Popular Science Monthly/Volume 48/February 1896/The Study of Inheritance I


A Review of the Writings of Francis Galton.



IT is much more easy to talk about inheritance than to study it. Of the books and essays which meet us at every turn, few have much basis in research, but those of Francis Galton are among the most notable exceptions. These books, which have appeared at intervals during the last twenty-five years, are not speculations but studies. They describe long exhaustive investigations, carried out by rigorous methods, along lines laid down on a plan which has been matured with great care and forethought.

The simplicity of their language is as notable as their subject. Dealing with conceptions which are both new and abstruse, the author finds our mother tongue rich enough for his purpose, and, while the reasoning often taxes all our powers, there is never any doubt as to the meaning of the words.

When in rare cases a technical term is inevitable, some familiar word is chosen with so much, aptness that it does its duty and presents the new conception better than a compound from two or three dead languages. The terms "mid-parent" or "mid," "fraternity," "nurture," and "Q" can not mislead or convey any idea except the right one.

The reviewer's debt to Galton is very great, and it is acknowledged with gratitude. Such acquaintance with the statistical method as he possesses he owes to the study of these books; especially the ones on Hereditary Genius (1869), on Natural. Inheritance (1889), and on Finger Prints (1893).

The attempt to question Galton's generalizations may therefore seem ungracious and presumptuous, but the uncertainties of vital statistics are proverbial; and it is not impossible that Galton's data may fail to cover all the ground needed to prove his general conclusions.

One of these generalizations is so far-reaching that, if it is well founded, it must lead to profound and fundamental changes in our view of the origin of species.

According to Darwin and Wallace, specific identity in living things is the outcome of the extermination, in the struggle for existence, of the individuals which depart too widely from that "type" which is on the whole best adapted to existing conditions. As these conditions change, the "type" is also slowly modified through a change in the character of this process of extermination. According to this view, the "type" is the outcome of the statistical "law of error," or the deviation from the mean, which holds good in the environment; and while the "events" are properties of the organism, the type is fixed by the external world, and not by anything in the organism itself.

Galton holds that specific identity is not due to the process of extermination, but to "organic stability." As I understand him, he holds that this fills up the gaps made by extermination, and thus keeps the type intact.

This "principle of stability," which is held to result in the permanency of types, is said to be quite independent of selection. "Genera and species may be formed without the slightest aid from either natural or sexual selection." "Organic stability is the primary factor by which the distinctions between genera are maintained."

Galton holds, furthermore, not only that specific stability is independent of selection, but that selection is "scarcely competent" to effect a change of type "by favoring mere varieties"—that is, by the extermination of the ordinary slight differences between individuals; and that it is only when a "sport" has made its appearance, only when the type has actually changed, that selection can exert any influence. According to this view, the agencies which cause "sports" are the real causes of the mutability of species, and natural selection can do nothing more than to exterminate disadvantageous sports, and thus favor advantageous ones.

The "organic stability" to which so much is attributed is held to be due to the fact that the child inherits in part from its parents, in part from more remote ancestors; and since the sum of its ancestry, or its "mid-parentage," is on the average nearer than any exceptional parents to the mean of the race, the children of selected parents are on the average more mediocre than their parents.

It is quite possible that Galton's data may be valuable and trustworthy, and that they may yet fail to prove this generalization; and I shall try to show that this is the case, although I am not sure that I fully grasp his point of view.

I assume that he regards a zoölogical type or species as something; something which is due to a "principle of stability," which is not the result of selection. This is assuredly the current interpretation of his statements, and it is from this standpoint that I shall examine his writings.

If this is not his opinion; if he really believes that this "principle" owes its existence to past selection; if he only deduces from his data the generalization that the results of past selection may persist after it has ceased to act; I see no ground for criticism, for his data assuredly prove this much, although I can not reconcile his statement that "the principle of stability is independent of selection" with the belief that it is the result of past selection.

Before entering upon the discussion of the subject it may be well to ask what evidence there is that the child does inherit from any ancestor except its parents, for descent from a long line of ancestors is not necessarily equivalent to inheritance from them, and it is quite possible that the conception of a "mid-parent" may be nothing but a logical abstraction.

Most of its support is derived from the phenomena of reversion or atavism; from the appearance in children of ancestral features which were not exhibited by the parents. While these phenomena are familiar and real, we may well doubt whether any of them are reversions in Galton's sense.

