Chapter 16 Chapter 9 The nature of hybrids 2

We now examine more closely the approximate nature of the difference between the species which gave rise to the sterility of the first cross and that of the hybrid.The degree of difficulty with regard to their union, and of obtaining offspring, in the case of the first cross, is evidently determined by several different causes.Sometimes it is impossible for the male reproductive plasm to reach the ovule due to physiological reasons, such as in plants where the pistil is too long so that the pollen tube cannot reach the ovary.We have also observed that when the pollen of one species is placed on the stigma of another distantly related species, although the pollen-tubes protrude, they do not penetrate the surface of the stigma.Furthermore, the male genitals, although they reach the female genitals, do not give rise to the formation of embryos, as seems to be the case in some of Trey's experiments with Fucus.There is as yet no explanation for these facts, any more than there can be any explanation for why some trees cannot be grafted on others. Finally, perhaps the embryo develops, but dies early.This last point has not been sufficiently noted; but Mr. Hewitt, who has a great deal of experience in the cross-breeding of pheasants and chickens, has told me in writing of observations he has made which have led me to Early death of the embryo is believed to be the most common cause of first cross sterility.Mr. Salter has recently published the results of an examination of 500 eggs produced by various crosses between three species of Gallus and their hybrids; Most of the eggs were fertilized; and in most of them the embryo was either partially developed and soon died, or nearly mature, but the chicks were unable to peck through the shell, and among the hatched chicks there were Four-fifths died within the first few days, or at the longest within a few weeks, "without any apparent cause, evidently due to mere incapacity to live"; so from 500 eggs only Raised twelve chicks.With respect to plants, the embryoid bodies of hybrids, presumably, often die in the same manner; at least we know that hybrids raised from very different species are often feeble, low, and die early; Max Wichura has recently published some notable examples of willow.It is here to be remarked that, in some cases of parthenogenesis, the embryos of unfertilized silk moth eggs, after passing through an early stage of development, die like embryos resulting from the crossing of different species.Until I had ascertained these facts, I was reluctant to believe that the embryo of hybrids often dies at an early age; for once hybrids are produced, as we have seen in the case of mules, they are generally healthy and long-lived.A hybrid, however, is under different environmental conditions before and after its generation: if hybrids arise and live where their parents live, they are generally under favorable conditions of life.But a hybrid inherits only one half of the mother's nature and constitution; so it may have been in some It is therefore liable to die at an early age; especially since all very young organisms are eminently sensitive to noxious or unnatural conditions of life.But, on the whole, the cause is more likely to lie in some defect in the original fertilization, which prevented the embryo from fully developing, than in the circumstances in which it was afterwards found.

The case seems to be quite different with regard to the sterility of hybrids with incomplete development of the germ plasm of both sexes.I have more than once presented a great number of facts, showing how easily the reproductive systems of animals and plants are severely affected, if removed from their natural condition.In fact, this is a major obstacle to the domestication of animals.There are many similarities between the sterility thus induced and that of hybrids.In both cases sterility has nothing to do with general health, and sterile individuals tend to be hypertrophied or extraordinarily lush.In both cases sterility occurs in various degrees; and the male plasm is most susceptible; but the female plasm is sometimes more susceptible than the male plasm.In both cases the tendency to sterility is to some extent consistent with the affinities of the taxonomic systems, since all groups of animals and plants are rendered sterile by the same unnatural conditions; and all groups species have a tendency to produce sterile hybrids.On the other hand, one species in a population will often resist great changes in environmental conditions without loss of fertility; and some species in a population will produce abnormally fertile hybrids, which, without experiment, no one He can say whether any particular animal can be fertile in pens, or whether any foreign plant can set seed freely under cultivation; nor can he say without experiment whether any two species of a genus can at all produce More or less sterile hybrids.Finally, if the plants are kept for several generations in conditions other than their natural ones, they are very susceptible to variation, and the cause of variation seems to be partly due to special influences on the reproductive system, though this influence is less than the cause of sterility. That effect is small.The same is true of hybrids, for the offspring of hybrids are also remarkably susceptible to variation in successive generations, as every experimenter has observed.

