Chapter 3 Chapter 1 Variations in Domestication

Causes of variability - Habits and effects of the use and non-use of organs - Correlative variation - Heredity - Characters of domestic varieties - Difficulty in distinguishing species from varieties - Domestic varieties originate from one or more species — Kinds of domestic pigeons, their differences and origins — Principles of selection upon which they were based in antiquity and their effects — Unknown origins of domesticated organisms — Planned and unconscious selection — Advantages of artificial selection. causes of variability One of the points which most attracts our attention in the comparison of individuals of the same variety or subvariety of the older cultivated plants and domestic animals, is that they generally differ from each other more than any species in a state of nature. Variation among individuals of a species or variety is large.Reflecting on the variety of cultivated plants and domesticated animals, which have been varied by long exposure to very different climates and managements, we must be led to the conclusion that this great variability, From the conditions of life under which our domesticated organisms are less uniform, and somewhat different from, those of our parent-species in a state of nature.There are also some possibilities, as proposed by Andrew Knight; he believes that this variability may be partly related to the excess of food.It seems obvious that organisms must be grown under new conditions for several generations before they can undergo much variation; and that organisms, once modified, can generally continue to vary during many generations.No instance of a mutating organism ceased to mutate under cultivation has ever been seen on record.The oldest cultivated plants, such as wheat, are still producing new varieties; the oldest domesticated animals are still rapidly improving or varying.

After long study of the subject, it is evident, so far as I can judge, that the conditions of life act in two ways--directly on the whole system or only on certain parts, and indirectly on the reproductive system.As regards direct action, we must remember that in each case, as Prof. Weismann has lately maintained, and as I have occasionally mentioned in Variations Under Domestication, it has two A factor: the nature of living things and the nature of conditions.The former seems to be the more important; for, so far as we can judge, almost similar variations can sometimes occur under dissimilar conditions; on the other hand, dissimilar variations can occur under nearly identical conditions.These effects are either constant or indeterminate for the offspring.The effect may be regarded as definite if all, or nearly all, the offspring of individuals raised under certain conditions during several generations are modified in the same manner.But it is extremely difficult to draw any conclusions about the range of variation so definitely induced.Many slight changes, however, such as size from the amount of food, color from the nature of food, thickness of skin and hair from climate, etc., can hardly be doubted.For each of the innumerable variations we see in the feathers of chickens, there must have been some effective cause; and if the same cause, through many generations, had acted alike on many individuals, all of them would probably have behaved in the same manner. Make mutations.The fact that a minute amount of venom from a gall-producing insect must produce complex and unusual galls when injected into a plant shows us what happens in plants if the properties of the sap are chemically altered. Strange change.

Indeterminate variability is more often the more common result of changed conditions than definite variability, and it will probably play a more important part in the formation of our domestic races.We see indeterminate variability in an infinite number of minute characters, which distinguish individuals within the same species, and which cannot be supposed to be inherited from parents, or more distant ancestors.Even larvae in litters, and seedlings emerging from the same capsule, sometimes show most marked differences from each other.Over long periods of time, in the same place, among millions of individuals reared on nearly the same food, there will appear very marked differences of structure which may be called monstrosities; Clear boundaries.All such changes in structure, whether very slight or very marked, occur in many individuals living together, and may be regarded as the indeterminate effect of the conditions of life acting on each individual, unlike cold for cold. The different effects that occur to different people are almost the same, due to their different physical conditions or constitutions, causing cough or cold, rheumatism, or inflammation of some organs.

Concerning what I have called the indirect effect of altered external conditions, i.e., the effects on the reproductive system, we may infer that the variability thus induced is partly due to the extreme sensitivity of the reproductive system to any change in external conditions, and partly, As has been pointed out by Kailoroud et al., because the variations produced by crossing between different species are similar to those which occur in plants and animals kept under new or unnatural conditions, a number of facts clearly point to the fact that reproductive How remarkably sensitive the system is to very slight changes in the surrounding conditions.It is the easiest thing to domesticate animals, but it is the most difficult thing to make them reproduce freely within the cage, even if the male and female mate.How many animals, even if raised in the original place, in a state of almost free, can't give birth!It is generally wrong to ascribe this condition to an impairment of instinct.Many cultivated plants appear to be extremely vigorous, but rarely, or never, set fruit!In a few occasions, a very small change has been found, such as more or less water in a certain special growth period, which can determine whether the plant is fruitful or not.On this curious subject; the detailed facts which I have gathered have been published elsewhere, and I shall not here describe them.But to show how wonderful the laws governing the reproduction of animals in the range are, I would like to say that carnivorous animals, even if brought from the heat, reproduce quite freely in the range of England; For example, they rarely give birth; whereas birds of prey, with rare exceptions, hardly ever lay fertilized eggs.The pollen of many exotic plants is as utterly useless as that of the most barren hybrids.On the one hand, we see many species of domesticated animals and plants, though often infirm, which reproduce freely within the enclosure; Long-lived and vigorous (of which I can cite innumerable instances), yet their reproductive system is, for unknown reasons, so severely affected that it is rendered useless; Not surprisingly, it is irregular and produces offspring more or less like its parents.I should add that some organisms reproduce freely under the most unnatural conditions (such as rabbits and minks kept in cages), which shows that their reproductive organs are not easily affected; so some animals and plants can withstand domestication. or cultivated, and the changes are very slight—probably no greater than that which occurs in a state of nature.

Some naturalists maintain that all variations are connected with the action of sexual reproduction; but this is certainly false; and I have, in another work, included among the species what horticulturists call "Sporting plants." A long list:--The plant will suddenly produce a shoot which, unlike other shoots of the same plant, will have new and sometimes markedly different characters.They may be called variations of buds, and may be propagated by grafting, cuttings, etc., and sometimes by seeds.They occur rarely in nature, but not uncommon in cultivated conditions.Since, on the same tree under uniform conditions, out of the thousands of shoots which grow yearly, a single shoot will suddenly appear with a new character; and since shoots from different trees under different conditions will sometimes produce almost Of the same varieties—for example, the buds on the peach tree give nectarines, and the common roses give moss roses—so that we can clearly see that in determining the particular type of each variation In general, the nature of the external conditions is of only a second importance to the nature of living beings;—or no more important in determining the nature of the flame than the nature of the spark which enables the combustible to burn.

