Chapter 11 Chapter VII Objections to the Theory of Natural Selection 1

Longevity--Variations not necessarily simultaneous--Variations apparently of no immediate use--Progressive development--Characters of little functional importance are most stable--The inability of natural selection to account for the initial stages of useful structures-- Reasons for Interfering with the Acquisition of Useful Structures by Natural Selection--Grades of Structure with Variations in Functions--Development of Widely Different Organs in Members of the Same Class from a Common Root--Improbable Reasons for Large and Sudden Variations . I am going to devote this chapter to the various objections to my view, because this will make some of the earlier discussions clearer; but it is not necessary to deal with all objections, since many are Proposed by authors who have not put their heart into understanding the problem.A well-known German naturalist, for example, asserts that the weakest part of my doctrine is that I regard all living beings as imperfect; That is perfect; the fact that indigenous organisms have given way to invasive alien organisms in many parts of the world demonstrates that this is the case.Even if living beings were at any one time in the past perfectly adapted to the conditions of their life, they could not do so again when the conditions were changed, unless they themselves also changed; The number and types of species have undergone many changes.

A recent critic, with some flaunting of mathematical precision, has insisted that longevity is of great benefit to all species, so that a believer in natural selection "should put his phylogenetic tree" in such a way that all descendants are better than their ancestors. Longer life that way to line up!But a biennial plant or a lower animal, if distributed in a cold country, dies every winter; Couldn't critics consider this situation, recently discussed by Mr. E. Ray Lankester, who concluded that, so far as the extreme complexity of the problem permits, His judgment is that longevity is generally related to the standard of each species in the institutional hierarchy, and also to the amount expended in reproduction and ordinary activity.Most of these conditions may have been determined by natural selection.

It has been argued that during the last three or four thousand years the animals and plants of Egypt have, so far as we know, not changed, and so presumably no living things anywhere in the world have changed.But, as Mr. GH Lewes says, this argument goes too far, for the ancient domestic races inscribed on Egyptian monuments, or mummified, closely resembled those living today. , or even the same; yet all naturalists admit that these domestic races have arisen by variations of their original forms.The many animals which have remained unchanged since the beginning of the Glacial Ages may perhaps serve as a most powerful example, since they have been exposed to great changes in climate, and have migrated over great distances; in Egypt, on the contrary, so far as we know, The living conditions have been exactly the same for the past thousands of years.The fact that little or no change has occurred since the Ice Age has probably had some validity against those who believed in an innate and necessary law of development, but against the doctrine of natural selection, the survival of the fittest. No force, for the doctrine implies that variations or individual differences of a favorable nature are preserved only when they occur; but this is only possible under certain favorable circumstances.

The famous paleontologist Buron asked at the end of the German edition of his translation: According to the principle of natural selection, how can a variety survive side by side with its parent species?If both are adapted to slightly different habits or conditions of life, they will probably be able to survive together; if we combine the polymorphic species (whose variability seems to be of a peculiar nature), and temporal variations, such as size, skin Leukemia, etc., aside, the other more stable varieties, so far as I can find, generally inhabit different places--uplands or lowlands, dry or wet regions.Also, in those animals which roam over great distances, and which interbreed freely, the varieties seem generally to be confined to different regions.

Bronn also maintains that different species never differ in not only one character, but in many parts; and how, he asks, are many parts of an organization often simultaneously modified by variation and natural selection?But it is not necessary to imagine that all parts of any living being change simultaneously.The most marked variations, which are most suitable for any purpose, are, as before said, probably acquired by successive modifications, even slight ones, first in one part and then in another; for these are all transmitted together. Yes, so it seems that we have developed at the same time.The best answer to this objection is afforded by those domestic races which have been modified for some particular purpose chiefly by the power of human selection.Look at the racehorse and the carthorse, or the greyhound and the mastiff.Their whole bodily and even mental characteristics have been altered; but if we could trace every stage of their history of change--the last few stages can be traced--we shall not see Vast and simultaneous changes, but only to see first one part and then another slightly mutated and improved.Even when man selects for a single character--the cultivated plant furnishes the best example of this--we must see that although the part--whether it be flower, fruit, or foliage, is greatly If the ground is changed, almost everything else is changed in some way.This can be attributed partly to the principle of correlated growth and partly to so-called spontaneous variation.

A still more serious objection has been raised by Brone, and more recently by Broca, that there are many characters which do not seem to be of any use to their owners, and therefore cannot be affected by natural selection.Bronn cites as illustrations the length of the ears and tails of the different species of hares and mice, the intricate folds of enamel in the teeth of many animals, and many similar cases; This issue has been discussed in the commended paper.He acknowledged the influence of natural selection, but he maintained that the main differences between the various families of plants were morphological characters, which did not appear to be very important for the prosperity of the species.As a result he believed that living things had an inbuilt tendency towards progressive and more perfect development.He especially cites the arrangement of cells in tissues, and the arrangement of leaves on stems, as examples of the inability of natural selection to operate.Added, I think, the number of the various parts of the flower, the position of the ovules, and the shape of the seed, which is of no use in dispersal, &c.