We can see, therefore, that the reproductive system is affected in a very similar manner when organic beings are placed under new and unnatural conditions, and when hybrids arise from the unnatural crossing of two species. , regardless of general health status.In the former case its conditions of life are disturbed, though often in such a slight degree that we do not perceive it; remains the same, but its constitution is disturbed by the mixing of two different constitutions and constitutions, including of course the reproductive system.For, when the two systems are mixed into one, there is nothing in its development, in its periodic activity, in the relation of its different parts and organs to each other, and to the conditions of life. It is almost impossible for some kind of disturbance to occur.If the hybrids were capable of crossing each other, they would transmit to their offspring the same hybrid system from generation to generation, and it is not surprising that their sterility, though varied to some extent, should not be eliminated. .There is even a tendency for their sterility to be increased, which, as has been said, is generally the result of too close inbreeding.Vicula has vigorously maintained the above view, that the sterility of hybrids is the result of the mixture of the two constitutions.

It must be admitted that there are certain facts concerning the sterility of hybrids which we do not understand from the above or any other view; The sterility of hybrids closely resembling either of the pure parents is increased.I dare not say that the foregoing arguments have touched the root of things; no explanation can be afforded why a living being, if placed under unnatural conditions, should become sterile.All I have tried to show is that two conditions, which are similar in some respects, can equally produce sterile results—in the one case from disturbed conditions of life, in the other because their institutions were disturbed by the mixing of the two.

The same parallel applies to similar, but very different, facts.It is an old and nearly universal belief that slight variations in the conditions of life are advantageous to all living beings, a belief founded on the mass of evidence which I have elsewhere cited.I have seen farmers and gardeners do this, often exchanging seeds, roots, etc., from different soils and different climates, and back again.Almost any change in habits can be of great benefit to animals during their recovery from illness.Again, with regard to both plants and animals, it has been most definitely established that crosses between more or less different individuals of the same species increase the vigor and fertility of their offspring; Inbreeding of close relatives of the species, if continued through several generations, while conditions of life remain the same, almost always results in a reduction, enfeebling, or sterility of the body.

Thus, on the one hand, slight variations in the conditions of life are advantageous to all organisms; on the other hand, a slight degree of interbreeding, that is, between males and females of the same species under slightly different conditions of life, or slightly modified, , will enhance the viability and fertility of offspring.But, as we have seen, organic beings long accustomed to certain identical conditions in a state of nature, when exposed to rather varied conditions, as in captivity, frequently become more or less sterile; and we It is well known that the cross between two forms, if they differ widely, or are of different species, will almost always produce a hybrid which is to some extent sterile.I am fully convinced that this double parallelism is by no means accidental or delusion.A man who can explain why elephants, and many other animals, are sterile in only partial captivity in their native lands, can explain the chief cause of the common barrenness of hybrids.At the same time he was able to explain why certain families of domesticated animals, often under new and inconsistent conditions, were perfectly fertile when crossed, though descended from different species which were presumably sterile when first crossed of.The above two parallel sets of facts seem to be united by some common, unidentified bond, connected essentially with the principle of life; according to Mr. Herbert Spencer, this The first principle is that life depends or exists on the continual action and reaction of various forces, which in nature are always inclined to equilibrium; when this tendency is slightly disturbed by any change, the life force is will be strengthened.

Dimorphism and trimorphism of interactions A brief discussion will here be made of this subject, which we shall find sheds some light on the problem of the nature of hybrids.Several plants belonging to different "orders" exhibit two forms, which are present in about equal numbers, and which do not differ in any way except in their reproductive organs; one form has long pistils and short stamens, and the other has Short, long stamens; these two types have pollen grains of different sizes.Trimorphic plants are of three types, differing likewise in the length of the pistils and stamens, in the size and color of the pollen-grains, and in some other respects; and each of the three types has two sets of stamens. , so there are six groups of stamens and three types of pistils in three types.These organs are so proportionate to each other in length that the stamen-half of two of the types is as high as the stigma of the third.I have shown, and this result has been confirmed by other observers, that the pollen of stamens of one type to be of equal height to fertilize the stigmas of the other is necessary, in order to obtain the full fertility of these plants.In dimorphic species, therefore, there are two unions which may be called legitimate and are fully fertile, and two unions which may be called illegitimate and more or less sterile.In trimorphic species six unions are legal, i.e. fully fertile,--twelve are illegitimate, i.e. more or less sterile.