Habits and effects of organ use and non-use; associated variation; heredity Changes in habit can have hereditary effects, as when a plant is moved from one climate to another, its period of flowering is altered.In animals, the frequent use or disuse of parts of the body has a more marked effect; for example, I have found that the wing-bones of the domestic duck are lighter in proportion to the whole skeleton than those of the wild duck, while the leg-bones of the domestic duck are more prominent among them. In proportion to the total skeleton, it is heavier than the leg bones of the mallard; this change may be safely attributed to the fact that the domestic duck flies less and walks less than its wild ancestors.The udders of cows and goats, which are better developed where they are accustomed to be milked than where they are not, and that this development is hereditary, may be another example of the effect of use.There is not anywhere among our domestic animals a species which does not have drooping ears; and it is probably probable that the drooping of the ears is due to the fact that the animals seldom suffer great frights and the ear muscles are not used.

Many laws govern variation, but a few are only vaguely understood, and shall be discussed a little later.Here I am only going to say what may be called correlated variation.Important changes in the embryo or larva will presumably cause changes in the mature animal as well.The correlation of quite different parts in monstrosities is curious; and of this subject many instances are recorded in the great work of St. Tyrrell the Younger.Breeders believe that long limbs are almost always accompanied by a long head.Some cases of association are quite curious; for example, cats which are all white and have blue eyes are generally deaf; but Mr. Tait has lately remarked that this is the case only in male cats.There are many striking examples of the association of body color and physical characteristics in animals and plants.According to the facts collected by Heusinger, white-coated sheep and pigs suffer from eating certain plants, which dark individuals are avoided; Prof. Wyman recently wrote to tell A good example of this fact I have: He asked some farmers in Virginia why they had all black pigs, and they told him that the pigs had eaten Lachnantbes, and the bones had turned reddish, The trotters will fall off, except for the black variety; and a Virginia grazier added, "We breed the black ones in the first litter because they alone have a good chance of survival."Hairless dogs with missing teeth; long- and coarse-haired animals, said to have a tendency to have horns or multiple horns; hairy-footed pigeons, with leather between the outer toes; short-billed pigeons, with small feet; long-billed pigeons, with large feet .Therefore, if a man chooses any characteristic, and thereby strengthens it, it is almost certain to change the constitution of other parts of the body unconsciously, by mysterious laws of correlation.

The consequences of various, unknown, or only vaguely understood laws of variation are infinitely complex and varied.For the treatises of several ancient cultivated plants, such as the hyacinth (Hyacinth), the potato, and even the dahlia, it is well worthwhile to make a careful study; Differences do surprise us.The whole constitution of living beings seems to have become plastic, and to deviate by a very slight degree from that of its parent-types. Variations which are not inherited are of no importance to us.But the number and variety of heritable differences of structure, whether slight or of considerable physiological importance, are truly infinite.The two volumes of treatises by Dr. Prosper Lucas are the fullest and best work on the subject.No breeder doubts how powerful hereditary tendencies are; Like produces like is their fundamental belief: only theoretical writers doubt this principle.When any deviation of structure frequently occurs, and is seen in both parent and child, we cannot say whether it is the result of the same cause acting on both; In millions of individuals, chance appears in the parents and reappears in the offspring, when pure opportunism almost compels us to attribute its recurrence to heredity.Everyone must have heard of Albinism, prickly skin, and hairy bodies that appear in several members of the same family.If strange and rare deviations of structure are indeed inherited, less strange and more common deviations may, of course, also be regarded as inherited.To regard the inheritance of various characters as the rule, and the non-inheritance as the anomaly, is probably the correct way of looking at the whole problem.

The laws governing heredity are largely unknown.No one can explain why the same characteristics between different individuals of the same species or between heterogeneous species are sometimes inherited and sometimes not; why offspring often reproduce certain traits of their grandfather or grandmother, or attach importance to the traits of more distant ancestors; Why a trait is often passed from one sex to both sexes, or to one sex only, and more commonly, but not exclusively, to the same sex.It is a fact of considerable importance to us that traits which occur in the male domestic animals are often transmitted to the male either absolutely or to a great extent.