The above objection is powerful.In the first place, though, great care should be exercised when we decide what structures are or were formerly useful in individual species.Secondly, it must always be remembered that when one part changes, the other parts also change, for reasons which are not quite clear, such as: the increase or decrease of the nourishment flowing to one part, the differences between the parts. Mutual oppression, a part developed earlier influencing a part developed later, and so on,--and other causes, which we do not understand, have led to many mysterious cases of correlative action.These effects may, for the sake of brevity, be included in the term law of growth.Thirdly, we must take into account the immediate and definite role of altered conditions of life, and of so-called spontaneous variations, in which the nature of the conditions of life obviously plays a very secondary part.The variation of buds—such as the moss rose on the common rose, or the nectarine on the peach tree, are good examples of spontaneous variation; but even in such cases, if we remember that a drop of worm Given the power of the venom in producing complex galls, we should not be quite sure that the above variations are not due to some change in the conditions of life.The effects of local variations in the properties of the sap must have some powerful cause for every slight individual difference, and for more marked variations which happen by chance; All the individuals of the species will almost necessarily undergo similar modifications.

In earlier editions of this book I underestimated the frequency and importance of variation due to spontaneous variability, which now seems likely.But it is impossible, I cannot believe, to attribute to this cause the innumerable structures so well adapted to the habits of life of the individual species.The well-conditioned racehorse or greyhound, before the principle of artificial selection was known, has exclamated some older naturalists, and I do not believe that it can be explained by this reason. It is worth giving examples to illustrate some of the above points.Concerning the supposed uselessness of the various parts and organs, even in the most familiar higher animals there are many such structures, so developed that no one doubts their importance, yet they The usefulness of has not yet been determined, or has only recently been determined.There is almost no need to say more on this point.Since Bronn gives the lengths of the ears and tails of several species of mice as examples of differences in structure without special use, though this is not a very important example, I may point out that, in Dr. Schobl's opinion, The external ear of the common mouse has many nerves distributed in a peculiar manner, and they undoubtedly serve as organs of touch: the length of the ear is therefore not insignificant.Again, we shall see that the tail is a highly useful organ of grasping in some species; its usefulness is therefore greatly affected by its length.

With regard to plants, since Negley's papers are already available, I will only make the following remarks.It will be admitted that the flowers of the orchidaceae have many curious structures which, a few years ago, were regarded as merely morphological differences, without any particular function; Fertilization is extremely important, and they were probably obtained through natural selection.Until recently no one would have imagined that the different lengths of stamens and pistils, and the way they are arranged, could be of any use in dimorphic or trimorphic plants, but we now know that they do.

In whole groups of some plants the ovules are erect, while in others they are upside down; and in a few plants, in the same ovary, one ovule is upright and the other is upside down.These positions may at first appear to be purely morphological, or of no physiological significance; but Dr. Hooker told me that, in the same ovary, some only the upper ovules are fertilized, and some only the lower ovules are fertilized; This is probably due to the different directions of the pollen tubes entering the ovary.If this is the case, then the position of the ovules, even when one is upright and one is upside down in the same ovary, is presumably the result of selection for any slight deviation in position, whereby fertilization and seed production are advantageously obtained.

