginary, and may be inserted anywhere, provided the length of the interval allows a considerable amount of divergent variation to accumulate. As all the modified offspring from a common, widespread species, belonging to a large genus, will often inherit jointly those advantages which made the parent successful in life, they generally increase both in number and in life. Ability to diverge in character: this is shown in the diagram by the dashed lines branching off from (A).The modified descendants from (A), and the more highly improved branches on the line of the system, will tend to take the place of the earlier and less improved branches, and thus destroy them; this is shown in the diagram by several Lower branches that do not reach the upper horizontal line are indicated.In some cases, no doubt, the process of variation has been confined to one systematic line, so that, though the divergent variation has increased quantitatively, the modified offspring have not increased in number.This can be shown if all the lines from (A) in the diagram are removed, leaving only the one from a1 to a10. The English racehorse and the English guide dog are similar, and their character is obviously Slowly diverging from the original species, neither giving off any new branches nor any new races. After 10,000 generations, it is assumed that species (A) produces three types a10, f10 and m10. Due to the divergence of characters in the past generations, they will be very different from each other and from the common ancestor, but they may not not equal.If we assume that the change between the two horizontal lines in the graph is extremely small, these three types may still be only very marked varieties; but we only need to assume that the process of change is more in steps or greater in magnitude. Turn these three types into dubious species or at least into definite species.This diagram, therefore, shows the steps from the smaller differences which distinguish varieties to the larger differences which distinguish species.Continuing the same process for more generations (as shown in the condensed and simplified diagram), we obtain eight species, denoted by lower case letters a14 to m14, all descended from (A) down.Thus, as I believe, as species multiplied, genera formed. In large genera there will probably always be more than one species which vary.In the diagram, I assume that the second species (1) has, in similar steps, produced, after ten thousand generations, two distinct varieties or species (w10 and z10), which are either varieties or species, It should be determined according to the assumed change amount indicated between the horizontal lines.After 14,000 generations, six new species n14 to z14 are assumed to have arisen.Species which, in any one genus, already differ widely in character from each other, will generally produce the greatest number of modified offspring; for they have the best chance in their natural composition of occupying new and widely divergent places: so in In the chart, I select the extreme species (A) and the near-extreme species (I) as the species with the greatest variation and having produced new varieties and new species.The other nine species in the original genus (indicated by capital letters) may have continued to produce unchanged offspring for long but unequal periods; this is shown in the diagram by upward dashed lines of unequal length. . But in the process of modification, as shown in the diagram, another principle, that of extinction, also plays an important part.For in every place where living beings abound, the operation