There is a more important rule, which I think may be believed, namely, that at whatever period of life a trait first appears, it tends to recur in the offspring at a comparable age, though sometimes earlier.In many cases this is quite true; for example, the hereditary character of the horns of the ox appears only in the period when its offspring are nearly mature; and we know that the characters of the silk-worm each appear at a comparable larval or pupal stage.But hereditary disease, and other facts, lead me to believe that the law holds on a larger scale, that a trait, though there is no apparent reason why it should appear at a certain age, is, at the time of its appearance, tend to be at the same time as the parent's first appearance.I believe this law is of the utmost importance in explaining the laws of embryology.These observations refer, of course, to the initial appearance of the character, and not to the original cause acting on the ovules or the male genitals; late, but apparently due to the action of the male reproductive plasm.

I have already spoken of the problem of atavism, and I would like to mention here a point often made by naturalists—namely, that our domestic varieties, when returned to their wild state, gradually but necessarily revert to their original origin. ancestral traits.It has been argued, therefore, that species in a state of nature cannot be inferred deductively from domestic races.I have endeavored, and failed, to inquire upon what definite facts the foregoing statement is so frequently and boldly made.It is indeed extremely difficult to prove its truth: we may safely assert that a very large number of unusually marked domestic varieties probably could not have survived in the wild.In many cases we do not know what the original ancestry looked like, so we cannot say whether the atavism that has occurred is nearly complete or not.In order to prevent the influence of interbreeding, it would probably be necessary to keep only a single variety in its new home.Nevertheless, since our varieties do sometimes reproduce some of the characters of the ancestral forms, it seems to me that the following is probably possible: If we could succeed in associating for many generations several families, such as cabbage, Some of the tribes naturalized or cultivated on extremely poor soils (but in this case some effects are due to some action of the poor soils), most or even all of them will reproduce the characters of their wild original ancestors .Whether the experiment was successful or not is not very important to our argument; for the experiment itself has changed the conditions of life.If it can be shown that when we bring domestic varieties under the same conditions, and keep them together in large groups, and let them cross freely, preventing any slight deviation in structure by intermingling, then if they also show a strong tendency to reversion —that is, the loss of their acquisition, in which case I would agree that nothing can be deduced from domestic varieties of species in nature.But there is hardly a shadow of evidence in favor of this view: to conclude that we cannot make our drive and race horses, longhorn and shorthorn cattle, chickens of every breed, and vegetables of every kind eaten, To reproduce indefinitely is against all experience. Characters of domestic varieties; difficulties in distinguishing varieties from species; origin of domestic varieties from one or more species When we look at hereditary varieties, or races, of domesticated animals and cultivated plants, and compare them with closely allied species, we generally see that the individual domesticated races are, as stated above, inferior in character to the true species. ) are as consistent.The characters of domestic races are often more or less monstrous; that is to say, although they differ very little in several respects from each other, and from other species of the same genus, yet, when they are compared with each other, there are often some differences in some parts of the body. exhibit a great degree of variation, especially when compared with their nearest allied species in a state of nature.Besides monstrous characters (and the complete fertility of varieties crossed—a question to be dealt with hereafter), domestic races of the same species differ from each other as much as closely allied species of the same genus differ from each other in a state of nature. similar, but the former differs to a lesser extent in most cases.We must admit this to be absolutely true, for some able experts regard many domesticated families of animals and plants as descendants of originally distinct species, and others only