Several plants belonging to different "orders" frequently produce two kinds of flowers--one open, with a common structure, and the other closed, incomplete.The two kinds of flowers sometimes appear very different in structure, yet they can also be seen to be mutual gradations on the same plant.Ordinary open flowers can be cross-fertilized; and thus the benefits of cross-fertilization are assured.Yet closed incomplete flowers are also of remarkably high importance, since they produce large quantities of seeds with great certainty at the expense of very little pollen.It has just been said that the two kinds of flowers are often quite different in structure.The petals of incomplete flowers are almost always made of remnants, and the pollen grains are reduced in diameter.In one Ononis columna five alternate stamens are rudimentary; in several species of Viola three are rudimentary, while the remaining two remain normal. function, but has been greatly reduced.In a species of Indian Violet (Violet) (don't know its name, because I have never seen this plant in full bloom here), of the thirty closed flowers, the sepals of six flowers rose from the The normal number of five slices degenerates to three slices.In a certain class in the family Malpigbiaceea, according to A.In the opinion of A. de Jussieu, the closed flower has a further modification, namely, that the five stamens opposite the sepals are all reduced, and only the sixth stamen opposite the petals is developed; In the common flowers of these species, this stamen is absent; the style is hypodeveloped; and the ovary is reduced from three to two.Although natural selection has been sufficiently powerful to prevent certain flowers from opening, and to reduce the excess of pollen by closing them, yet the particular variations mentioned above cannot be so determined, but must be regarded as a function of growth. The consequences of the law, the functional inactivity of certain parts during the process of pollen reduction and flower closure, may also be included in the law of growth. The important effects of the laws of growth are so much to be reckoned with, that I would like to give some other instances in which the same parts or organs vary according to their relative positions on the same plant.According to Schacht, the leaves of the Spanish chestnut and some fir trees differ in their angle of division between nearly horizontal and erect branches.In common rue and some other plants, the central or top flower often opens first, and this flower has five lobes and five petals, and the ovary is also five-chambered; while all the other flowers of these plants are Four counts.In the genus Adoxa in the UK, the flowers on the top generally have only two cockles, while the rest of it is four-numbered, and the surrounding flowers generally have three cockles, while the other parts are five-numbered .In many plants of the Composita and Umbelliferae (and certain other plants), the peripheral flowers have a much more developed corolla than the central flowers; and this often seems to be associated with the underdevelopment of the reproductive organs.There is also the still more curious fact, already mentioned, that the peripheral and central achenes or seeds often differ greatly from each other in shape, colour, and other characters.In Carthamus and certain other Compositae only the central achene has a pappus; while in Hyoseris the same inflorescence bears three achenes of different shapes.In certain Umbelliferae, according to Tausch, the outer seeds are straight and the central ones are anatropous, which de Candor regards as Traits are of high taxonomic importance in other species.Prof. Braun cited a genus of Fumariaceae, whose flowers in the lower part of the spike bear ovate, ribbed, one-seeded nutlets; and in the upper part of the spike It bears lanceolate, two-capsular, two-seeded siliques.In these cases, apart from the well-developed ejaculi for attracting the attention of insects, natural selection, so far as we can judge, has played no part, or only a very minor part.All such variations are the result of the relative position and interaction of the parts; and there can hardly be any doubt that if all the flowers and foliage on the same plant had suffered as much as those on certain parts, If