except only those which spring out of that other plague of the Darwinian, the too numerous cases of imperfect adjustment to environment.

For about a generation following the publication of the Origin, writers on evolution were inclined to content themselves with constructing ingenious theories on the basis of Darwin's evidence, piecing out one untested hypothesis with another, and, in general, following a dialectical method which fairly merited Mr. Bateson's sarcastic paraphrase: "If,' say we with much circumlocution, the course of Nature followed the lines we have suggested, then, in short, it did.'' As he put the case ten years ago: —

6

“So far, indeed, are the interpreters of Evolution from adding to this [Darwin's] store of facts, that in their hands the original stock becomes even less, until only the most striking remain. It is wearisome to watch the persistence with which these are revived for the purpose of each new theorist. How well we know the offspring of Lord Morton's mare, the bitch Sappho, the Sebright Bantams, the Himalaya Rabbit with pink eyes, the white Cats with their blue eyes, and the rest! Perhaps the time has come when even these splendid observations cannot be made to show much more. Surely their use is now rather to point the direction in which we must go for new facts."

The last decade has changed all this. A few of the younger men who have come up since the days of ignorance have turned their backs upon the older questions, and have gone to work on the two great presuppositions of Darwinism, heredity and variation, making them always a question of fact, and not of logic, in a way that would have delighted Darwin's heart.

As a result of this work, unless all signs fail, the next few years should see an advance in the theory of evolution comparable with that which is just now making the physicist the thaumaturgist of science.

Variation is, therefore, except for Darwin's work, almost a new subject; so new, that important facts concerning the commonest animals and plants are still ungathered. Indeed, so inconsiderable is the amount which has yet been written from the modern standpoint, and that little is so easily come at, that almost any one who enjoys playing with mathematics, or any amateur gardener with a turn for experimenting, can, with a few months' reading, put himself in the way of making worthy contributions to science.

Luckily, too, for all writers on heredity and variation, and perhaps still more fortunately for the interest of their readers, the two turn out to be, not, as used to be said, two antagonistic principles, but merely different aspects of the same problem. Nature seems always to be striving to give to each creature seed after its kind. She never quite succeeds, and, in so far as she fails, we call her failure variation. She rarely fails seriously, and such measure of success as she attains we term heredity. A single illustration will serve to show how close the two stand to each other and to everyday life. It must be a matter of common observation that different parts of the body are so correlated that, for example, long arms nearly always accompany long legs, and usually a long face also. Modern standards of accuracy, however, demand something more definite than general impressions that certain things are apt to occur together. So the powerful mathematical analysis, which is perhaps the most distinctive feature of recent work in this field, has yielded, among other things, the index of correlation, a convenient numerical measure of the strength of the tie between the variations of any two organs of the body. On the other hand, the index of correlation for the same organs between parent and offspring is a measure of the force of heredity. Professor Karl Pearson finds that this correlation is least between mother and daughter, somewhat greater between

mother and son, greater still between father and daughter, and greatest of all between father and son. Thus it appears and this is corroborated by other evidence that men not only transmit more to their children of either sex than do women, but also inherit more even from their mothers; a striking justification of our immemorial emphasis on inheritance in the male line. Per contra, if women inherit less from their fathers and grandfathers, by so much more are they the daughters of the race. Professor Pearson's discovery recalls a piece of biological speculation out of fashion now and much frowned upon in certain ters

quar

to the effect that males furnish everywhere the variable and progressive element of a species, while the great stream of racial inheritance flows through the females; a theory which would explain the differences between men and women by supposing that in the one Nature tries her little fliers, in the other, she salts down her gains.

For the anonymous author of Doubts about Darwinism the old dilemma still offers nothing better than a choice of horns on which to spit himself, in spite of all the good work of the last dozen years, with which, to be sure, he shows no very striking acquaintance. Like the worthy clergymen who, a generation ago, used to refute the evolutionists on the basis of a sight acquaintance with the commoner domestic animals, the "SemiDarwinian can only fall back on special act of creative energy whenever he finds a "gap." It is always possible, of course, that the teleologist is right, though even his ready-made explanation has its own difficulties. Teleology, how ever, is not science; and there never would have been any science if men had been contented with giving the easy explanation, as there never was any until they stopped giving it.