regard They are considered variants.This doubt would not be so persistently repeated if there were any marked difference between a domestic race and a species.It has often been said that differences in character between domestic races have no genus value.We may show that this statement is incorrect; but naturalists are quite divided when it comes to ascertaining what characters should be of generic value; and all these evaluations are at present empirical.When it is shown how genera originate in nature, it will be seen that we have no right to expect to find in our domestic races such an amount of variation as is often found in genera. In attempting to estimate the amount of difference in structure between closely related domestic races, we are at once thrown into perplexity by not knowing whether they have descended from one or several parent-species.It would be interesting if this could be clarified; for example, if it could be shown that the well-known grey-hounds, bloodhounds, terriers, spaniels ( The spaniel and the bulldog, both descended from a single species, have a serious influence on us, making us less familiar with the many closely allied natural species—for example, the many species of foxes—that inhabit various parts of the world. ——It is the unchanging statement that creates great doubts.I do not believe, as we shall see, that the whole of the differences in these several breeds of dogs arose from domestication; Some distinct families of domesticated species have presumed, and even strong, evidence of descent from a wild parent. It has often been assumed that man has selected domesticated animals and domesticated plants with a great propensity for genetic variation, able to withstand all kinds of climates.I do not dispute that these qualities have greatly increased the value of most domesticated organisms; but how, when savages first domesticated an animal, could they know whether it would vary in successive generations, and how Can you know if it can withstand other climates?Will the little variability of the ass and goose, the little resistance of the reindeer to heat, and the common camel to cold, prevent them from being domesticated?I cannot doubt that, if animals and plants were taken from a state of nature, corresponding in number, origin, and taxonomic order to our domestic productions, and supposing that they had been reproduced for as many generations under domestic conditions, the variation on the average would vary. As much variation as has ever occurred in the parent-species of our existing domesticated organisms. Whether most of the animals and plants domesticated from antiquity have been descended from one or from several wild species, no definite conclusion can now be drawn.The arguments of those who believe that domesticated animals are of many origins rest chiefly on the fact that the breeds of domesticated animals which we find in antiquity, on Egyptian steles and in lake dwellings in Switzerland, were extremely diverse; The species are very similar, or even identical.But this only pushes back the history of civilization, and shows that animals have been domesticated much longer than was ever supposed.The Swiss lake dwellers cultivated several varieties of wheat and barley, peas, poppies (for oil), and flax; and they also had several domesticated animals.They also traded with other peoples.These clearly indicate, as Heer says, that they had a very advanced civilization at such an early stage; At that time, the animals kept by various tribes in various places probably mutated and produced different races.Since the discovery of flint implements in surface formations in many parts of the world, all geologists have believed in the very remote existence of savages; bred. The origin of most domesticated animals may remain forever obscure.But I can say here that I have studied domestic dogs all over the world, and painstakingly gathered all the known facts, and I have come to the conclusion that several wild species of the dog family have been domesticated, and their blood in some cases Downstairs are mixed together and run in the veins of our domestic breeds.As to sheep and goats, I cannot yet form a decisive opinion. From the facts which Mr. B1yth has written to me to inform me of the habits, voice, constitution, and structure of the Indian zebu, it is almost certain that their original progenitor was different from the European cattle; Experts of ability believe that the European cattle had two or three wild ancestors - but it is not known whether they deserve to be called a species.This conclusion, and that of the interspecies distinction between zebu and common cattle, have in fact been confirmed by the admirable studies of Prof. Rutimeyer.Contrary to the opinions of several writers concerning horses, I generally believe that all the horse races belong to the same species, for reasons which cannot be given here.I have bred, bred, mated, and studied the skeletons of nearly every breed of English chicken, and