they are affected by the same internal and external conditions, then they will all be changed in the same way. In countless other cases we see variations of structure, considered by botanists to be of generally high importance, occurring only in certain flowers on the same plant, or in different plants growing close together under the same external conditions.As such variations do not seem to be of particular use to the plant, they have not been subject to natural selection.The reason for this is not quite clear, and they cannot even be attributed to any approximation of relative positions, etc., as in the last class of cases mentioned above.Here I cite only a few examples.It is common for flowers to appear irregularly in numbers of four or five on the same plant, and I need not give further examples; but, since the number of parts is small, the variation in number Rare, so I will give the following example, according to de Candol, of the flower of Papaver bracteatum, having two cockles and four petals (which is the common form of the genus Papaver), or three etc. slices and six petals.The manner in which the petals are folded in the bud is a remarkably stable morphological character in most flora; but Professor Asa Gray says that in certain species of the genus Mimulus their flowers almost often like both the Rhinanthidea and the Antirrhinidea, to which Physalis genus belongs.Saint-Tirrell gives the following example: The flowers of certain species of a division of Rutaceta with a single ovary, Zanthoxylon, on the same plant or even on the same panicle, But give birth to one or two ovaries.The capsules of the genus Helianthemum are either one- or three-locular; but the deformed Helianthemum, (H. mutabila) "has a thin septum, somewhat broad, between the pericarp and the placenta".Concerning the flowers of Saponaria officinalis, it has, according to Dr. Masters' observation, a marginal placenta and a free central placenta.At last Saint-Tirrell had found, near the southern extremity of the range of Gompbia oleaformis, two forms which at first he had no doubt of being two distinct species, but which he afterwards saw growing on the same bush. above, and added: "In the same individual the ovary and style, sometimes on the erect shaft, sometimes at the base of the pistil." We thus know that many of the morphological changes in plants are due to the laws of growth and the interaction of parts, and have nothing to do with natural selection.But on the basis of Negli's doctrine that living beings have an inherent tendency to develop toward perfection or progress, can it be said that, in cases of such marked variation, plants are advancing toward a state of high development?On the contrary, from the mere fact that the above-mentioned parts differ or vary greatly in the same plant, I can infer that such variations, however great they may be of generally taxonomic importance, are of extreme indifference in the plants themselves. of.The acquisition of a useless part cannot really be said to raise the order of living beings in nature; and as for the incomplete and closed flower described above, if any new principle must be invoked to explain it, it must be the principle of degeneration, rather than an evolutionary principle; the same must be true of many parasitic and degenerate animals.We are as yet ignorant of the cause of the above-mentioned particular variation; but if this unknown cause were to operate almost uniformly over a long period, we may infer that the effect would also be nearly uniform; and in this case , all the individuals of the species will vary in the same manner. From the fact that the above-mentioned characters are not important to the safety of the species, any slight variation in them would not have been accumulated and increased by natural selection.A structure which has been developed by long continuation of selection, when it has lost its usefulness to the species, is generally liable to variation, as we see in rudimentary organs; for it is no longer subject to the same selection. Power dominates.But from the nature of living beings and the nature of the external conditions, variations which are not essential to the safety of the species, if they occur, they can, and apparently often do, be transmitted in almost the same state