We may observe the sterility of various dimorphic and trimorphic plants when they are illegitimately inseminated, that is to say, with pollen from stamens not equal in height to the pistils, as in The same happens in the crosses of different species, showing great degrees of variation, down to absolute and complete sterility.The degree of sterility in crosses of different species depends markedly on the favourability of living conditions, and I have found that the same is true of illegitimate unions.It is well known that if the pollen of a different species is placed on the stigma of a flower, its own pollen is subsequently placed, even after a considerable period, on the same stigma, so strongly does it act. The effect of the dominance of the land so as to generally annihilate the effect of the foreign pollen; so also with several types of pollen of the same species, when the legal pollen and the illegitimate pollen are placed on the same stigma, the former has a strong predominance over the latter. .I affirmed this on the basis of the fertilization of certain flowers, first of all I inseminated some flowers illegally, and twenty-four hours later I legally inseminated them with the pollen of a variety of a peculiar colour, and all The seedlings were all of the same colour; this indicates that the legal pollen, though applied twenty-four hours later, was able to destroy or prevent the action of the illegitimate pollen applied earlier.In addition, the mutual interaction between the same two species often has very different results.So also with trimorphic plants; for instance: the mesistylized form of Lythrum Salicaria is very easily fertilized illegally by the long-stamened pollen of the short-stylared form, and produces many seeds; but, Fertilization of short-styled forms with pollen from the long stamens of the mesostyled forms failed to produce a single seed.

In all these cases, and in others which may be added, some forms of the same undoubted species, when illegitimately combined, behave exactly as if two distinct species were crossed.This led me to observe carefully for four years the many seedlings grown from several illegitimate unions.The main result is that none of these plants, which may be called illegitimate, are fully fertile.Long-style and short-style illegitimate plants can be bred from dimorphic plants, and three illegitimate types can be bred from trimorphic plants.These plants can be legally combined in a legal manner.When this is done, there is no apparent reason why the plants should not produce as many seeds as their parents when legally fertilized.But this is not the case.The plants were all sterile, though to varying degrees.Some are so extremely and irremediably sterile that a single seed or even a seed capsule has not been produced in four years.The sterility of these illegitimate plants when combined in a legal manner may be strictly compared with that of hybrids when crossed with each other.On the other hand, if a hybrid is crossed with either of the pure parents, its sterility will generally be much lessened; and the same is the case when an illegitimate plant is fertilized by a legal one.Just as the sterility of hybrids and the difficulties of the first cross between two parents are not always parallel, so some illegitimate plants are extremely sterile, but the sterility of the union that produced them Sex is by no means big.Hybrids, more so illegitimate plants, vary inherently in the degree of sterility raised from the same seed-capsule.Finally, many hybrids bloom profusely and long, but others, which are more sterile, flower less, and they are feeble and wretchedly small; and the illegitimate offspring of various dimorphic and trimorphic plants have exactly the same situation.

In short, the illegitimate plants and hybrids share the closest identity in character and habits.That is to say, illegal plants are hybrids, but such hybrids are produced by improper unions of certain types within the same species, while ordinary hybrids are produced by improper unions between so-called different species. , which is hardly an exaggeration.We have also seen that the first illegitimate unions, and the first crosses of different species, have in every respect a very close resemblance.To illustrate, or to make clearer, an illustration; let us suppose that a botanist has discovered two distinct (and in fact there are) varieties of the long-styled form of the trimorphic purple Phyllanthus, and that he decides to cross the to test whether they are different species.He will probably find that they produce only one-fifth as many seeds as normal, and that in the other respects mentioned above they appear to be two different species.But, to confirm this, he raised plants from the seeds of his hypothetical hybrids, whereupon he found that the seedlings were wretchedly small and extremely sterile, and that they behaved in other respects like ordinary hybrids. .He would then declare that he had indeed proved, on the general view, that his two varieties were real and distinct species, as any in the world; but he was quite mistaken.