I feel almost certain to say that all of them are descended from the wild Indian chicken (Gallus bankiva); Mr. Bryce and others, having studied this chicken in India, have concluded that, with respect to ducks and rabbits, there are some breeds so different from each other that the evidence clearly shows that they have all been descended from the common wild ducks and rabbits. Certain writers have absurdly exaggerated to extremes the doctrine of the origin of several domestic races from a few primitive ancestors.They believed that every purebred domestic race had its own wild primitive form, even if their distinguishable characters were the slightest.According to this, at least twenty species of bison, twenty species of wild sheep, and several species of wild goat must have existed in Europe alone, and several species must have existed in England alone.Another author believes that there were as many as eleven wild species of sheep previously peculiar to England!If we remember that England now has no endemic mammals, France has only a few mammals different from those in Germany, Hungary, Spain, etc., but each of these countries has several endemic breeds of cattle, sheep, etc., so we It must be admitted that many breeds of domestic animals must have originated in Europe; otherwise where did they come from?It is the same in India.Even the domestic dog breeds of the world (which I admit have descended from several wild breeds) undoubtedly have a great deal of genetic variation; dog), or the Blenheim spaniel, etc., so unlike all wild canines, who would believe that animals closely resembling them ever existed in a state of nature?It has often been said at will, that all our dog-races have descended from the crossing of a few original species; Without the origin of the several domestic races of our country, we must admit that some extreme forms, such as the greyhound, the bloodhound, the bulldog, etc., have existed in the wild.What's more, we have exaggerated the possibility of interbreeding to produce different races.There are many instances found on record showing that, if we carefully select some individuals exhibiting our desired character, we can help to vary a race by accidental crossing; Sexuality is very difficult.Sir, J. Sebright experimented specifically for this purpose, and failed.The offspring of the first cross between two pure breeds are sometimes quite uniform in character (as I have found with pigeons), so that everything seems simple; After several generations of crosses, so few of the offspring are like each other, the job is obviously difficult. Breeds of domestic pigeons, their differences and origins I believe that research with special groups is the best method, and after careful consideration, domestic pigeons were selected.I have bred every breed I could buy or acquire, and I have received generous donations of various skins from many parts of the world, especially from the Honorable Hon. W. Elliot from India, the Honorable Murray (Hon. C. Murray) sent from Persia.Many treatises have been published on pigeons in several different languages, some of them very old and therefore of great importance.I have been associated with several well-known pigeon breeders, and have been admitted to two London clubs, and the variety of pigeons is astonishing.From a comparison of the English carrier and the Short-faced tumbler there is seen a curious difference between their beaks, and consequently in their skulls.The messenger pigeon, especially the male, has a curiously developed fleshy protuberance on the skin around the head; with this are very long eyelids, very large external nostrils, and a wide mouth.The shape of the beak of the short-faced tumbler is almost like that of a finch; and the common tumbler has a curious hereditary habit of flying high in dense flocks and doing somersaults.The dwarf pigeon (runt) has a huge body, a thick long beak, and large feet; some subspecies of the dwarf pigeon have a very long neck; some have very long wings and tails, and some have very short tails.The Barbary pigeon (barb) is similar to the messenger pigeon, but the beak is not long, but short and broad.The body, wings, and legs of the pouter are exceptionally long, and the crop is unusually well developed, which can amaze and even amuse when it swells triumphantly.The turbit, with its short, conical beak, and a row of inverted feathers under its breast, has a habit of constantly dilating the upper part of its esophagus slightly.The feathers of the jacobin stand upside down along the back of the neck in a pocket; judging from the size and proportion of the body, its wing-feathers and tail-feathers are quite long.The calls of the trumpeter and laugher are, as their names suggest, very different from those of other breeds.The fantail (falntail) has thirty or even forty tail-feathers instead of twelve or fourteen--the normal number of tail-feathers in all members of the large pigeon family; , The fine breed can touch the head and the