to many offspring which have otherwise been modified.For many mammals, birds, or reptiles, the presence or absence of hair, feathers, or scales is not very important; yet hair has been transmitted to nearly all mammals, feathers to all birds, and scales to all true reptiles. .Where a structure, whatever it may be, is common to many allied forms, it is considered by us to be of high taxonomic importance, and the consequence is often assumed to be of vital importance to the species.I am therefore inclined to believe that the differences in form which we regard as important--such as the arrangement of leaves, the distinction of flowers and ovaries, the position of ovules, etc.--appeared in many cases at first as wandering variations, and later By the nature of the living beings and the nature of the surrounding conditions, and by the interbreeding of different individuals, but not by natural selection, sooner or later stabilized; for, as these morphological characters do not affect the safety of the species, any slight deviation from them Neither is governed or accumulated by the action of natural selection.We thus arrive at the curious result that characters of least importance to the life of the species are of most importance to the taxonomist; but, when we shall later discuss the systematic principles of classification, we shall see It's so contradictory at first glance. Although we have no good evidence for an innate tendency towards progressive development in living organisms, yet, as I have attempted to show in Chapter 4, the continual action of natural selection will necessarily produce progressive development. The higher standard of the human body is best defined by the degree to which the organs have been specialized or differentiated: natural selection has tended to accomplish this end, because the more specialized or differentiated the organs, the more efficient are their functions. The eminent zoologist, Mr. Mivart, has lately collected the objections which I and others have raised to the theory of natural selection advocated by Mr. Wallace and myself, and has explained them with admirable skill and force.Those dissentings, so arranged, make for a formidable lineup; and since Mr. Mivart has not planned to enumerate facts and arguments contrary to his conclusions, the reader will have to pay a great deal of reason and memory to weigh the evidence on both sides. Great effort.When discussing particular cases, Mr. Mivart leaves aside the effects of enhanced use and disuse of body parts, which I have often maintained to be of high importance, and in Variations Under Domestication, I believe I have discussed this issue in more detail than any other author.At the same time, he often thinks that I have not estimated variations independent of natural selection, whereas in the work just referred to I have collected more definite instances than in any other work known to me.My judgment is not necessarily sound, but having read Mr. Mivart's book carefully, and comparing, paragraph by paragraph, what he says with what I say on the same subject, I have never believed so strongly in this The conclusions drawn in the book are generally true. Of course, in such a complicated issue, many partial mistakes are inevitable. All of Mr. Mivart's objections will be discussed in this book, or have already been discussed.One of the new arguments which struck many readers was that "Natural selection cannot account for the first stages of useful construction."This problem is closely connected with the gradation of characters which often accompanies a change of function, such as the transformation of the swim-bladder into lungs, etc., already discussed under the two headings of the previous chapter; Several examples put forward by Mr. Vater, choose the most representative ones, and discuss them in a little detail, because of the limited space, we cannot discuss everything he put forward. The giraffe, because of its extremely high stature, with its long neck, forelegs, and tongue, is all made up beautifully for eating the taller branches of trees.It thus has access to food in the same place which other ungulates cannot reach; this must have been of great advantage to him in times of famine, and the Niata cattle of South America show us how little structural