Some of the above facts concerning dimorphic and trimorphic plants are important, first, because it illustrates that physiological tests of first-cross fertility and of reduced fertility in hybrids are not safe for differentiating species. Criterion: Second, because we can conclude that there is some unknown bond connecting the sterility of illegitimate unions with the sterility of their illegitimate offspring, and leads us to extend the same view to the first cross and hybrids; thirdly, because we see that the same species may exist in two or three forms, which do not differ in any way in structure or constitution in relation to external conditions, but which in certain ways When combined, they are sterile, which seems to me especially important.For we must remember that it is precisely the union of the sexes of two individuals of the same type, e.g., of two long-styled types, that produces sterility; It is precisely the combination of the male and female reproductive qualities inherent in the two different types.The case, therefore, appears at first to be the reverse of that of the ordinary unions of individuals of the same species, and of the crosses of different species.Whether this is the case, however, is doubtful; but I shall not dwell here on this ambiguous question. At any rate, we may presumably infer from the examination of dimorphic and trimorphic plants that the sterility of hybrids of different species and of their hybrid offspring is entirely determined by the nature of the male and female germ plasm, rather than by the structural or any difference in general constitution.The same conclusion may be drawn from a consideration of reciprocal crosses, in which males of one species are unable or with great difficulty to unite with females of a second species, and yet cross in reverse. But it's totally easy.That excellent observer, Gartner, has similarly concluded that the sterility of the crosses of species is due only to differences in their reproductive systems. Fertility of varieties crossed and of their hybrid offspring not universal As a very plausible argument, it may be maintained that there must be some essential difference between species and varieties, since varieties, however much they may differ in appearance from one another, can cross with considerable ease, and produce perfectly fertile species. offspring.With some exceptions to be mentioned, I fully admit that this is the rule.But much difficulty surrounds the subject, for, when looking for varieties produced under a state of nature, if two forms, hitherto regarded as varieties, are found to have any degree of sterility in crosses, it is very difficult to find them. Most naturalists would immediately rank them as species.For example, the blue and red weeds, which are considered varieties by most botanists, are, according to Gartner, so sterile in crosses that he ranks them as undoubted species.If we proceed in this circle of debate, we must necessarily admit that all varieties arising under natural conditions are fertile. If we turn to some of the varieties which have arisen, or are supposed to have arisen, under domestication, we are still involved in some doubts.For, for example, when we say that certain native domestic dogs of South America cannot easily combine with European dogs, an interpretation arises in everyone's mind, and it is probably the correct one, that these dogs were originally are descended from different species.But the complete fertility of many domestic races, such as the pigeon or the cabbage, which differ widely in appearance, is a remarkable fact, especially when we remember how many species, though most closely allied to each other, But when crossed they are extremely sterile; this is a more noteworthy fact.The fertility of domestic varieties, however, is not so surprising, from the following considerations.First, It may be observed that the amount of external difference between two species is not a sure index of the degree of their mutual sterility, so that in the case of varieties the external difference is not a sure index either, as regards species, The cause must lie entirely in their reproductive systems, and the changing conditions of life which act upon domesticated animals and cultivated plants have little tendency to alter their reproductive systems to induce mutual sterility, so we have good grounds for admitting The directly opposite doctrine of Pallas, that the conditions of domestication generally eliminate the tendency to sterility; thus, species in a state of nature, when crossed, would probably be to