In some cases we can prove that a so-called reversion is simply the manifestation of a feature which is latent in the structure of all the normal individuals of the species. The occurrence of a distinct premaxillary bone in man is an example of this sort of reversion. It is the outcome of the arrest of normal development, and this arrest might have happened to any member of the species. We do not know what causes the arrest, but the view that it is due to some adverse circumstance which has kept the individual from completing its development is much more simple and probable than the view that the child inherits its distinct premaxilla from any ancestor except its parents.

When the son of a beardless boy grows up and acquires a beard, we may be permitted to say that he has inherited his grandfather's beard; but this is only a figure of speech, and he actually inherits the beard which his father might have acquired had he lived; nor would the case of a child descended from a series of ten or a hundred beardless boys and beardless women be any different. If we were to propagate a plant by cuttings, for ten or a hundred generations, under conditions which did not permit it to flower, and were finally to put the last of the series where it did flower, we should not be justified in saying that it did not inherit its flower from the preceding cutting; nor would the case be any different if, for some reason, this preceding cutting could not be made to bloom.

The phenomena of polymorphism in insects and in hydroids present illustrations of the normal inheritance of latent characters, but we find in them no ground for the assertion that the ancestral characteristics of the medusa are not inherited from the hydroid which produces it.

The sum of the visible features of the parent, plus the sum of its latent potencies, may be called a "mid-parent" for statistical purposes, if we see fit, but there is no evidence that this midparent is anything else than.the actual parent.

With this introductory note, we may now enter upon the study of Galton's works, the central point of which is as follows:

If we select any one characteristic of a natural group of animals—such a characteristic as the weight of the individuals, or the ratio between the lengths of their arms and legs, or anything else which admits of exact numerical statement—it will be found that while no two members of the group are exactly alike, they nevertheless conform to a type, and show the existence of a standard, the mean or average, to which the majority adhere pretty closely, while other members of the group are more abnormal, and show marked deviation from the mean.

If the cases tabulated are numerous enough, the individuals will conform, so far as this quality is concerned, to what is known in statistical science as the law of frequency of error. This agreement will be so close, when great numbers of instances are examined, that the number of individuals which depart from the mean to any specified degree may be computed mathematically.

For example, the chest measurements of 5,738 soldiers gave the following results:

Inches. Measured. Computed.
33 5 7
34 31 29
35 141 110
36 322 323
37 732 732
38 1,305 1,333
39 1,867 1,838
40 1,882 1,987
41 1,628 1,675
42 1,148 1,096
43 645 560
44 160 221
45 87 69
46 88 16
47 7 3
48 2 1

If the number of events had been five hundred thousand or five million instead of five thousand, the agreement between the computed and observed frequency of each degree of departure from the mean would have been very much closer. When the number of cases is unlimited, the agreement is perfect.

Galton gives the following illustration of the significance of a type:

Suppose a large island inhabited by a single race, who intermarry freely, and who have lived for many generations under constant conditions, then the average height of the adult male of that population will undoubtedly be the same year after year. Also—still arguing from the experience of modern statistics, which are found to give constant results in far less carefully guarded examples—we should undoubtedly find year after year the same proportion maintained between the number of men of different heights. I mean if the average stature was found to be sixty-six inches, and if it was also found in any one year that one hundred per million exceeded seventy-eight inches, the same proportion of one hundred per million would be closely maintained in all other years.

An equal constancy of proportion would be maintained between any other limits of height we please to specify, as between seventy-one and seventy-two inches, between seventy-two and seventy-three, and so on. Now, at this point the law of deviation from an average steps in. It shows that the number per million, whose heights range between seventy-one and seventy-two inches, or between any other limits we please to name, could be predicted from the previous datum of the average, and of any other one fact, such as that of one hundred per million exceeding seventy-eight inches.

Suppose a million of the men to stand in turns with their backs against a vertical board of sufficient height, and their heights to be dotted off upon it. The line of average height is that which divides the dots into two equal parts, and stands, in the case we have assumed, at the height of sixty-six inches. The dots will be found to be ranged so symmetrically on either side of the line of average that the lower half of the board will be almost a precise reflection of the upper. Next, let a hundred dots be counted from above downward, and let a line be drawn below them. According to the conditions, this line will stand at the height of seventy-eight inches. Using the data afforded by these two lines, it is possible by the help of the law of deviation from an average to reproduce with extraordinary closeness the entire system of dots on the board.

This law of deviation from an average is not restricted to vital phenomena, but holds true of all events which are the resultants of variable conditions, which remain the same through all the events recorded. If the marks on the board had been made by bullets fired at a horizontal line stretched in front of a target, they would have been distributed according to the same law, their average value would be constant, and the deviations of the several events from the average would be governed by the same law, which is identical with that which governs runs of luck at a gaming table.