Differences, in times of famine, can also make a big difference in preserving animal life.This cow grazes on grass like other cattle, only because of its protruding lower jaw, it cannot be forced to eat twigs and twigs during the recurring dry season like ordinary cows and horses. Reeds, etc.; so at these times, if the owner does not feed them, the Niatay cattle will die.Before discussing Mr. Mivart's objection, it may be well to show once more how natural selection operates in all ordinary cases.Man has altered some of his animals without regard to peculiarities of construction, as in the case of race horses and greyhounds, by simply selecting from among the fastest individuals for preservation and breeding, or as in the case of racehorses and greyhounds. In the case of fighting cocks, breeding is done simply by selecting from the winning chickens.In a state of nature, too, in the nascent state of the giraffe, those individuals which can feed from the highest places, and in times of famine even from an inch or two above the rest, are often preserved; for They can roam the whole area in search of food.Individuals of the same species often differ slightly in the proportional lengths of the various parts of the body, and this is described in many works of natural history, where detailed measurements are given.These slight differences in proportion, arising from the laws of growth and variation, are of little use, or of little importance, to many species.But for giraffes in their nascent state, if one takes into account their possible living habits at that time, the situation is different; because individuals with a certain part or several parts of the body are generally able to survive if they are somewhat longer than ordinary individuals.When such individuals are crossed, the offspring left either inherit the same physical characteristics, or tend to be further modified in the same manner; and the less fit individuals in these respects are the most liable to perish. From this we see that nature does not need to separate pairs of individuals, as man has planned to improve breeds; natural selection preserves and thus separates all good individuals, leaves them to cross freely, and destroys all inferior individuals. Lose.According to the long continuation of this process - which corresponds exactly to what I have called unconscious selection in man - and which is no doubt combined in a most important It seems that it can definitely be transformed into a giraffe. Mr. Mivart has raised two objections to this conclusion. One objection is that an increase in the body obviously requires an increase in food supply. interests are very problematic.”But since there are actually large herds of giraffes in South Africa, and since some of the largest antelopes in the world, taller than cattle, live there in herds, why should we doubt The middle classes, which suffered severe famines as they do today, existed there before.There must have been some benefit to the nascent giraffe at various stages of body enlargement to have access to a food supply that had been left out of the reach of other ungulates in the area.Let us also not lose sight of the fact that the enlargement of the body is a defense against almost every other carnivore except the lion; What Mr. Chauncey Wright said can be used as a watchtower.It is for this reason that, according to Sir. S. Baker, it is more difficult to sneak up on a giraffe than on any animal.The giraffe, in turn, uses its long neck as a tool of attack or defense by violently jostling its stump-horned head.The preservation of individual species can seldom be determined by any one advantage, but must be determined by a combination of all advantages, great and small. Mr. Mivart asks (and this is his second objection), if natural selection has such a force, and if there is such an advantage in being able to bite the leaves aloft, why is it that, except for the giraffe and the slightly shorter-necked species, Besides the camel, the guanaco, and the long-headed camel (macrauchenia), did any other ungulates not acquire a long neck and a tall body?Or, why didn't any member of the group get a long kiss?Since South America was once inhabited by innumerable giraffes, the answer to the above question is not difficult, and can be best