of natural selection must be to select those forms which are more advantageous than others in the struggle for life, and the improved descendants of any one species will always have a tendency: at every stage of the system In order to expel and eliminate their ancestors and their original ancestors.It must be remembered that the struggle is generally most violent between those forms which are nearest to each other in habits, constitution, and structure.The forms intermediate between earlier and later states, that is, between the less and more improved states of the same species, and the original parent-species themselves, therefore, generally have unequivocal Many whole offshoots on the system line will be so extinct that they are overcome by later and improved offshoots.But if the modified descendants of a species enter a different country, or are quickly adapted to an entirely new country, where there is no struggle between the offspring and the ancestors, both may continue to exist go down. Assuming that the amount of variation represented by our diagram is considerable, species (A) and all earlier varieties will perish, and will be replaced by eight new species a14 to m14; and species (1) will be replaced by six new species (n14 to z14) instead. We can go further.The original species of the genus are supposed to be unequal to each other, as is generally the case in nature; species (A) is more closely related to B, C, and E than to the others; species (I) to G , H, K, and L are more closely related to the other species than to the other species, and assuming that (A) and (I) are common and widely distributed species, they must have been more closely related than most other species in the same genus. has several advantages.Their modified descendants, totaling fourteen species at fourteen thousand generations, have inherited a part of the same virtue: they have also been modified and improved in various ways at each stage of the system, so that in The natural composition of the areas in which they inhabit became adapted to many of the positions associated with them.It is very probable, therefore, that they will not only take the place of parent-species (A) and (I) and exterminate them, but also some of the original species which are nearest to the parent-species.Therefore, the original species that can be passed down to the 14,000th generation is extremely rare.We may assume that only one of the two species (E and F) most distantly related to the other nine original species (E and F) was able to transmit their descendants to the last stage of the system. In our chart, the number of new species descended from eleven original species is now fifteen.Due to the tendency of natural selection to diverge, the extreme amount of difference in character between a14 and z14 is far greater than the maximum amount of difference between the eleven stock species.Also, the relative distances between the new species are very different.Among the eight descendants descended from (A), a14, q14, and p14 are relatively close relatives because they were all recently branched from a10; b14 and f14 were branched from a5 in an earlier period, so differ in some degree from the above three species; finally O14, i14, m14 are closely related to each other, but differ greatly from the preceding five species because they diverged at the beginning of the process of variation. Depending on the difference, they can become a subgenus or a definite genus. The six descendants descended from (1) will form two subgenera or two genera.But as the original species (1) differed