Galton has described an apparatus which mimics in a very pretty way the conditions on which deviations from a mean depend. It is a long, shallow box set on end and glazed in front, leaving a depth of about a quarter of an inch behind the glass. Strips are placed in the upper part to act as a funnel. Below the outlet of the funnel stand a succession of rows of pins stuck squarely into the backboard, and below these again are a series of vertical compartments. A charge of small shot is inclosed. When the frame is held topsy-turvy, all the shot runs to the upper end; then when it is turned back into its working position the desired action commences.

The shot passes through the funnel and, issuing from its narrow end, scampers deviously down through the pins in a curious and interesting way: each one of them darting a step to the right or left, as the case may be, every time it strikes a pin. The pins are so placed that every descending shot strikes a pin in each successive row. The cascade issuing from the funnel broadens as it descends, and at length every shot finds itself caught in a compartment immediately after freeing itself from the last row of pins. The outline of the columns of shot that accumulate in the successive compartments approximates to the mathematical law of frequency, and is closely of the same shape, however often the experiment is repeated.

The outlines of the columns would become more nearly identical with the normal law of frequency if the rows of pins were much more numerous, the shot smaller, and the compartments narrower; also, if a larger quantity of shot were used.

The principle on which the action of the apparatus depends is that a number of small and independent accidents befall each shot in its career. In rare cases a long run of luck continues to favor the course of a particular shot toward either outside place, but in the large majority of instances the number of accidents that cause deviation to the right balance in a greater or less degree those that cause deviation to the left. Therefore most of the shot finds its way into the compartments that are situated near to a perpendicular line drawn from the outlet of the funnel, and the frequency with which shots stray to different distances to the right and left of that line diminishes in a much faster ratio than these distances increase.

Types which are based upon vital statistics have peculiar interest, since they persist from generation to generation, according to what is known as the law of specific stability, while they also undergo slow changes according to the principle of the mutability of species.

Individuals come and go, but the type persists, and its slow changes may be pictured as quite independent of and more substantial than the procession of individuals which files past only to vanish from the world.

The statistical comparison of vital types affords a means for studying the phenomena of inheritance by the exact methods of mathematics, and it is capable of yielding definite and valuable results, so far as the vital phenomena which are studied can be treated as if they stood alone, but the attempt to generalize from vital statistics and to deduce general laws of inheritance from them is attended by peculiar difficulties, due in great part to the fact that the data which are studied are not separable from the organism which exhibits them. Stature or size or weight may be treated abstractly for statistical purposes, but the stature of an organism is not an abstraction, for the organism is not only a bundle of properties, but a unit as well, and its stature is only one among many features which are all beautifully co-ordinated with each other in such a way as to promote the welfare of the species. A generalization which ignores this fact and treats stature as an abstraction may, while proved by statistics, be untrustworthy as a contribution to our knowledge of inheritance.

In popular language, specific stability may be said to be due to inheritance, and specific mutability to variation; but in this connection these words have only a loose meaning. In so far as they convey the impression that the stability of species and the mutability of species are antagonistic to each other, or are due to two distinct and opposing influences, these terms are unfortunate, for we have good ground for holding that they are due to the same influence—the extermination of certain individual peculiarities, and the preservation of others by natural selection.

The older naturalists held that adherence to type is due to some innate principle of specific stability which is an essential and immutable attribute of each species of living things; but the accumulation of conclusive evidence of the mutability of species has driven this conception out of the field. Most naturalists now regard the type as nothing but that normal which is most perfectly fitted to the environment, and they hold that it is kept true through the extinction of aberrant individuals by selection.

According to this view, which seems to be supported by ample evidence, the stability of species is due to survival—to the same mechanism which brings about the mutability of species.

Galton is led by his statistical studies of vital characters to a view which bears an odd resemblance to that of the older naturalists; for, according to him, the principle of stability which results in the permanency of types is quite independent of selection.

He shows, for example, by the statistical study of stature, that the type of human stature is very constant from generation to generation, although the statistics of marriage show that there is no controlling tendency for persons of like stature to marry. He also shows that the children of parents who are both tall or both short do not on the average have the stature of their parents, but are nearer than they to the mean for the race. These facts, and others like them, are held to prove the existence of a principle of stability independent of selection.