answered by an example.In every meadow in England, if a tree grows on it, we see its low branches cut to equal height by being eaten by horses or cattle; if, say, the sheep kept there, What advantage could it be to them to have acquired a slightly longer neck? In every country some species of animals were almost certainly able to eat higher leaves than others; A species is able to elongate its neck for this purpose through natural selection and the effects of enhanced use.In South Africa, the competition for the leaves of acacias and other tall trees must have been between giraffes and giraffes, not between giraffes and other ungulates. Why, in other parts of the world, have no long necks or long snouts been given to the various animals belonging to this "order"?This cannot be definitively answered; but it is as wrong to wish for a definitive answer to this question as it is to wish for a definitive answer to the question of why certain things in the history of mankind did not happen in this country but in another. reasonable.We are ignorant of the conditions which determine the number and range of each species; we cannot even conjecture what changes of structure would be favorable to the increase of its individual numbers in a new country.In general, however, we can see various causes for the development of the long neck or snout.Reaching foliage at considerable heights (not climbing, since ungulates are not specially constructed for climbing trees) implies considerable enlargement of the body; Why is it so rare in South Africa, where there are so many large quadrupeds, even though the vegetation there is so luxuriant?We don't know; why was the end of the Tertiary period more suitable for their survival than the present?We don't know either.Whatever its cause, we can see that some places and periods have greatly favored the development of a gigantic quadruped like the giraffe, more than others. In order for an animal to acquire a particular and great development in one structure, it is almost inevitable that several other parts should also be modified and adapted to each other.Though every part of the body varies slightly, the necessary parts do not always vary in the proper direction and to the proper degree.With regard to the different species of our domestic animals, we know that the parts of their bodies vary in different ways and in different degrees; and we know that some species are more susceptible to variation than others.Even when favorable variations have occurred, natural selection has not necessarily been able to act upon them to produce a structure which is manifestly advantageous to the species.For instance, if the number of individuals living in a place is determined chiefly by the infestation of predators, or by external and internal parasites, etc.—as often seems to be the case,— —then natural selection would then play little, or be greatly hindered, in modifying any particular structure for food.Finally, natural selection is a slow process, so that in order to have any noticeable effect, equally favorable conditions must persist over a long period of time.Except for these general and vague reasons, we cannot really explain why ungulates in many parts of the world have not acquired very long necks, or other organs, for gnawing leaves on high branches. Objections of the same nature as above have been raised by many authors.In each case, various causes, besides the general ones above stated, may interfere with the acquisition by natural selection of structures supposedly advantageous to a particular species.Why, one author asked, did the ostrich not acquire the ability to fly?但是，只要略略一想便可知道，要使这种沙漠之鸟具有在空中运动它们巨大身体的力量，得需要何等多的食物供应。海洋岛（Oceanic Islands ）上有蝙蝠和海豹，然而没有陆栖哺乳类；但是，因为某些这等蝙蝠是特别的物种，它们一定在这等岛上住得很长久了。所以莱尔爵士问道，为什么海豹和蝙蝠不在这些岛上产出适于陆栖的动物呢？并且他举出一些理由来答复这个问题。但是如果变起来，海豹开始一定先转变为很大的陆栖食肉动物；蝙蝠一定先转变为陆栖食虫动物；对于前者，岛上没有可捕食的动物；对于蝙蝠，地上的昆虫虽然可以作为食物，但是它们大部分已被先移住到大多数海洋岛上来的，而具数量很多的爬行类和鸟类吃掉了。构造上的级进变化，如果在每一阶段对于一个变化着的物种都有利，只有在某种特别的条件下才会发生。一种严格的陆栖动物，由于时时在浅水中猎取食物，随之在溪或湖里猎取食物，最后可能变成一种如此彻底的水栖动物，以致可以在大洋里栖息。但海豹在海洋岛上找不到有利于它们逐步再变为陆栖类型的条件。至于蝙蝠，前已说过，为了逃避敌害或避免跌落，大概最初像所谓飞鼠那样地由这树从空中滑翔那树，而获得它们的翅膀；但是真正的飞翔能力一旦获得之后，至少为了上述的目的，决不会再变回到效力较小的空中滑翔能力里去。蝙蝠确像许多鸟类一样，由于不使用，会使翅膀退化缩小，或者完全失去；但是在这种情形下，它们必须先获得单凭后腿的帮助而能在地上跑得很快的本领，以便能够与鸟类或别的地上动物相竞争；而蝙蝠似乎特别不适于这种变化的。