greatly from (A), and (I) stood almost at one extreme in the original genus, the six offspring from (I) were, by heredity only, similar to those from (A). ) from which the eight offspring diverged were quite different; furthermore, we suppose that the two groups of organisms continued to diverge in different directions.And the intermediate species connecting the original species (A) and (I), which is a very important point, except for (F), have also been completely extinct, and have left no descendants.The six new species descended from (I), and the eight new species descended from (A), must therefore be classified in very distinct genera, and may even be classified in distinct subfamilies. I believe, therefore, that two or more genera have descended, by variation, from two or more species of the same genus.These two or more parent-species may again be presumed to have descended from a species in an earlier genus.In our diagrams, this is represented by a dotted line under a capital letter, the branches of which converge downward to a point; this point represents a species which is the putative progenitor of several new subgenera or genera.The character of the new species F14 deserves some consideration, assuming that its character has not diverged significantly, and that it still retains the form of (F), unchanged or only slightly.In this case its affinities with the other fourteen new species were of a strange and distant character.As it is descended from a form between the parent-species (A) and (I), now supposed to be extinct and unknown, its characters probably intermediate in some degree between those of the two species. Between the two groups of descendants handed down.But the characters of these two groups have diverged from their parent-types, so that the new species (F14) is not directly between the parent-species, but between the parent-types of the two groups; every naturalist You can probably imagine this situation. In this diagram, the horizontal lines are assumed to represent a thousand generations, but they could also represent a million or more generations: it could also represent a part of the continuous stratum of the Earth's crust containing the remains of extinct organisms, as we have seen in In the chapter on Geology, this question will have to be dealt with, and, I think, we will then see what this diagram has to say about the affinities of extinct organisms,--whose often Belonging to the same order, family, or genus with living organisms now living, but often intermediate in character more or less between the various groups of living organisms now living; we can understand this fact, because extinct species exist in various groups In different distant eras, there were only minor divergences in the branches of the system line at that time. I see no reason to limit the process of variation now explained to the formation of genera.In the diagram, if we suppose the amount of variation represented by each successive group on the dotted line of divergence to be enormous, the forms marked a14 to p14, b14 and f14, and o14 to m14 form three very distinct groups. belongs to.We would also have two very distinct genera descended from (I), which were quite different from the descendants of (A).The two groups of the genus form two different families, or different orders, according to the amount of divergent variation shown in the diagram.These two new families or orders are descended from two species of the original genus, which are supposed to have descended from some older and unknown forms. We have seen that, everywhere, it is the