tail, and the fat glands are very degenerated.In addition, some varieties with relatively small differences can also be cited. There are several varieties of bones, and the development of the length, width, and curvature of the facial bones is quite different.There is highly marked variation in the raster shape, as well as in width and length, of the mandible.The number of caudal and sacral vertebrae varies; the number of ribs also varies, as do their relative widths and the presence or absence of protrusions.The size and shape of the holes in the sternum are highly variable; so are the openings and relative lengths of the two branches of the furcula.Relative width of cleft, relative length of eyelids, nostrils, tongue (not always strictly related to beak length), size of crop and upper esophagus; development and degeneration of fat glands; first row of wing feathers and The number of tail-feathers; the length of wings and tail relative to each other and to the body; the relative length of legs to feet;There is variation in the period when the feathers are fully fledged, as is the state of the fluff of the young pigeons after hatching.Eggs vary in shape and size.There are marked differences in the flying posture and in some breeds the voice and temperament.Finally, there are certain breeds in which the males and females differ slightly from each other. In all, at least twenty species of pigeons could be selected, which, if shown to an ornithologist, and told that they are wild birds, he would certainly rank them among a well-defined species.Again, I do not believe that any ornithologist would in such a case place the English carrier, the short-faced tumbler, the dwarf, the barbary, the pout, and the fantail into the same genus; This is especially the case when several purely hereditary sub-breeds of these breeds—what he would call species—are shown to him. Great are the differences between the breeds of pigeons, but I am fully convinced that the general opinion of naturalists is correct, that they are all descended from the rock-pigeon (Columba livia), which under the name also includes Several endemic races, or subspecies, that differ only slightly from each other.Since some of the reasons that lead me to this belief apply to some extent to other situations as well, I shall state them here in general terms.If these several breeds were not varieties, and were not descended from the rock-pigeon, they must have descended from at least seven or eight original progenitors; Of domestic breeds; for example, if two breeds are crossed, how can a protruding pigeon be produced if one of the parents does not have the character of crop?These presumed primitive ancestors must have been rock pigeons, which neither bred nor liked to perch in trees.But, besides this rock-pigeon and its geographical subspecies, only two or three other wild rock-pigeons are known, and none of them possesses any character of the domestic breeds.Those primordial progenitors, therefore, are two possibilities: either the place where the pigeon was first domesticated is still alive, unknown to ornithologists; Not unknown; or they must all have become extinct in the wild.But the easy-flying bird that breeds on rocky cliffs is unlikely to be extinct; and the common rock-pigeon, having the same habits as the domestic breeds, is found even on some of the smaller English islands, or on the shores of the Mediterranean. Not extinct.It seems, therefore, a rash conjecture to suppose that so many species, having the like habits of the domestic breeds, have become extinct.Also, several of the above-mentioned domestic breeds have been transported all over the world, so some must have been brought back to their country of origin; No species has become wild except where it has become wild.Moreover, all recent experience shows that it is difficult to keep wild animals breeding freely under domestication; yet, on the basis of the multi-source of the domestic pigeon, it must be assumed that at least seven or eight species were completely destroyed by barbarians in ancient times. It is domesticated and can reproduce in large numbers in cages. It is a strong argument, and applicable in several other cases, that while the above-mentioned breeds generally correspond to the wild rock-pigeon in constitution, habits, voice, colour, and most of the structure, some other parts must be highly abnormal.的；我们在鸠鸽类的整个大科里，找不到一种像英国传书鸽的、或短面翻飞鸽的、或巴巴利鸽的喙；像毛领鸽的倒羽毛；像突胸鸽的嗉囊；像扇尾鸽的尾羽。因此必须假定，不但半开化人成功地彻底驯化了几个物种，而且他们也有意识地或者偶然地选出了特别畸形的物种；此外，还必须假定，这些物种以后都完全绝灭了或者湮没无闻了。看来这许多奇怪的意外之事是完全不会有的。 有关鸽类颜色的一些事实值得考察：岩鸽是石板青色的，腰部白色；但是印度的亚种——斯特里克兰的青色岩鸽（C. intermedia）的腰部却是青色的，岩鸽的尾端有一暗色横带，外侧尾羽的外缘基部呈白色。翅膀上有两条黑带。一些半家养的品种和一些真正的野生品种，翅上除有两条黑带之外，更杂有黑色方斑。全科的任何其他物种都不同时具有这几种斑纹。在任何一个家养品种里，只要是充分养得好的鸽子，所有上述斑纹，甚至外尾羽的白边，有时都是充分发达的。而且，当两个或几个不同品种的鸽子进行杂交后，虽然它们不具有青色或上述斑纹，但其杂种后代却很容易突然获得这些性状。现在把我观察过的几个例子之一述说于后：我用几只极其纯粹繁殖的白色扇尾鸽同几只黑色巴巴利鸽进行杂交——巴巴利鸽的青色变种是如此稀少，我不曾听到过在英国有一个这样事例——它们的杂种是黑色、褐色和杂色的。我又用一只巴巴利鸽同斑点鸽（spot ）进行杂交，斑点鸽是白色的，尾红色，额部有一红色斑点，这是众所周知的极其纯粹繁殖的品种；而它们的杂种却呈暗黑色并具斑点。