上述这等推想无非要指出，在每一阶段上都是有利的一种构造的转变，是极其复杂的事情；并且在任何特殊的情形里没有发生过渡的情况，毫不值得奇怪。 最后，不止一个作者问道，既然智力的发展对一切动物都有利，为什么有些动物的智力比别的动物有高度的发展呢？为什么猿类没有获得人类的智力呢？对此是可以举出各种各样的原因来的；但都是推想的，并且不能衡量它们的相对可能性，举出来也是没有用处的。对于后面的一个问题，不能够希望有确切的解答，因为还没有人能够解答比这更简单的问题——即在两族未开化人中为什么一族的文化水平会比另一族高呢；文化提高显然意味着脑力的增加。 我们再回头谈谈米伐特先生的其他异议。昆虫常常为了保护自己而与各种物体类似，如绿叶或枯叶、枯枝、一片地衣、花、棘刺、鸟粪以及别种活昆虫，但关于最后一点留在以后再讲。这种类似经常是奇异地真切，并不限于颜色，而且及于形状，甚至昆虫支持它的身体的姿态。在灌木上取食的尺蠖，常常把身子峤起、一动也不动地像一条枯枝，这是这一种类似的最好事例。模拟像鸟粪那样物体的情形是少有的，而且是例外的。关于这一问题，米伐特先生说道：“按照达尔文的学说，有一种稳定的倾向趋于不定变异，而且因为微小的初期变异是朝向一切方面的，所以它们一定有彼此中和和最初形成极不稳定的变异的倾向，因此，就很难理解，如果不是不可能的话，这种无限微小发端的不定变异，怎么能够被自然选择所掌握而且存续下来，终于形成对一片叶子、一个竹枝或其他东西的充分类似性。”但是在上述的一切情形里，昆虫的原来状态与它屡屡访问的处所的一种普通物体，无疑是有一些约略的和偶然的类似性的。只要考虑一下周围物体的数量几乎是无限的，而且昆虫的形状和颜色是各式各样的，就可知道这并不是完全不可能的事。某些约略的类似性对于最初的发端是必要的，由此我们能够理解为什么较大的和较高等的动物（据我所知，有一种鱼是例外）不会为了保护自己而与一种特殊的物体相类似，只是与周围的表面相类似，而且主要是颜色的相类似。假定有一种昆虫本来与枯枝或枯叶有某种程度的类似，并且它轻微地向许多方面进行变异，于是使昆虫更像任何这些物体的一切变异便被保存下来，因为这些变异有利于昆虫逃避敌害，但是另一方面，其他变异就被忽略，而终于消失；或者，如果这些变异使得昆虫完全不像模拟物，它们就要被消灭。如果我们不根据自然选择而只根据彷徨变异来说明上述的类似性，那末米伐特先生的异议诚然是有力的；但实际情况并非如此。 华莱斯先生举出一个竹节虫（Ceroxylus laceratus）的例子，它像“一枝满生鳞苔的杖”。这种类似如此真切，以致大亚克（Dyak）土人竟说这种叶状瘤是真正的苔。米伐特先生认为这种“拟态完全化的最高妙技”是一个难点，但我看不出它有什么力量。昆虫是鸟类和其他敌害的食物，鸟类的视觉大概比我们的还要敏锐，而帮助昆虫逃脱敌害的注意和发觉的各级类似性，就有把这种昆虫保存下来的倾向；并且这种类似性愈完全，对于这种昆虫就愈有利。考虑到上述竹节虫所属的这一群里的物种之间的差异性质，就可知道这种昆虫在它的身体表面上变得不规则，而且多少带有绿色，并不是不可能的；因为在各个群里，几个物种之间的不同性状最容易变异，而另一方面，属的性状，即一切物种所共有的性状最为稳定。 格林兰（Greenland）的鲸鱼是世界上一种最奇异的动物，鲸须或鲸骨是它的最大特征之一。鲸须生在上颚的两侧，各有一行，每行约三百片，很紧密地对着嘴的长轴横排着。在主排之内还有一些副排。所有须片的末端和内缘都磨成了刚毛，刚毛遮盖着整个巨大的颚，作为滤水之用，由此而取得这些巨大动物依以为生的微小食物。格林兰鲸鱼的中间最长的一个须片竟长达十英尺、十二英尺甚至十五英尺；但在鲸类的不同物种里它的长度分为诸级，据斯科列斯比（Scoresby ）说，中间的那一须片在某一物种里是四英尺长，在另一物种里是三英尺长，又在另一物种里是十八英寸长，而在长吻鳁鲸（Balaenoptera rostrata）里其长度仅九英寸左右。鲸骨的性质也随物种的不同而有所差异。 关于鲸须，米伐特先生说道：当它“一旦达到任何有用程度的大小和发展之后，自然选择才会在有用的范围内促进它的保存和增大。但是在最初，它怎样获得这种有用的发展呢？”在回答中我们可以问，具有鲸须的鲸鱼的早期祖先，它们的嘴为什么不应像鸭嘴那样地具有栉状片呢？鸭也像鲸鱼一样，依靠滤去泥和水以取得食物的；因此这一科有时候被称为滤水类（Criblatores ）。我希望不要误解我说的是鲸鱼祖先的嘴确曾具有像鸭的薄片喙那样的嘴。我只是想表明这并不是不可信的，并且格林兰鲸鱼的巨大鲸须板，也许最初通过微小的级进步骤从这种柿状片发展而成，每一级进步骤对这动物本身都有用途。 琵琶嘴鸭（Spatula clypeata ）的喙在构造上比鲸鱼的嘴更巧妙而复杂。根据我检查的在其上颚两侧各有188 枚富有弹性的薄栉片一行，这些栉片对着喙的长轴横生，斜列成尖角形。它们都是由颚生出，靠一种韧性膜附着在颚的两侧，位于中央附近的栉片最长，约为三分之一英寸，突出边缘下方达0.14英寸长。在它们的基部有斜着横排的栉片构成短的副列。这几点都和鲸鱼口内的鲸须板相类似。但接近嘴的先端，它们的差异就很大，因为鸭嘴的栉片是向内倾斜，而不是下向垂直的。琵琶嘴鸭的整个头部，虽然不能和鲸相比，但和须片仅九英寸长的、中等大的长吻鳁鲸比较起来，约为其头长的十八分之一；所以，如果把琵琶嘴鸭的头放大到这种鲸鱼的头那么长，则它们的栉片就应当有六英寸长，——即等于这种鲸须的三分之二长。琵琶嘴鸭的下颚所生的栉片在长度上和上颚的相等，只是细小些；因为有这种构造，它显然与不生鲸须的鲸鱼下颚有所不同。另一方面，它的下颚的栉片顶端磨成细尖的刚毛，却又和鲸须异常类似。锯海燕属是海燕科的另一个成员，它只在上颚生有很发达的栉片，突出颚边之下，这种鸟的嘴在这一点上和鲸鱼的嘴相类似。 从琵琶嘴鸭的喙这种高度发达的构造（根据我从沙尔文先生［Mr．Salvin］送给我的标本和报告所知道的），仅就适于滤水这一点来说，我们可以经由湍鸭（Merganetta armata）的喙，并在某些方面经由鸳鸯（Aix sponsa）的喙，一直追踪到普通家鸭的喙，其间并没有任何大的间断。家鸭喙内的栉片比琵琶嘴鸭喙内的栉片粗糙得多，并且牢固地附着在颚的两侧；在每侧上大约只有五十枚，不向嘴边下方突出。它们的顶端呈方形，并且镶着透明坚硬组织的边，好像是为了轧碎食物似的。下颚边缘上横生着无数细小而突出很少的突起线。作为一个滤水器来说，虽然这种喙比琵琶嘴鸭的喙差得多，然而每个人都知道，鸭经常用它滤水的。我从沙尔文先生那里听到，还有其他物种的栉片比家鸭的栉片更不发达；但我不知道它们是否把它当作滤水用的。 现在谈一下同科的另一群。埃及鹅（Chenalopex）的喙与家鸭的喙极相类似；但是栉片没有那么多，那么分明，而且向内突出也不那样厉害；然而巴利特先生（Mr. E. Bartlett）告诉我说，这种鹅“和家鸭一样，用它的嘴把水从喙角排出来”。但是它的主要食物是草，像家鹅那样地咬吃它们。家鹅上颚的栉片比家鸭的粗糙得多，几乎混生在一起，每侧约有二十七枚，末端形成齿状的结节。颚部也满布坚硬的圆形结节。下颚边缘由牙齿形成锯齿状，比鸭喙的更突出，更粗糙，更锐利。家鹅不用喙滤水，而完全用喙去撕裂或切断草类，它的喙十分适于这种用途，能够靠近根部把草切断，其他任何动物几乎都不及它。另外还有一些鹅种，我听到巴利特先生说，它们的栉片比家鹅的还不发达。 由此我们看到，生有像家鹅喙那样的喙、而且仅供咬草之用的鸭科的一个成员，或者甚至生有栉片较不发达的喙的一个成员，由于微小的变异，大概会变成为像埃及鹅那样的物种的，——由此更演变成像家鸭那样的物种，——最后再演变成像琵琶嘴鸭那样的物种，而生有一个差不多完全适于滤水的喙；因为这种鸟除去使用喙部的带钩先端外，并不使用喙的任何其他部分以捉取坚硬的食物和撕裂它们。我还可补充地说，鹅的喙也可以由微小的变异变成为生有突出的、向后弯曲的牙齿的喙，就像同科的一个成员秋沙鸭（Merganser）的喙那样的，这种喙的使用目的大不相同，是用作捕捉活鱼的。 再回头来讲一讲鲸鱼，无须鲸（Hyperoodon bidens）缺少有效状态的真牙齿，但是据拉塞丕特（Lacepede）说，它的颚散乱地生有小形的、不等的角质粒点。所以假定某些原始的鲸鱼类型在颚上生有这等相似的角质粒点，但排列得稍微整齐一些，并且像鹅喙上的结节一样，用以帮助捉取和撕裂食物，并不是不可能的。如果是这样的话，那么就几乎不能否认这等粒点可以通过变异和自然选择，演变成像埃及鹅那样的十分发达的栉片，这种栉片是用以滤水和捉取食物的；然后又演变成像家鸭那样的栉片；这样演变下去，一直到像琵琶嘴鸭那样的专门当作滤水器用的构造良好的栉片。从栉片达到长吻鳁鲸须片的三分之二长这一个阶段起，在现存鲸鱼类中观察到的级进变化可以把我们向前引导到格林兰鲸鱼的巨大须片上去。这一系列中的每一步骤，就像鸭科不同现存成员的喙部级进变化那样，对于在发展进程中其器官机能慢慢变化着的某些古代鲸鱼都是有用的，对此毫无怀疑的余地。我们必须记住，每一个鸭种都是处于剧烈的生存斗争之下的，并且它的身体的每一部分的构造一定要十分适应它的生活条件。 比目鱼科（Pleuronectidra）以身体不对称著称。它们卧在一侧，——多数物种卧在左侧，有些卧在右侧；与此相反的成鱼也往往出现，下面，即卧着的那一侧，最初一看，与普通鱼类的腹面相类似：它是白色的，在许多方面不如上面那一侧发达，侧鳍也常常比较小。它的两眼具有极其显著的特征；因为它们都生在头部上侧。在幼小的时候，它们本来分生在两侧，那时整个的身体是对称的，两侧的颜色也是相同的。不久之后，下侧的眼睛开始沿着头部慢慢地向上侧移动；但并不是像从前想像的那样是直接穿过头骨的。显然，除非下侧的眼睛移到上侧，当身体以习惯的姿势卧在一侧时，那只眼睛就没有用处了。还有，这大概是因为下侧的那一只眼容易被沙底磨损的缘故。比目鱼科那种扁平的和不对称的构造极其适应它们的生活习性，这种情形，在若干物种如鳎（soles）、鲽（flounders ）里也极其普通，就是很好的说明。由此得到的主要利益似乎在于可以防避敌害，而且容易在海底取食。然而希阿特说，本科中的不同成员可以“列为一个长系列的类型，这系列表示了它们的逐渐过渡，从孵化后在形状上没有多大改变的庸鲽（Hippog-lossus pinguis）起，一直到完全卧倒在一侧的鳎为止”。