species of the larger genera that most frequently occur in varieties, or incipient species.This is indeed a case to be expected; for natural selection, acting by the preponderance of one form over the others in the struggle for existence, acts principally on those forms which already have some predominance; A large group indicates that its species have inherited some common advantages from a common ancestor.The struggle for the production of new and modified offspring, therefore, takes place chiefly among all the great groups striving to increase in number.One large group will gradually overcome another large group, reducing its number, and thus reducing its chances of continuous variation and improvement. Dividing out and occupying many new positions, there is often a tendency to crowd out and eliminate the earlier, less improved subpopulations.Small and weak groups and subgroups eventually perish.Looking into the future, we can predict that the groups of organisms which are now large and victorious, and which are the least destroyed, that is, least stricken with extinction, will continue to increase for a long time to come.But which groups will win the final victory no one can predict; for we know of groups which were formerly highly developed, but are now extinct.Looking further into the future, we can also predict that, as the larger groups continue to multiply, a large number of smaller groups will eventually become extinct, leaving no modified offspring; Only a very small number of species can pass on their offspring to the distant future.I shall return to this subject in the chapter on "Classification," but I may do it here again, on this view, since only a very small number of more ancient species have descended to the present day, and since the same species All the descendants of the genus formed into a class, and we can then understand why, in each of the major divisions of the animal and vegetable kingdoms, there are so few classes at present.Although there are only a few mutated descendants of extremely ancient species, in the distant geological ages, there were also many genera, families, orders and classes distributed on the earth, and their prosperity was almost the same as it is today. On the Degree of Progressive Tendency of Biological Systems The operation of "Natural Selection" consists entirely in the preservation and accumulation of variations which are advantageous to each being, under all the organic and inorganic conditions of its life, during all its life.The final result of this is an ever-improving relationship of every living being to its external conditions.This improvement necessarily entails a gradual improvement in the constitution of the majority of living beings throughout the world.But here we are confronted with a very complicated problem, for what is meant by the progress of institutions has not yet been satisfactorily defined among naturalists.In vertebrates, the degree of intelligence and the closeness of structure to humans clearly indicate their progress.It may be supposed that the magnitude of the changes undergone by the parts and organs from embryonic development to maturity would seem to serve as a standard of comparison; After growing up, it becomes incomplete, so that the mature animal cannot be said to be higher than its larva.The criterion laid down by Von Baer, ​​which seems to be the most widely and best applicable, refers to the amount of differentiation of the parts of the same organism,--here, I should add, the adult state- —and the degree of specialization of their different functions; that