In his more recent work on the patterns at the tips of human fingers, he says that since it has been shown (chapter xii) that the character of the finger prints is practically identical in Englishmen, Welshmen, Jews, negroes, and Basques, the same familiar patterns appearing in all of them with much the same degree of frequency, and that persons belonging to different classes, such as students in science and students in art, farm laborers, men of culture, and the lowest idiots in the London district, show no decided difference in their finger prints, it seems to be proved that no sensible amount of correlation exists between any of the patterns on the one hand and any of the bodily faculties or characteristics on the other. It seems absurd, therefore, to hold that, in the struggle for existence, a person with, say, a loop on his right middle finger has a better chance of survival or a better chance of early marriage than one with an arch. Consequently, genera and species are here seen to be formed without the slightest aid from either natural or sexual selection, and these finger patterns are apparently the only peculiarity in which panmyxia, or the effect of promiscuous marriage, admits of being studied on a large scale.

He says that results of panmyxia in finger-markings corroborate his arguments in Natural Inheritance and elsewhere to show that "organic stability" is the primary factor by which the distinctions between genera are maintained. Consequently, the progress of evolution is not a smooth and uniform progression, but one that proceeds by jerks, through successive "sports" (as they are called), some of them implying considerable organic changes, and each in its turn being favored by natural selection.

Galton's explanation of this specific stability is as follows: The child inherits in part from the parents, in part from more remote ancestors; and since the sum of its ancestry, or, as Galton terms it, the mid-parentage, is on the average nearer than the exceptional parents to the mean of the race, the children of selected parents are on the average more mediocre than their parents.

I have tried to show that, while the child is descended from a long line of ancestors, it inherits from none but the two parents, and that it can only be said in a figurative sense to inherit from more remote ancestors. I shall soon refer to proofs that the persistency of inherited types is due to natural selection, and not to any principle of organic stability independent of selection.

If this is true, if the stability of specific types is due to the survival of the fittest, why do we have a type and not a fixed standard? If speed and strength and courage are good things, why is not every individual as swift as the swiftest, as brave as the bravest, and as strong as the strongest? Why does not every individual have every useful quality developed to the highest excellence to which it may attain in any individual of the species? Why should we find that diversity among individuals which usually passes under the name of "variation"?

We can measure strength and can treat it abstractly, and we can artificially select and breed from the strongest members of a stock, neglecting all other features; but this is not what takes place in Nature.

Here the most favored individuals are not the strongest, but the ones in which all the qualities of the species are most perfectly co-ordinated with each other in relation to the external world. Excessive strength may involve deficiency in some other essential, and the mean or average strength of the species is that degree of strength which is most in harmony with the mean degree of development of all the other characteristics of the species; and the individuals who depart too widely from this mean, either through excess of strength or deficient strength, are the ones which are exterminated.

Galton has himself given such a clear statement of the way a type is established by selection that it can not be improved upon, and I quote it in his own words:

"Suppose," he says, "that we are considering the stature of some animal that is liable to be hunted by certain beasts of prey in a particular country. So far as he is big of his kind, he would be better able than the mediocres to crush through the thick grass and foliage whenever he was scampering for his life, to jump over obstacles, and possibly to run somewhat faster than they. So far as he is small of his kind, he would be better able to run through narrow openings, to make quick turns, and to hide himself. Under the general circumstances it would be found that animals of some particular stature had on the whole a better chance of escape than any other; and if their race is closely adapted to these circumstances in respect to stature, the most favored stature would be identical with the mean of the race. Though the impediments to flight are less unfavorable to this (stature) than to any other, they will differ in different experiences. The course of an animal might chance to pass through denser foliage than usual, or the obstacles in his way may be higher. In that case an animal whose stature exceeded the mean would have an advantage over mediocrities. Conversely the circumstances might be more favorable to a small animal. Each particular line of escape might be most favorable to some particular stature, and, whatever this might be, it might in some cases be more favored than any other. But the accidents of foliage and soil in a country are characteristic and persistent, and may fairly be considered as approximating to a typical kind. Therefore those which most favor the animals of the mean stature will be more frequently met with than those which favor any other stature, and the frequency of the latter occurrence will diminish rapidly as the stature departs from the mean.

"It might well be that natural selection would favor the indefinite increase of numerous separate faculties if their improvement could be effected without detriment to the rest: then mediocrity in that faculty would not be the safest condition. Thus an increase of fleetness would be a clear gain to an animal liable to be hunted by beasts of prey, if no other useful faculty v/as thereby diminished.

"But a too free use of this 'if' would show a jaunty disregard of a real difficulty. Organisms are so knit together that change in one direction involves change in many others; these may not attract attention, but they are none the less existent. Organisms are like ships of war, constructed for a particular purpose in warfare as cruisers, line-of-battle-ships, etc., on the principle of obtaining the utmost efficiency for their special purpose. The result is a compromise between a variety of conflicting desiderata, such as cost, speed, accommodation, stability, weight of guns, thickness of armor, quick steering power, and so on. It is hardly possible in a ship of any established type to make an improvement in any one of these respects without a sacrifice in other directions. If the fleetness is increased, the engines must be larger, and more space must be given up to coal, and this diminishes the remaining accommodation.