a

For the two naturalists, on the other hand, the way out of the old difficulty lies through the newer studies of inher

itance and variation. But the two authors tend so far to opposite opinions on most theoretical questions that they are, in a general way, the spokesmen for the somewhat diverse schools into which students of the double problem are divided by the two sorts of variation. Dr. Vernon gives an account of all important discoveries in variation, heredity, adaptation, and related subjects since Darwin, with so much of Darwin's own work as is necessary for a background. But while he treats all aspects of the question in due proportion, his chief interest is with the stricter Darwinism which puts most stress on normal variation. Professor Morgan comes to his somewhat unorthodox opinions by way of the remarkable studies in the regeneration of lost parts, an account of which he brought out two years ago. So that his first concern is with the problem of adaptation, where incidentally he disposes most effectively of the teleology of the Semi-Darwinian. For him the study of discontinuous variations and their inheritance has been the most significant aspect of recent work. Both authors, therefore, cover a good deal the same ground; Professor Morgan with the more critical attitude and the greater interest in the general question, Dr. Vernon with more attention to new facts and methods for their own sake. He assumes that his public is already on reading terms with Darwin, while Professor Morgan begins at the beginning, and devotes half his space to matters which the other takes for granted. Dr. Vernon is, on the whole, the easier reading; largely because his collection of facts is many times greater; in some degree because, as a general rule, English men of science write better than Americans. Between the two, modern aspects of organic evolution get pretty well discussed.

But to return to our old dilemma. There is, from the side of continuous variation, a great deal which goes to show that, all theory aside, Natural Selection

does seize on small differences which seem to us of no great importance, and does use them to hold one species to its most efficient form, or to modify another to fit a new set of conditions. To take

but one example out of many, Dr. Bumpus found that of 136 storm-beaten English sparrows, the 72 that revived differed appreciably from the 64 that died. In general, the aberrant individuals perished, and those nearest the typical size and shape survived. But besides this, the survivors were shorter and lighter than the others, longer of leg and breast-bone, and larger of skull. Yet who would not have said a priori that half a gram more of average weight would not be rather an advantage than otherwise when it came to weathering a storm, or that a little inferiority in length of leg could possibly make the slightest difference one way or the other! Still more striking, perhaps, is the case of Mr. Weldon's crabs, in which Natural Selection is modifying a species under our very eyes. It appears from measurements of thousands of individuals, and after all imaginable precautions for avoiding error, that the small shore crab of Plymouth Sound, England, is growing narrower of body, the ratio of breadth to length falling off about two per cent in five years. This change is due to the selective destruction of the broader individuals under the rapidly changing conditions of their environment. As the water of the Sound becomes dirtier year by year with the growth of the cities near by, the narrower crabs are slightly better able to filter it through their gill-chambers, and have, therefore, by so much the advantage over the others in the struggle for existence. And since their days, on the average, are longer in the water than their competitors', they leave more descendants to inherit their advantage, with the result that the race, continually recruited from the offspring of the "fitter" individuals, is maintaining itself in a situation where many species once common have been

exterminated. The obvious conclusion is that here are the beginnings of two new species. Given time enough, there should be a new sparrow to fit American weather, and a new crab to fit the mud of Plymouth Sound. It is easy enough for the philosopher to say offhand that the selection of such little differences can never go beyond the production of local races; but how, after all, does he know?

On the other hand, from the side of discontinuous variation, we have learned that almost any plant or animal may suddenly exhibit new characters. A perfect tulip appears with all its parts arranged by fours, when, by all precedent, they should go by threes; and we men, who are sometimes thought, very erroneously, to be above all bodily change,

even we are somewhat given to having more ribs or fewer than is thought quite correct, and six or seven digits in place of the usual five. Equally striking facts of the same sort were, of course, known to the older naturalists. But they missed seeing how common they are; in part, no doubt, because the analysis of the idea of discontinuity had not, in their day, shown that variation may be indefinitely small and yet entirely discontinuous. Size in man, for example, is one of the most variable qualities known, - the dime-museum giant is well up to ten times the weight of the dwarf, — but the variation is continuous, in the sense that all intermediate sizes occur, and those nearest the mean are most numerous. Eye-color, on the other hand, though a very small and unimportant matter, is discontinuous. Nearly all eyes can be assigned at a glance either to the brownblack group or to the gray-blue-green group. Eyes, therefore, are either dark or light, almost never intermediate. Not only, therefore, do we now know that abrupt variation is very much more common than used ever to be suspected, but we have, besides, good reason to think that almost any species, after plodding quietly along for ages, may, all of a sud