is, what Milne Edwards calls the completeness of the physiological division of labour.But if we look at fish, for example, we can see how obscure the matter is: some naturalists rank among them the closest to amphibians, the shark, as the highest, while others rank the common The teleosts of the genus are ranked highest, since they are most strictly fish-like and least like other vertebrates.We may still more clearly see the obscurity of the matter in regard to plants, which of course do not contain the criterion of wisdom at all; At the same time, there are some botanists who think that plants in which the several organs of flowers are greatly varied and reduced in number are the highest, and this view is probably more reasonable. If we take the degree of differentiation and specialization of the several organs of mature organisms (here including the development of the brain for intellectual purposes) as a criterion of institutional highness, then natural selection will obviously point towards this criterion: because all organisms Scholars all admit that the specialization of organs is beneficial to organisms, because specialization can make the functions better; therefore, the accumulation of variations towards specialization is within the scope of natural selection.On the other hand, if we only remember that all organic beings strive to increase at a high rate, and to seize every unoccupied or not fully occupied place in the constitution of nature, we can see that natural selection is quite probable to gradually make an organism Appropriate to a situation where several organs would be superfluous or useless: in this case the degradation of the hierarchy of the system takes place.Whether or not there has been any progress in living organisms as a whole, from the most distant geological ages to the present, will be more conveniently discussed in the chapter "Geological Succession". 但是可以提出反对意见：如果一切生物既然在等级上都是这样倾向上升，为什么全世界还有许多最低等类型依然存在？在每个大的纲里，为什么有一些类型远比其他类型更为发达？为什么更高度发达的类型，没有到处取代较低等类型的地位并消灭它们呢？拉马克相信一切生物都内在地和必然地倾向于完善化，因而他强烈地感到了这个问题是非常难解的，以致他不得不假定新的和简单的类型可以不断地自然发生。现在科学还没有证明这种信念的正确性，将来怎么样就不得而知了。根据我们的理论，低等生物的继续存在是不难解释的；因为自然选择即最适者生存，不一定包含进步性的发展——自然选择只利用对于生物在其复杂生活关系中有利的那些变异。那末可以问，高等构造对于一种浸液小虫（infusorian anin lcule），以及对于一种肠寄生虫，甚至对于一种蚯蚓，照我们所能知道的，究有什么利益？如果没有利益，这些类型便不会通过自然选择有所改进，或者很少有所改进，而且可能保持它们今日那样的低等状态到无限时期。地质学告诉我们，有些最低等类型，如浸液小虫和根足虫（rhizopods），已在极长久的时期中，差不多保持了今日的状态。但是，如果假定许多今日生存着的低等类型，大多数自从生命的黎明初期以来就丝毫没有进步，也是极端轻率的；因为每一个曾经解剖过现今被列为最低等生物的博物学者们，没有不被它们的确系奇异而美妙的体制所打动。 如果我们看一看一个大群里的各级不同体制，就可以知道同样的论点差不多也是可以应用的；例如，在脊推动物中，哺乳动物和鱼类并存；在哺乳动物中，人类和鸭嘴兽并存；在鱼类中，鲨鱼和文昌鱼（Amphioxus ）并存，后一种动物的构造极其简单，与无脊椎动物很接近。但是，哺乳动物和鱼类彼此没有什么可以竞争的；哺乳动物全纲进步到最高级，或者这一纲的某些成员进步到最高级，并不会取鱼的地位而代之。生理学家相信，脑必须有热血的灌注才能高度活动，因此必须进行空气呼吸；所以，温血的哺乳动物如果栖息于水中，就必须常到水面来呼吸，很不便利。关于鱼类，鲨鱼科的鱼不会有取代文昌鱼的倾向，因为我听弗里茨·米勒说过，文昌鱼在巴西南部荒芜沙岸旁的唯一伙伴和竞争者是一种奇异的环虫（annelid）。哺乳类中三个最低等的目，即有袋类、贫齿类和啮齿类，在南美洲和大量猴子在同一处地方共存，它们彼此大概很少冲突。总而言之，全世界生物的体制虽然都进步了，而且现在还在进步着，但是在等级上将会永远呈现许多不同程度的完善化；因为某些整个纲或者每一纲中的某些成员的高度进步，完全没有必要使那些不与它们密切竞争的类群归于绝灭。在某些情形里，我们以后还要看到，体制低等的类型，由于栖息在局限的或者特别的区域内，还保存到今日，它们在那里遭遇到的竞争较不剧烈，而且在那里由于它们的成员稀少，阻碍了发生有利变异的机会。 最后，我相信，许多体制低等的类型现在还生存在世界上，是有多种原因的。在某些情形里，有利性质的变异或个体差异从未发生，因而自然选择不能发生作用而加以积累。大概在一切情形里，人对于最大可能的发展量，没有足够的时间。在某些少数情形里，体制起了我们所谓的退化。但主要的原因是在于这样的事实，即在极简单的生活条件下，高等体制没有用处——或者竟会有害处，因为体制愈纤细，就愈不容易受调节，就愈容易损坏。 再来看一下生命的黎明初期，那时候一切生物的构造，我们可以相信都是极简单的，于是可以问：身体各部分的进步即分化的第一步骤是怎样发生的呢？赫伯特·斯潘塞先生大概会答复说，当简单的单细胞生物一旦由于生长或分裂而成为多细胞的集合体时，或者附着在任何支持物体的表面时，他的法则“任何等级的同型单位，按照它们和自然力变化的关系，而比例地进行分化”，就发生作用了。但是，既没有事实指导我们，只在这一题目上空想，几乎是没有什么用处的。但是，如果假定，在许多类型产生以前，没有生存竞争因而没有自然选择，就会陷入错误的境地：生长在隔离地区内的一个单独物种所发生的变异可能是有利的，这样，全部个体就可能发生变异，或者，两个不同的类型就可能产生，但我在《绪论》将结束时曾经说过，如果承认我们对于现今生存于世界上的生物间的相互关系极其无知，并且对于过去时代的情形尤其如此，那末关于物种起源问题还有许多不能得到解释的地方，便不会有人觉得奇怪了。 