some extent sterile, but their domesticated offspring, when crossed, would be sterile. will become fully fertile.In plants, cultivation has no tendency to produce sterility between different species, and in the several well-documented cases already mentioned, certain plants have been adversely affected, for they have become self-fertile. fertile, while still retaining the ability to fertilize and be fertilized by other species.If Pallas' theory that sterility is eliminated by long-continued domestication is accepted (which is almost indisputable), it is highly improbable that long-continuation of the same conditions of life would likewise induce sterility ; even in some cases, species possessed of a peculiar constitution, by which occasional sterility has occurred.We can thus understand, as I believe, why domesticated animals do not produce mutually sterile varieties, and why plants, except in a few cases to be enumerated, do not produce sterile varieties. It seems to me that the real difficulty in the matter at hand is not why the domestic breeds, when crossed, have not become mutually sterile, but why natural varieties, by undergoing perpetual changes, acquire the grades of species, Thus in general sterility occurs.We are still far from knowing its cause precisely; and it is not surprising when we see how profoundly ignorant we are of the normal and abnormal operations of the reproductive system; Innumerable competitors, engaged in a struggle for existence, are exposed for a long time to more uniform conditions of life than the domestic varieties; and thus inevitably produce very different results.For we know that it is not uncommon for wild animals and plants to become sterile if they are taken from nature and domesticated or cultivated; , is probably equally markedly sensitive to the effects of unnatural hybridization.On the other hand, domesticated organisms, from the mere fact that they have been domesticated, are not inherently highly sensitive to changes in their conditions of life, and are today generally able to resist repeated changes in conditions of life without diminishing their fertility. , so that the breeds produced by domestic organisms, if crossed with other varieties from the same origin, would be expected to be seldom injuriously affected in their reproductive functions by the act of crossing. I have said that the crosses of the varieties of the same species seem necessarily to be all fertile.A certain degree of sterility is, however, evidence of a certain degree of sterility in the few instances which I shall briefly describe below.This evidence is at least as valuable as our belief in the sterility of innumerable species.This evidence, too, is obtained from those who maintain the objection, that in all cases fertility and sterility are taken as safe criteria for distinguishing species.Gartner bred for several years a dwarf yellow-seeded variety of maize in his garden, and a tall red-seeded variety nearby; Dioecious, but never a natural hybrid.So he inseminated thirteen ears of the other with pollen from one type of corn, but only one ear produced some seeds.Only five seeds were produced, and since the plants are dioecious, the operation of artificial insemination would have no deleterious effect here, and I am sure no one will doubt that these varieties of maize belong to different species; important It is to be noted that the hybrid plants thus produced are themselves perfectly fertile; so that even Gaitner dared not admit that the two varieties were distinct species. Girou de Buzareingues, who crossed three gourd varieties, which, like maize, are dioecious, asserted that the more they differed from each other, the less easily was mutual fertilization.How reliable are these tests, I do not know; but Sagarit ranks the types tested as varieties, and his classification is chiefly based on tests of sterility, and Nordan makes the same. in conclusion. The following case is still more remarkable, which at first glance seems incredible, but which has been carried out over many years by Gartner, such a good observer and opponent of the nine species of the genus Mullein. The result of the experiments, namely, that the crosses of the yellow and white varieties produce fewer seeds than the crosses of the same colored varieties of the same species.He further asserted that, when the yellow and white varieties of one species were crossed with those of another species, the crosses between the same-colored varieties produced more seeds than those between different-colored varieties.Mr. Scott has also experimented