"Evolution may produce an altogether new type of vessel that shall be more efficient than the old one, but when a particular type has become adapted to its functions, through long experience, it is not possible to produce a mere variety of its type that shall have increased efficiency in some one particular without detriment to the rest. So it is with animals."

This quotation from Galton shows how a type may be established by selection, and it also shows why it is not possible to make any great and permanent change in the type of one characteristic of an organism unless changes at the same time occur in the type of other characters of the same organism. It also follows that a breeder of domesticated animals or cultivated plants who devotes his attention to one characteristic exclusively must soon reach a point where no further improvement in this quality is practicable unless the spceies is at the same time modified in other respects. This fact does not prove that specific stability is due to anything else than selection. It only proves that no great change is possible without the co-ordinated modification of correlated features, and this is just what we should expect as the effect of long ages of selection.

The passage I have quoted from Galton seems to indicate that, after all, he may believe that the specific types of zoölogy and botany are nothing more than the persistent effects of past selection, and that his statement that "organic stability is independent of selection" may refer to present selection only.

These statements are clear and explicit, however, and they have been interpreted by most readers as a flat contradiction of the view that the mechanism which leads to the formation of new types is identical, on its vital side, with that which preserves established types; the view that the differences between the two are differences in the external world.

He says (Nature, September 4, 1885): "It is some years since I made an extensive series of experiments in the produce of seeds of different sizes, but of the same species. . . . It appears from these experiments that the offspring did not tend to resemble their parent seeds in size, but to be always more mediocre than they; to be smaller than they if the parents were large; to be larger than the parents if the parents were very small," and that the analysis of the family records of heights of 205 human parents and 930 children fully confirms and goes far beyond the conclusions obtained from seeds, as it gives with great precision and unexpected coherence the numerical value of the regression toward mediocrity. He says that this regression is a necessary result of the fact that "the child inherits partly from his parents, partly from his ancestors. Speaking generally, the further his genealogy goes back, the more numerous and varied will his ancestors become, until they cease to differ from any equally numerous sample taken at haphazard from the race at large. Their mean stature will then be the same as that of the race; in other words, it will be mediocre." He illustrates this by comparing the result of the combination in the child of the mean stature of the race with the peculiarities of its parents to the result of pouring a uniform proportion of pure water into a vessel of wine. It dilutes the wine to a certain fraction of its original strength, whatever that strength may have been.

He then goes on to the deduction that the law of regression to the type of the race "tells heavily against the full hereditary transmission of any rare and valuable gift, as only a few of the many children would resemble the parents. The more exceptional the gift the more exceptional will be the good fortune of a parent who has a son who equals, and still more if he has a son who surpasses, him. The law is even-handed; it levies the same heavy succession tax on the transmission of badness as well as goodness. If it discourages the extravagant expectations of gifted parents that their children will inherit all their powers, it no less discountenances extravagant fears that they will inherit all their weaknesses and diseases. . . . Let it not for a moment be supposed that the figures invalidate the general doctrine that the children of a gifted pair are much more likely to be gifted than the children of a mediocre pair; what it asserts is that the ablest of the children of one gifted pair is not likely to be as gifted as the ablest of all the children of many mediocre pairs."

In his recent work on Finger Prints he says: "It is impossible not to recognize the fact so clearly illustrated by these patterns in the thumbs that natural selection has no monopoly of influence in the construction of genera, but that it could be wholly dispensed with, the internal conditions acting by themselves being sufficient. Not only is it impossible to substantiate a claim for natural selection that it is the sole agent in forming genera, but it seems, from the experience of artificial selection, that it is scarcely competent to do so by favoring mere varieties in the sense in which I understand the term. Mere varieties from a common typical center blend freely in the offspring, and the offspring of every race where statistical characters are constant necessarily tend, as I have shown, to regress toward their common typical center. A mere variety can never establish a sticking point in the forward course of evolution." He therefore holds that, while specific stability is due to inheritance from a long series of ancestors, the transmutation of species is due to the sudden appearance of "sports" which, if useful, are seized upon and perpetuated by selection.

He says that a sport is a substantial change of type effected by a number of small changes of typical center, each more or less stable, and each being in its turn favored and established by natural selection to the exclusion of its competitors.

"The distinction between a mere variety and a sport is real and fundamental."

This generalization, based upon definite numerical data, is so fundamental and far-reaching that a critical discussion of the evidence is most important.