性状的趋同 H． C．沃森先生认为我把性状分歧的重要性估计得过高了（虽然他分明是相信性状分歧的作用的），并且认为所谓性状趋同同样也有一部分作用。如果有不同属的但系近属的两个物种，都产生了许多分歧新类型，那末可以设想，这些类型可能彼此很接近，以致可以把它们分类在一个属里；这样，两个不同属的后代就合二而成为一属了。就大不相同的类型的变异了的后代来说，把它们的构造的接近和一般相似归因于性状的趋同，在大多数场合里都是极端轻率的。结晶体的形态，仅由分子的力量来决定，因此，不同的物质有时会呈现相同的形态是没有什么奇怪的。但就生物来说，我们必须记住，每一类型都是由无限复杂的关系来决定的，即由已经发生了的变异来决定的，而变异的原因又复杂到难于究诘，——是由被保存的或被选择的变异的性质来决定的，而变异的性质则由周围的物理条件来决定，尤其重要的是由同它进行竞争的周围生物来决定的，——最后，还要由来自无数祖先的遗传（遗传本身就是彷惶的因素）来决定，而一切祖先的类型又都通过同样复杂的关系来决定。因此，很难相信，从本来很不相同的两种生物传下来的后代，后来是如此密接地趋同了，以致它们的整个体制变得近乎一致，如果这种事情曾经发生，那么在隔离极远的地层里，我们就可以看到毫无遗传联系的同一类型会重复出现，而衡量证据正和这种说法相反。 自然选择的连续作用，结合性状的分歧，就能产生无数的物种的类型，华生先生反对这种说法。如果单就无机条件来讲，大概有很多物种会很快地适应于各种很不相同的热度和湿度等等；但我完全承认，生物间的相互关系更为重要；随着各处物种的继续增加，则有机的生活条件必定变得愈益复杂。结果，构造的有利分歧量，初看起来，似乎是无限的，所以能够产生的物种的数量也应该是无限的。甚至在生物最繁盛的地区，是否已经充满了物种的类型，我们并不知道；好望角和澳洲的物种数量如此惊人，可是许多欧洲植物还是在那里归化了。但是，地质学告诉我们，从第三纪早期起，贝类的物种数量，以及从同时代的中期起，哺乳类的数量并没有大量增加，或根本没有增加。那末，抑制物种数量无限增加的是什么呢？一个地区所能维持的生物数量（我不是指物种的类型数量）必定是有限制的，这种限制是由该地的物理条件来决定的；所以，如果在一个地区内栖息着极多的物种，那末每一个物种或差不多每一个物种的个体就会很少；这样的物种由于季节性质或敌害数量的偶然变化就容易绝灭。绝灭过程在这种场合中是迅速的，而新种的产生永远是缓慢的。想像一下一种极端的情况吧，假如在英国物种和个体的数量一样多，一次严寒的冬季或极干燥的夏季，就会使成千上万的物种绝灭。在任何地方，如果物种的数量无限增加，各个物种就要变为个体稀少的物种，两个稀少的物种，由于常常提到的理由，在一定的期间内所产生的有利变异是很少的；结果，新种类型的产生过程就要受到阻碍。任何物种变为极稀少的时候，近亲交配将会促其绝灭；作者们以为立陶宛的野牛（Aurochs）、苏格兰的赤鹿、挪威的熊等等的衰颓，皆由于这种作用所致。最后，我以为这里还有一个最重要的因素，即一个优势物种，在它的故乡已经打倒了许多竞争者，就会散布开去，取代许多其他物种的地位。得·康多尔曾经阐明，这些广为散布的物种一般还会散布得极广；结果，它们在若干地方就会取代若干物种的地位，而使它们绝灭，这样，就会在全世界上抑制物种类型的异常增加。胡克博士最近阐明，显然有许多侵略者由地球的不同地方侵入了澳洲的东南角，在那里，澳洲本地物种的类量就大大地减少了。这些论点究有多大价值，我还不敢说；但把这些论点归纳起来，就可知道它们一定会有在各地方限制物种无限增加的倾向。 本章提要 在变化着的生活条件下，生物构造的每一部分几乎都要表现个体差异，这是无可争论的；由于生物按几何比率增加，它们在某年龄、某季节或某年代，发生激烈的生存斗争，这也确是无可争论的；于是，考虑到一切生物相互之间及其与生活条件之间的无限复杂关系，会引起构造上、体质上及习性上发生对于它们有利的无限分歧，假如说从来没有发生过任何有益于每一生物本身繁荣的变异，正如曾经发生的许多有益于人类的变异那样，将是一件非常离奇的事。但是，如果有益于任何生物的变异确曾发生，那么具有这种性状的诸个体肯定地在生活斗争中会有最好的机会来保存自己；根据坚强的遗传原理，它们将会产生具有同样性状的后代。我把这种保存原理，即最适者生存，叫做“自然选择”。“自然选择”导致了生物根据有机的和无机的生活条件得到改进；结果，必须承认，在大多数情形里，就会引起体制的一种进步。然而，低等而简单的类型，如果能够很好地适应它们的简单生活条件，也能长久保持不变。 根据品质在相应龄期的遗传原理，自然选择能够改变卵、种籽、幼体，就像改变成体一样的容易。 在许多动物里，性选择，能够帮助普通选择保证最强健的、最适应的雄体产生最多的后代。性选择又可使雄体获得有利的性状，以与其他雄体进行斗争或对抗；这些性状将按照普遍进行的遗传形式而传给一性或雌雄两性。 自然选择是否真能如此发生作用，使各种生物类型适应于它们的若干条件和生活处所，必须根据以下各章所举的证据来判断。但是我们已经看到自然选择怎样引起生物的绝灭；而在世界史上绝灭的作用是何等巨大，地质学已明白地说明了这一点。自然选择还能引致性状的分歧；因为生物的构造、习性及体质愈分歧，则这个地区所能维持的生物就愈多，——我们只要对任何一处小地方的生物以及外地归化的生物加以考察，便可以证明这一点。所以，在任何一个物种的后代的变异过程中，以及在一切物种增加个体数目的不断斗争中，后代如果变得愈分歧，它们在生活斗争中就愈有成功的好机会，这样，同一物种中不同变种间的微小差异，就有逐渐增大的倾向，一直增大为同属的物种间的较大差异、或者甚至增大为异属间的较大差异。 我们已经看到，变异最大的，在每一个纲中是大属的那些普通的、广为分散的、以及分布范围广的物种；而且这些物种有把它们的优越性——现今在本上成为优势种的那种优越性——传给变化了的后代的倾向。正如方才所讲的，自然选择能引致性状的分歧，并且能使改进较少的和中间类型的生物大量绝灭。根据这些原理，我们就可以解释全世界各纲中无数生物间的亲缘关系以及普遍存在的明显区别。这的确是奇异的事情，——只因为看惯了就把它的奇异性忽视了——即一切时间和空间内的一切动物和植物，都可分为各群，而彼此关联，正如我们到处所看到的情形那样，——即同种的变种间的关系最密切，同属的物种间的关系较疏远而且不均等，乃形成区（section）及亚属；异属的物种间关系更疏远，并且属间关系远近程度不同，乃形成亚科、科、目、亚纲及纲。任何一个纲中的几个次级类群都不能列入单一行列，然皆环绕数点，这些点又环绕着另外一些点，如此下去，几乎是无穷的环状组成。如果物种是独立创造的，这样的分类便不能得到解释；但是，根据遗传，以及根据引起绝灭和性状分歧的自然选择的复杂作用，如我们在图表中所见到的，这一点便可以得到解释。 同一纲中一切生物的亲缘关系常常用一株大树来表示。我相信这种比拟在很大程度上表达了真实情况。绿色的、生芽的小枝可以代表现存的物种；以往年代生长出来的枝条可以代表长期的、连续的绝灭物种。在每一生长期中，一切生长着的小枝都试图向各方分枝，并且试图遮盖和弄死周围的新技和枝条，同样地物种和物种的群在巨大的生活斗争中，随时都在压倒其他物种。巨枝分为大枝，再逐步分为愈来愈小的枝，当树幼小时，它们都曾一度是生芽的小枝；这种旧芽和新芽由分枝来连结的情形，很可以代表一切绝灭物种和现存物种的分类，它们在群之下又分为群。当这树还仅仅是一株矮树时，在许多茂盛的小枝中，只有两三个小枝现在成长为大枝了，生存至今，并且负荷着其他枝条；生存在久远地质时代中的物种也是这样，它们当中只有很少数遗下现存的变异了的后代，从这树开始生长以来，许多巨枝和大枝都已经枯萎而且脱落了；这些枯落了的、大小不等的枝条，可以代表那些没有留下生存的后代而仅处于化石状态的全目、全科及全属。正如我们在这里或那里看到的，一个细小的、孤立的枝条从树的下部分叉处生出来，并且由于某种有利的机会，至今还在旺盛地生长着，正如有时我们看到如鸭嘴兽或肺鱼之类的动物，它们由亲缘关系把生物的两条大枝连络起来，并由于生活在有庇护的地点，乃从致命的竞争里得到幸免。芽由于生长而生出新芽，这些新芽如果健壮，就会分出枝条遮盖四周许多较弱枝条，所以我相信，这巨大的“生命之树”（Tree of Life）在其传代中也是这样，这株大树用它的枯落的枝条填充了地壳，并且用它的分生不息的美丽的枝条遮盖了地面。