with the species and varieties of Mullein: though he could not confirm Gartner's results concerning the interbreeding of different species, he found that heterochromatic varieties of the same species produced more species than homochromatic varieties. Seeds are less, the ratio is 86:100.These varieties, however, differ in no way except in the color of the flowers, and one variety is sometimes developed from the seeds of another. The accuracy of Korreuter's work has been confirmed by every subsequent observer, who demonstrated the remarkable fact that a particular variety of common tobacco, if crossed with a species of the same size, More fertile than other variants.He experimented with the five forms commonly called varieties, and with the most severe experiment, the cross-cross experiment, and found that their hybrid offspring were all perfectly fertile.But one of these five varieties, whether used as the male parent or the female parent, will never produce a hybrid like that produced by crossing the other four varieties with Nicotiana glutinosa. The land is sterile.The reproductive system of this variety must therefore have been modified in some way and to some extent. From these facts it can no longer be maintained that varieties must be perfectly fertile when crossed.From the difficulty of ascertaining the sterility of varieties in a state of nature, since a putative variety, if it were proved to be sterile to some degree, would almost universally be ranked as a species:- In character, and according to domestic breeds, varieties have not been kept for long periods under uniform conditions of life;—from these few observations we may conclude that the fertility or non-fertility at the time of crossing does not make a fundamental difference between varieties and species.The general sterility of hybrid species is not to be regarded as a special acquisition or endowment, but may safely be regarded as accompanying a change of an unknown nature in their male and female germ-plasm. Hybrids vs. Mixed Breeds Beyond Fertility The offspring of crossed species and crossed varieties may be compared in several respects besides fertility.Gartner, who had ardently wished to draw a definite line between species and varieties, could find only few, and in my opinion quite insignificant difference.On the other hand, they agree extremely closely in many important points. Here I will discuss this very briefly.The most important difference is that hybrids are more variable in the first generation than hybrids, but Gartner maintains that hybrids from long-cultivated species are often variable in the first generation; Notable examples of this fact have been seen.Gartner further maintains that hybrids between very closely allied species are more likely to vary than hybrids between very dissimilar species;It is well known that when mixed-breeds and more fertile hybrids are bred for several generations, the variability in the offspring of both is great; but a few instances can still be given in which hybrids or mixed-breeds have long remained indistinguishable. traits.The variability in successive generations, however, is probably greater in mixed-breeds than in hybrids. The greater variability of mixed-breeds than hybrids does not seem at all surprising, since the parents of mixed-breeds are varieties, and mostly domestic varieties (very few experiments have been done with natural varieties), which means that there The variability is recent, and means that the variability produced by the act of crossing has often continued and increased.It is a curious fact that the variability of hybrids in the first generation is insignificant, as compared with that in successive generations thereafter, and it is worth noting.For this is connected with one of the causes of ordinary variability which I have proposed; the idea that, since the reproductive system is markedly sensitive to changed conditions of life, it cannot use its powers under such conditions. Inherent ability to produce offspring closely resembling the parental type in all respects.The first generation of hybrids descended from species whose reproductive system had never been disturbed in any way (except in long-cultivated species), so they were not susceptible to variation; highly variable. Returning again to the comparison of hybrids and hybrids: Gartner says that hybrids reproduce the characters of either parent-form more readily than hybrids; but, if this is true, it is certainly only a difference of degree.Again, Gartner expressly states that hybrids produced from long-cultivated plants