随后我用巴巴利鸽和扇尾鸽之间一个杂种，同巴巴利鸽和斑点鸽之间的一个杂种进行杂交，它们产生了一只鸽子，具有任何野生岩鸽一样美丽的青色、白腰、两条黑色的翼带以及具有条纹和白边的尾羽！假如说一切家养品种都是从岩鸽传下来的，根据熟知的返祖遗传原理，我们就能够理解这些事实了。但是，如果我们不承认这一点，我们就必须采取下列两个完全不可能的假设之一。第一，所有想像的几个原始祖先，都具有岩鸽那样的颜色和斑纹，所以各个品种可能都有重现同样颜色和斑纹的倾向，可是没有一个别的现存物种具有这样的颜色和斑纹。第二，各品种，即使是最纯粹的，也曾在十二代、或至多二十代之内同岩鸽交配过：我说在十二代或二十代之内，是因为不曾见到一个例子表明杂种后代能够重现二十代以上消失了的外来血统的祖代性状。在只杂交过一次的品种里，重现从这次杂交中得到的任何性状的倾向，自然会变得愈来愈小，因为在以后各代里外来血统将逐渐减少；但是，如果不曾杂交过，则在这个品种里，就有重现前几代中已经消失了的性状的倾向，因为我们可以看出，这一倾向同前一倾向正好完全相反，它能不减弱地遗传到无数代。论述遗传问题的人们常常把这两种不同的返祖情形混淆在一起了。 最后，根据我自己对于最不同的品种所作的有计划的观察，我可以说，所有鸽的品种间杂种都是完全能育的，然而两个十分不同的动物种的种间杂种，几乎没有一个例子能够确切证明，它们是完全能育的，有些著者相信，长期继续的家养能够消除种间不育性的强烈倾向，根据狗以及其他一些家养动物的历史来看，如果把这一结论应用于彼此密切近似的物种，大概是十分正确的。但是，如果把它扯得那末远，以假定那些原来就具有像今日的传书鸽、翻飞鸽、突胸鸽和扇尾鸽那样显著差异的物种，还可以在它们之间产生完全能育的后代，那就未免过于轻率了。 根据这几个理由——人类先前不可能曾使七个或八个假定的鸽种在家养状况下自由地繁殖；——这些假定的物种从未在野生状态下发现过，而且它们也没有在任何地方变为野生的；——这些物种，虽然在多方面如此像岩鸽，但同鸽科的其他物种比较起来，却显示了某些极变态的性状；——无论是在纯粹繁育或杂交的情况下，一切品种都会偶尔地重现青色和各种黑色斑纹；——最后，杂种后代完全能生育；——把这些理由综合起来，我们可以稳妥地论断，一切家养品种都是从岩鸽及其地理亚种传下来的。 为了支持上述观点，我补充如下：第一，已经发现野生岩鸽在欧洲和印度能够家养：并且它们在习性和大多数构造的特点上和一切家养品种相一致。第二，虽然英国传书鸽或短面翻飞鸽在某些性状上和岩鸽大不相同，然而，把这两个族的几个亚品种加以比较，特别是把从远地带来的亚品种加以比较，我们可以在它们和岩鸽之间造成一条几乎完整的系列；在其他场合里我们也能做到这样，但不是在一切品种里都能做到这样。第三，每一品种的主要赖以区别的性状都是显著容易变异的，如传书鸽的肉垂和喙的长度，翻飞鸽的短喙，扇尾鸽的尾羽数目；对于这一事实的解释，等我们论到“选择”的时候便会明白了。第四，鸽类曾受到许多人极细心的观察、保护和爱好。它们在世界的若干地方被饲育了数千年；关于鸽类的最早记载，如来普修斯教授（Prof. Lepsius）曾经向我指出的，约在公元前3000年埃及第五皇朝的时候；但伯奇先生（Mr. Birch）告诉我说，在此之前的一个皇朝已有鸽名记载在菜单上了。在罗马时代，照普利尼（Pliny）所说的，鸽的价格极高；“而且，他们已经达到了这种地步，他们已经能够核计它们的谱系和族了”。印度亚格伯汗（Akbar Khan）非常重视鸽，大约在1600年，养在宫中的鸽就不下两万只，宫廷史官写道：“伊朗王和都伦王曾送给他一些极稀有的鸽”；又写道：“陛下使各种类进行杂交，前人从未用过这方法，这把它们改良到可惊的程度。”差不多在这同一时代，荷兰人也像古罗马人那样也爱好鸽子、这些考察对解释鸽类所发生的大量变异是无上重要的，我们以后讨论“选择”时就会明白了。同时我们还可知道，为什么这几个品种常常具有畸形的性状。雄鸽和雌鸽容易终身相配，这也是产生不同品种的最有利条件；这样，就能把不同品种饲养在一个鸟槛里了。 我已对家养鸽的可能起源作了若干论述，但还是十分不够的；因为当我最初养鸽并注意观察几类鸽子的时候，清楚地知道了它们能够多么纯粹地进行繁育，我也充分觉得很难相信它们自从家养以来都起源于一个共同祖先，这正如任何博物学者对于自然界中的许多雀类的物种或其他类群的鸟，要作出同样的结论，有同样的困难。有一种情形给我印象很深，就是几乎所有的各种家养动物的饲养者和植物的栽培者（我曾经和他们交谈过或者读过他们的文章），都坚信他们所养育的几个品种是从很多不同的原始物种传下来的。像我曾经询问过的那样，请你也向一位知名的赫里福德（Hereford ）的饲养者问一问：他的牛是否从长角牛传下来的，或是二者是否都来自一个共同祖先，其结果将受到嘲笑。我从未遇见过一位鸽、鸡、鸭或兔的饲养者，不充分相信各个主要品种是从一个特殊物种传下来的。凡蒙斯（Van Mons）在他的关于梨和苹果的论文里，全然不信几个种类，如“立孛斯东·皮平”（Ribston-Pippin）苹果或“考得林”苹果（Codlin-apple），能够从同一株树上的种籽生出来。其他例子不胜枚举，我想，解说是简单的：根据长期不断的研究，他们对几个族间的差异获得了强烈的印象；他们熟知各族微有变异，因为他们选择此等轻微差异而得到了奖赏，但是他们对于一般的论点却是一无所知，而且也不肯在头脑里把许多连续世代累积起来的轻微差异综合起来。那些博物学者所知道的遗传法则，比饲养者所知道的还少得多，同时对于悠长系统中的中间环节的知识也不比饲养者知道得多些，可是他们都承认许多家养族是从同一祖先传下来的——当他们嘲笑自然状态下的物种是其他物种的直系后代这个观念时，难道不应该学一学谨慎这一课吗？ 古代所依据的选择原理及其效果 现在让我们对于家养族是从一个物种或从几个近似物种产生出来的步骤简要地讨论一下。有些效果可以归因于外界生活条件的直接和一定的作用，有些效果可以归因于习性；但是如果有人用这等作用来说明驾车马和赛跑马、长驱猎狗和嗅血警犬、传书鸽和翻飞鸽之间的差异，那就未免冒失了。我们的家养族的最显著的特色之一，是我们所看到的它们确实不是适应动物或植物自身的利益，而是适应人的使用或爱好。有些于人类有用的变异大概是突然发生的，即一步跃进的；例如，许多植物学者相信，生有刺钩的恋绒草（fullers teasel）——这些刺钩是任何机械装置所不及的——只是野生川续断草（Dipsactus）的一个变种而已，而且这种变化可能是在一株实生苗突然发生的。矮脚狗（turnspit dog）大概也是这样起源的；我们知道安康羊（Ancon sheep）的情形也是如此。但是，当我们比较驾车马和赛跑马、单峰骆驼和双峰骆驼、适于耕地和适于山地牧场的、以及毛的用途各异的不同种类的绵羊时；当我们比较以各种用途为人类服务的许多狗品种时，当我们比较如此顽强争斗的斗鸡和很少争斗的品种时，比较斗鸡和从来不孵卵的卵用鸡时，比较斗鸡和极其小形而美丽的矮鸡（bantam）时，当我们比较无数的农艺植物、蔬菜植物、果树植物以及花卉植物的族时，它们在不同的季节和不同的目的上最有益于人类，或者如此美丽非凡而赏心悦目；我想，我们必须于变异性之外，作更进一步的观察。我们无法想像一切品种都是突然产生的，而一产生就像今日我们所看到的那样完善和有用；的确，在许多情形下，我们知道它们的历史并不是这样的。这关键就在于人类的积累选择的力量；自然给与了连续的变异，人类在对他们自己有用的一定方向上积累了这些变异。在这种意义上，才可以说人类为自己制造了有用的品种。 这种选择原理的伟大力量不是臆想的。确实有几个优秀的饲养者，甚至在一生的时间里，就大大地改变了他们的牛和绵羊品种。要充分理解他们所干的是些什么，阅读若干关于这个问题的论文，和实际观察那些动物，几乎是必要的，饲养者习惯地说动物的体制好像是可塑性的东西，几乎可以随意塑造。如果有篇幅，我能从极其有才能的权威者的著作中引述许多关于这种效果的记载。尤亚特（Youatt ）对农艺家们的工作，可能比几乎任何别人更为通晓，而且他自己就是一位极优秀的动物鉴定者，他说选择的原理“可以使农学家不仅能够改变他的畜群性状，而且能够使它们发生完全的变化。选择是魔术家的魔杖，用这只魔杖，可以随心所欲地把生物塑造成任何类型和模式”。萨默维尔勋爵（Lord Somerville ）谈到饲养者养羊的成就时，曾说：“好像他们用粉笔在壁上画出了一个完美的形体，然后使它变成为活羊。”在撒克逊，选择原理对于美利奴羊（merino sheep ）的重要性已被充分认识，以致人们把选择当做一种行业：把绵羊放在桌子上，研究它，就像鉴赏家鉴定绘画那样；在几个月期间内，一共举行三次，每次在绵羊身上都作出记号并进行分类，以便最后选择出最优良的，作为繁育之用。 英国饲养者所得到的实际成就，可以从价格高昂的优良谱系的动物来证明；这些优良动物几乎被运送到世界各地去。这种改良，一般决不是由于不同品种的杂交；一切最优秀的饲养者都强烈地反对这样的杂交，除了有时行于密切近似的亚品种之外。而且在杂交进行以后，严密的选择甚至比在普通场合更不可缺少。如果选择仅仅在于分离出某些很独特的变种，使它繁殖，那末这一原理很明显地就几乎不值得注意了；但它的重要性却在于使未经训练过的眼睛所绝对觉察不出的一些差异——我就觉察不出这些差异——在若干连续世代里，向一个方向累积起来而产生出极大的效果。在一千人里不见得有一个具有准确的眼力和判断力，能成为一个卓越的饲养家。如果赋有此等品质，并且多年研究他的课题，同时以不屈不挠的耐性终生从事这一工作，他就会得到成功，而且能作出巨大改进；如果他一点也不具有这些品质，则必定要失败。很少人会立即相信，甚至要成为一个熟练的养鸽者，也必须有天赋的才能和多年的经验。 园艺家也依据相同的原理；但植物比动物的变异常常更是突发的。没有人会假定我们最精选的生物，是从原始祖先由一次变异而产生的。在若干场合我们有正确的记录可做证明；如普通醋栗（Common gooseberry）的大小是逐渐增加的，就是一个很小的例证。把今日的花同仅仅十年或三十年前所画的花相比较，我们就可看到花卉栽培家对许多花做出了可惊的改进。当一个植物的族一旦很好地固定下来以后，种籽繁育者并不是采选那些最好的植株，而仅仅是巡视苗床，拔除那些“无赖汉”，他们把那些脱离固有标准型的植株叫做无赖汉。对于动物，事实也同样采用这种选择方法；无论何人，都不会这样粗心大意，用最劣的动物去进行繁殖。 关于植物，还有另一种方法可以观察选择的累积效果——就是在花园里比较同种的不同变种的花所表现的多样性；在菜园里把植物的叶、荚、块茎或任何其他有价值部分，在与同一变种的花相比较时所表现的多样性；在果园里把同种的果实在与同种的一些变种的叶和花相比较时所表现的多样性。看看甘蓝的叶是何等相异，而花又是何等极其相似；三色堇的花是何等相异，而叶又是何等相似；各类醋栗果实的大小、颜色、形状、茸毛是何等相异，而它们的花所表现的差异却极其微小。这并不是说在某一点上差异很大的变种就在一切其他各点上就全无差异；我经过慎重观察之后才说这种情形是决无仅有的。相关变异法则的重要性决不可忽视，它能保证某些变异的发生；但是，按照一般法则，我们无论对于叶、花还是对于果实的微小变异进行连续选择，就会产生出主要在这些性状上有所差异的族，这是无可怀疑的。 选择原理成为有计划的实践差不多只有七十五年的光景，这种说法也许有人反对。近年来对于选择的确比以前更加注意，并且关于这一问题，发表了许多论文，因而其成果也相应地出得快而且重要。但是，要说这一原理是近代的发现，就未免与真实相距甚远了。我可以引用古代著作中若干例证来说明那时已经认识了这一原理的充分重要性。在英国历史上的蒙昧未开化时代，常有精选的动物输入，并且制订过防止输出的法律；明令规定，马的体格在一定尺度之下就要加以消灭，这同艺园者拔除植物的“无赖汉”可以相比。我看到一部中国古代的百科全书清楚记载着选择原理。有些罗马古代著作家们已经拟定了明确的选择规则。从创世纪的记载里，可以清楚地知道在那样早的时期已经注意到家养动物的颜色了。未开化人现在有时使他们的狗和野生狗类相杂交，以改进狗的品种，他们从前也曾这样做过，这可以在普利尼的文章里得到证实。南非洲的未开化人依据挽牛的颜色使它们交配，有些爱斯基摩人对于他们的驾车狗也这样做。利文斯登（Livingstone）说，未曾与欧洲人接触过的非洲内地的黑人极重视优良的家畜。某些这种事实虽然并不表示真正的选择已在实行，但它们表示了在古代已经密切注意到家养动物的繁育，而且现今的最不开化的人也同样注意这一点，既然好品质和坏品质的遗传如此明显，要是对于动、植物的繁育还不加注意，那的确是一件奇怪的事了。 无意识的选择 日前，优秀的饲养者们都按照一种明确的目的，试图用有计画的选择来形成优于国内一切种类的新品系或亚品种。但是，为了我们的讨论目的，还有一种选择方式，或可称为无意识的选择，更为重要。每个人都想拥有最优良的个体动物并繁育它们，这就引起了这种选择。例如，要养向导狗（pointer）的人自然会竭力搜求优良的狗，然后用他自己拥有的最优良的狗进行繁育，但他并没有持久改变这一品种的要求或期待。然而，我们可以推论，如果把这一程序继续若干世纪，将会改进并且改变任何品种，正如贝克韦尔（Bakewell ）、科林斯（Collins）等等根据同样的程序，只是进行得更有计划些，便能在他们一生的时期内大大地改变了他们的牛的体型和品质。除非在很久以前，对问题中的品种就进行正确的计量或细心的描绘，以供比较，缓慢而不易觉察的变化就永远不能被辨识。然而，在某些情形下，同一品种的没有变化的或略有变化的个体生存在文明落后的地区也是有的，在那里品种是很少改进的。有理由相信，查理斯王的长耳猎狗自从那一朝代以来已经无意识地大大被改变了。某些极有才能的权威家相信，侦犬（setter）直接来自长耳猎狗，大概是在徐徐改变中产生的。我们知道英国的向导狗在上一世纪内发生了重大变化，并且人们相信这种变化的发生主要是和猎狐狗（fox hound）杂交所致；但是和我们的讨论有关系的是：这种变化是无意识地、缓慢地进行着的，然而效果却非常显著，虽然以前的西班牙向导狗确实是从西班牙传来的，但博罗先生（Mr. Borrow ）告诉我说，他没有看见过一只西班牙本地狗和我们的向导狗相像。 经过同样的选择程序和细心的训练，英国赛跑马的体格和速度都已超过了亲种阿拉伯马，所以，依照古特坞赛马的规则，阿拉伯马的载重量被减轻了。斯潘塞勋爵和其他人曾经指出，英格兰的牛同先前养在这个国家的原种相比较，其重量和早熟性都大大增加了。把各种旧论文中论述不列颠、印度、波斯的传书鸽、翻飞鸽的过去和现在的状态加以比较，我们便可以追踪出它们极缓慢地经过的各个阶段，通过这些阶段，而到达了和岩鸽如此大不相同的地步。 尤亚特举了一个最好的例证说明一种选择过程的效果，这可以看作是无意识的选择，因为饲养者没有预期过的、或甚至没有希望过的结果产生了。这就是说，产生了两个不同的品系。尤亚特先生说，巴克利先生（Mr. Buckley）和伯吉斯先生（Mr. Burgess）所养的两群莱斯特绵羊（Leicester sheep）“都是从贝克韦尔先生的原种纯正繁殖下来的，论时间已在五十年以上。熟悉这一问题的任何人都完全不会怀疑，上述任何一个所有者曾在任何情况下把贝克韦尔先生的羊群的纯粹血统搞乱，但是这二位先生的绵羊彼此间的差异却如此之大，以致它们的外貌就像完全不同的变种。 如果现在有一种未开化人，很野蛮，甚至从不考虑家养动物后代的遗传性状，然而当他们常常遇到饥谨或其他灾害时，他们还会把合乎任何特殊目的的特别对他们有用的动物小心地保存下来。这样选取出来的动物比起劣等动物一般都会留下更多的后代；所以这样，一种无意识的选择便在进行了。我们知道，甚至火地岛（Tierra del Fuego）的未开化人也重视他们的动物，在饥荒的时候，他们甚至杀吃年老妇女，在他们看来，这些年老妇女的价值并不比狗高。 在植物方面，通过最优良个体的偶然保存可以逐渐得到改进，不论它们在最初出现时是否有足够的差异可被列入独特的变种。也不论是否由于杂交把两个或两个以上的物种或族混合在一起，都可以清楚地辨识出这种改进过程。我们现在所看到的诸如三色堇、蔷薇、天竹葵、大理花以及其他植物的一些变种，比起旧的变种或它们的亲种，在大小和美观方面都有所改进，从来没有人会期望从野生植株的种籽得到上等的三色堇或大理花。也没有人会期望从野生梨的种籽培育出上等软肉梨，即使他可能把野生的瘦弱梨苗培育成佳种，如果这梨苗本来是从栽培系统来的。在古代虽有梨的栽培，但据普利尼的描述看来，它们的果实品质是极劣的。我曾看到园艺著作中对于园艺者的惊人技巧表示惊叹，他们能从如此低劣的材料里产生出如此优秀的结果。不过这技术是简单的，就其最终结果来说，几乎都是无意识地进行的。这就在于永远是把最有名的变种拿来栽培，播种它的种籽，当碰巧有稍微较好的变种出现时，便进行选择，并且这样继续进行下去。但是，我们的最优良果实在某种很小程度上虽然有赖于古代艺园者自然地进行选择和保存他们所能寻得的最优良品种，然而他们在栽培那些可能得到的最好梨树时，却从未想到我们要吃到什么样的优良果实。 正如我所相信的，这样缓慢地和无意识地累积起来的大量变化，解释了以下的熟知事实：即在许多情形下，我们对于花园和菜园里栽培悠久的植物，已无法辨认其野生原种。我们大多数的植物改进到或改变到现今于人类有用的标准需要数百年或数千年，因此我们就能理解为什么无论澳大利亚、好望角或十分未开化人所居住的地方，都不能向我们提供一种值得栽培的植物。拥有如此丰富物种的这些地区，并非由于奇异的偶然而没有任何有用植物的原种，只是因为该地植物还没有经过连续选择而得到改进，以达到像古文明国家的植物所获得的那样完善的程度。 关于未开化人所养的家养动物，有一点不可忽略，就是它们至少在某些季节里，几乎经常要为自己的食物而进行斗争。在环境极其不同的两个地区，体质上或构造上微有差异的同种个体，在这一地区常常会比在另一地区生活得好些；这样，由于以后还要加以更充分说明的“自然选择”的过程，便会形成两个亚品种。这种情形或者可以部分地说明为什么未开化人所养的变种，如某些著者说过的，比在文明国度里所养的变种，具有更多的真种性状。 根据上述人工选择所起的重要作用来看，即刻可以明了我们家养族的构造或习性为什么会适应于人类的需要或爱好。我想，我们还能进一步理解，我们家养族为什么会屡屡出现畸形的性状，为什么外部性状所表现的差异如此巨大，而相对地内部器官所表现的差异却如此微小。除了可以看得见的外部性状外，人类几乎不能选择、或只能极其困难地选择构造上的任何偏差；其实他们对内部器官的偏差是很少注意的。除非自然首先在一定程度上向人类提供一些轻微变异，人类决不能进行选择。在一个人看到一只鸽子尾巴在某种轻微程度上已发育成异常状态之前，他不会试图育出一种扇尾鸽；在他看到一只鸽的嗉囊的大小已经有些异乎寻常之前，他也不会试图育出一种突胸鸽；任何性状，在最初发现时愈畸形或愈异常，就愈能引起人的注意。但是，我毫不怀疑要用人类试图育出扇尾鸽的这样说法，是完全不正确的。最初选择一只尾巴略大的鸽子的人；决不会梦想到那只鸽子的后代经过长期连续的、部分是无意识选择和部分是有计划选择之后，会变成什么样子。一切扇尾鸽的始祖恐怕只有略微展开的十四枝尾羽，就像今日的爪哇扇尾鸽那样，或者像其他独特品种的个体那样地具有十七枝尾羽。最初的突胸鸽嗉囊的膨胀程度并不比今日浮羽鸽食管上部膨胀程度为大，而浮羽鸽的这种习性并不被一切养鸽者所注意，因为它不是这个品种的主要特点之一。 不要以为只有某种构造上的大偏差才能引起养鸽者的注意，他能觉察极小的差异，而且人类本性就在于对他的所有物的任何新奇，即使是轻微的，也会予以重视。决不可用几个品种已经固定后的现今价值标准，去对以前同一物种诸个体的轻微差异所给予的价值进行判断。我们知道鸽现在还会发生许多轻微的变异，不过此等变异却被当作各品种的缺点、或离开完善标准的偏差而遭舍弃。普通鹅没有产生过任何显著的变种；图卢兹（Toulouse ）鹅和普通鹅只在颜色上有所不同，而且这种性状极不稳定，但近来却被当做不同品种在家禽展览会上展览了。 这些观点，对于时常说起的——即我们几乎不知道任何家畜的起源或历史的说法，似乎可以给予解释了。但是实际上，一个品种好像语言里的一种方言一样，几乎无法说它有明确的起源。人保存了和繁育了构造上微有偏差的个体，或者特别注意了他们的优良动物的交配，这样便改进了它们，并且已改进的动物便慢慢地会传布到邻近的地方去。但是它们很少有一定的名称，而且对于它们的价值也很少重视，所以它们的历史就要遭到忽视。当通过同样的缓慢而逐渐的过程得到进一步改进的时候，它们将传布的更远，并且会被认为是特殊的和有价值的种类，在这时它们大概才开始得到一个地方名称。在半文明的国度里，交通还不太发达，新亚品种的传布过程是缓慢的，一旦有价值的各点被人认识后，我称之为无意识选择的原理就会常常倾向于慢慢地增加这一品种的特性，不论那特性是什么；品种的盛衰依时尚而定，恐怕在某一时期养得多些，在另外时期养得少些；依照居民的文明状态，恐怕在某一地方养得多些，在另外一地方养得少些。但是，关于这种缓慢地、不定的、不易觉察的变化的记载，很少有机会被保留下来。 人工选择的有利条件 我现在要稍微谈谈人工选择的有利的或不利的条件。高度的变异性显然是有利的，因为它能大量地向选择供给材料，使之顺利发生作用；即使仅仅是个体差异，也是充分够用的，如能给予极其细心的注意，也能向着几乎任何所希望的方向积累起大量变异。但是，因为对于人们显著有用的或适合他们爱好的变异只是有时偶然出现，所以个体如果饲养的愈多，变异出现的机会也就愈多。因此，数量对于成功来说，是高度重要的。马歇尔（Marshall ）曾依据这一原理对约克郡各地的绵羊作过如下叙述：“因为绵羊一般为穷人所有，并且大都只是小群的，所以它们从来不能改进。”与此相反，艺园者们栽培着大量的同样植物，所以他们在培育有价值的新变种方面，就比业余者一般能得到更大的成功。一种动物或植物的大群个体，只有在有利于它们繁殖的条件下才能被培育起来。如果个体稀少，不管它们的品质怎样，都得让其全部繁育，这就会有效地妨碍选择。但最重要的因素大概是，人类必须高度重视动物或植物的价值，以致对品质或构造上的最微小偏差都会给予密切注意；要是没有这样的注意，就不会有什么成效了。我曾见到人们严肃地指出，正好在艺园者开始注意草莓的时候，它开始变异了，这就是极大的幸运。草莓自被栽培以来，无疑是经常发生变异的，不过对微小的变异未曾给予注意罢了。然而，一旦艺园者选出一些个体植株，它们具有稍微大些的、稍微早熟些的或稍微好些的果实，然后从它们培育出幼苗，再选出最好的幼苗，并用它们进行繁育，于是（多少在种间杂交的帮助下），许多可赞美的草莓变种就被培育出来了，这就是近半世纪来所培育出的草莓变种。 在动物方面，防止杂交是形成新族的重要因素，至少在已有其他动物族的地方是如此。关于这一点，把土地封闭起来是有作用的。漂泊的未开化人，或者开阔平原上的居住者，所饲养的同一物种很少有超过一个品种的。鸽能终身配合，这对于养鸽者大有便利。因此，它们虽混养在一个鸽槛里，许多族还能改进并能保纯；这样条件一定大有利于新品种的形成。我可以补充地说，鸽能大量而迅速地被繁殖，把劣等的鸽杀掉以供食用，自然就把它们淘汰了。相反的，猫由于有夜间漫游的习性，不容易控制它们的交配，虽然妇女和小孩喜爱它，但很少看到一个独特的品种能够长久保存；我们有时看到的那些独特品种，几乎都是从外国输入的。虽然我并不怀疑某些家养动物的变异少于另外一些家养动物的变异，然而猫、驴、孔雀、鹅等的独特品种的稀少或竟然没有，则主要是由于选择未曾起作用：猫，由于难控制其交配；驴，由于只有少数为穷人所饲养，并且很少注意它们的繁育；但是近年来在西班牙和美国的某些地方，因为仔细地进行了选择，这种动物已意外地变化了和改进了；孔雀，由于不很容易饲养，而且也没有大群的饲养；鹅，由于只在两种目的上有价值，即供食用和取羽毛，特别是由