are more prone to reversion than hybrids produced from species in a state of nature; Explanation given: Vicula, who had experimented with wild species of poplar, doubted whether hybrids would reproduce the traits of the parental type; Noudan, on the contrary, insisted in emphatic terms that atavism in hybrids was almost a In accordance with the general tendency, his experiments were mainly carried out on cultivated plants.G. Turner goes on to say that when any two species, though closely allied to each other, are crossed with a third species, the hybrids thereof will be very different from each other, whereas two very different varieties of one species, if crossed with another species, , and their hybrids are not very different from each other.But as far as I know, this conclusion is based on one experiment; and it seems to be contrary to the results of several experiments made by Korreut. These are such insignificant differences between hybrids and mixed plants as Gartner can point out.On the other hand, hybrids and hybrids, especially those derived from allied species, obey the same laws, according to Gartner.When two species cross, one of the species sometimes has the dominant power to force the hybrid to resemble itself.The same is true, I believe, of varieties of plants; and certainly of animals also, that one variety often has a superior power of transmission over another.The hybrid plants that arise from crosses are generally closely resembling each other; so do the hybrid plants that arise from crosses.Repeated crosses with either parent in successive generations, whether hybrid or mixed, will cause them to reproduce the characters of either pure parent-type. These observations obviously apply to animals as well; but in animals the problem is considerably more complicated in part by the existence of secondary sexual characters: especially by the fact that in crosses between species and varieties one sex is more important than the other. The problem is further complicated by the fact that sex strongly has the transmitting force of superiority.For example, I think those authors who assert that the ass has superior power of transmission over the horse are right, so that both the mule and the hinny are more ass-like than horse-like; The force of transmission is so strongly preponderant that the offspring of a jackass and a mare, the mule, resembles a donkey more than the offspring of a jackass and a stallion, the mule. 某些作者特别着重下述的假定事实：即只有混种后代不具有中间性状，而密切相似于双亲的一方；但是这种情形在杂种里也曾经发生，不过我承认这比在混种里发生的少得多。看一看我所搜集的事实，由杂交育成的动物，凡与双亲一方密切相似的，其相似之点似乎主要局限于性质上近于畸形的和突然出现的那些性状——如皮肤白变症，黑变症（melanism）、无尾或无角、多指和多趾；而与通过选择慢慢获得的那些性状无关。突然重现双亲任何一方的完全性状的倾向，也是在混种里远比在杂种里更易发生。混种是由变种传下来的，而变种常常是突然产生的，并且在性状上是半畸形的；杂种是由物种传下来的，而物种则是慢慢而自然地产生的。我完全同意普罗斯珀·卢卡斯博士的见解，他搜集了有关动物的大量事实后，得出如下的结论：不论双亲彼此的差异有多少，就是说，在同一变种的个体结合中，在不同变种的个体结合中，或在不同物种的个体结合中，子代类似亲代的法则都是一样的。 除了能育性和不育性的问题以外，物种杂交的后代和变种杂交的后代，在一切方面似乎都有普遍的和密切的相似性。如果我们把物种看作是特别创造出来的，并且把变种看作是根据次级法则（Secondary laws）产生出来的，这种相似性便会成为一个令人吃惊的事实。但这是和物种与变种之间并没有本质区别的观点完全符合。 本章提要 充分不同到足以列为物种的类型之间的第一次杂交以及它们的杂种，很一般地但非普遍地不育，不育性具有各种不同的程度，而且往往相差如此微小，以致最谨慎的试验者根据这一标准也会在类型的排列上得出完全相反的结论。不育性在同一物种的个体里是内在地易于变异的，并且对于适宜的和不适宜的生活条件是显著敏感的。不育性的程度并不严格遵循分类系统的亲缘关系，但被若干奇妙的和复杂的法则所支配。在同样的二个物种的互交里不育性一般是不同的，有时是大为不同的。在第一次杂交以及由此产生出来的杂种里，不育性的程度并非是永远相等的。 在树的嫁接中，某一物种或变种嫁接在其他树上的能力，是伴随着营养系统的差异而发生的，而这些差异的性质一般是未知的；与此同样，在杂交中，一个物种和另一物种在结合上的难易，是伴随着生殖系统里的未知差异而发生的。想像为了防止物种在自然状况下的杂交和混淆，物种便被特别赋予了各种程度的不育性，和想像为了防止树木在森林中的接合，树木便被特别赋予了各种不同而多少近似程度的难以嫁接的性质，同样是毫无任何理由的。 第一次杂交和它的杂种后代的不育性不是通过自然选择而获得的。在第一次杂交的场合，不育性似乎决定于几种条件：在某些事例里，主要决定于胚胎的早期死亡。在杂种的场合，不育性显然决定于它们的整个体制被二个不同类型的混合所扰乱了；这种不育性和暴露在新的和不自然的生活条件下的纯粹物种所屡屡发生的不育性，是密切近似的。能够解释上述情形的人们，就能够解释杂种的不育性。这一观点有力地被另一种平行现象所支持，即是：第一，生活条件的微小变化可以增加一切生物的生活力和能育性；第二，暴露在微有不同的生活条件下的、或已经变异了的类型之间的杂交，将有利于后代的大小、生活力和能育性。关于二型性和三型性植物的不合法的结合的不育性以及它们的不合法后代的不育性所举出的一些事实，大概可以确定以下情形，即有某种未知的纽带在所有情形里连结着第一次杂交的不育性程度和它们的后代的不育性程度。对于二型性这些事实的考察，以及对于互交结果的考察，明白地引出了如下的结论：杂交物种不育的主要原因仅仅在于雌雄生殖质中的差异。但是在不同物种的场合里，为什么在雌雄生殖质极其一般地发生了或多或少的变异后，就会引致它们的相互不育性，我们还不明白；然而这一点和物种长期暴露在近于一致的生活条件下，似有某种密切的关联。 任何二个物种的难以杂交和它们的杂种后代的不育性，纵然起因不同，在大多数情形下应当是相应的，这并不奇怪；因为二者都决定于杂交的物种之间的差异量。第一次杂交的容易和如此产生的杂种的能育，以及嫁接的能力——虽然嫁接的能力是决定于广泛不同的条件的——在一定范围内部应当与被试验类型的分类系统的亲缘关系相平行，这也不奇怪；因为分类系统的亲缘关系包括着一切种类的相似性。 被认为是变种的类型之间的第一次杂交，或者充分相似到足以被认为是变种的类型之间的第一次杂交，以及它们的混种后代，一般都是能育的，但不一定如常常说到的那样，必然如此。如果我们记得，我们是何等易于用循环法来辩论自然状态下的变种，如果我们记得，大多数变种是在家养状况下仅仅根据对外在差异的选择而产生出来的，并且它们并不曾长久暴露在一致的生活条件下；则变种之有几乎普遍而完全的能育性，就不值得奇怪了。我们还应当特别记住，长久继续的家养具有消弱不育的倾向，所以这好像很少能诱发不育性。除了能育性的问题之外，在其他一切方面杂种和混杂种之间还有最密切而一般的相似性——就是说在它们的变异性方面，在反复杂交中彼此结合的能力方面，以及在遗传双亲类型的性状方面，都是如此。最后，虽然我们还不知道第一次杂交的和杂种的不育性的真实原因，并且也不知道为什么动物和植物离开它们的自然条件后会变成为不育的，但是本章所举出的一些事实，对我来说，似乎与物种原系变种这一信念并不矛盾。