den, take to varying in the most unexpected manner. Once a species gets to kicking over the traces, the new forms are likely to come with a rush, and the same mutation to appear independently over and over again. De Vries, studying mutations of the evening primrose, found among 50,000 plants in eight generations, 359 of one "incipient species," 229 of another, 158 of a third, and smaller numbers of four more, all distinct and self-consistent. Moreover, two of these new primroses grew wild, and maintained themselves under natural conditions unswamped by intercrossing with the stock from which they came.

By all the rules of logic, half-a-dozen plants or animals of a new variety breeding freely with a hundred times their numbers of the older sort ought shortly to disappear. The reason why they do not is that a discontinuous variant is likely to transmit its peculiarity completely, or else not at all. Though Darwin knew this in a general way, the first accurate statement of the matter, like many another fertile idea, came from Mr. Francis Galton. Galton pointed out long ago that there are at least three kinds of heredity, shown conveniently in the transmission of coat-color among horses. If a pure white horse is mated with a pure black one, the colt may follow one parent to the exclusion of the other, and be entirely black or entirely white, the missing color remaining latent, to appear, perhaps, in a subsequent generation. This is alternative or discontinuous inheritance. The latent quality is now termed "recessive; the other "dominant." Or the colt may fuse completely the parental qualities and be gray, blended or continuous inheritance. Or it may exhibit both colors unblended, as a black and white piebald, particulate or mosaic inheritance. The observer of mankind will easily recall a sufficiency of cases of the two extreme We expect children to be blends of the diverse qualities of their parents,

sorts.

and usually do find them a hodgepodge of ancestral characters, -the nose of one, the temper of another, on the average copying their forbears in due proportion; but as to separate qualities, the heirs of single individuals. Striking physical peculiarities and unusual mental gifts are thought to be very liable to entail, and to come down through half-adozen generations unblended and unimpaired. Good cases of mosaic inheritance are not so common. Eye-color is almost always a discontinuous heritage, but once in a while an iris is flecked with two colors, or marked with two concentric bands, and, more rarely, the two eyes of a pair are not mates.

All these facts had, of course, been known time out of mind. Galton, however, analyzed the matter and provided a terminology. He also taught the world not to mix the evidence for different sorts of inheritance, and he formulated his Law of Ancestral Heredity, the most important contribution to the theory of the subject up to the last year of the nineteenth century. With that year came the final discovery which was to gather up and interpret a thousand scattered facts, the final chapter of a story which began a generation before.

-

Gregor Mendel was Abbot of Brünn in Moravia when Darwin was at work on the Origin. He does not appear to have had any unusual interest in the problem of evolution; indeed, his main concern was with an essentially pre-Darwinian question, the nature of plant hybrids. With this problem as an avocation from his serious clerical duties, the abbot busied himself in the garden of his cloister; a leisurely, clear-headed, middle-aged churchman in whom a great scientist was spoiled. For eight years he experimented with varieties of the common pea, and in 1865 communicated to the Society of Naturalists in Brünn the substance of the discovery which is hereafter to be known as Mendel's Law, "the greatest discovery in biology since Darwin." Unfortu

nately, at that time, the Brünn Society, like the rest of the world, had other things on its mind. The controversy over Darwin and evolution was then merrily under way, and the world promptly forgot the one thing which was needed to complete Darwin's work. He, it is clear, never saw Mendel's paper. If he had, a good many books would have remained unwritten. Mendel himself appears never to have understood the full value of his own idea. Except for one short paper written in 1869 he made no effort to follow the matter out, and devoted the remaining twenty years of his life to theology and the weather, — fields where his great talent for experiment could hardly have had free vent. He died in 1884 with no suspicion that, within twenty years, his modest paper would stand alongside of Animals and Plants and Natural Inheritance, and himself, as a student of heredity, with Dar win and Galton. Somehow or other, Mendel's discovery escaped attention until four years ago, when De Vries reached it independently. Two years later Mr. Bateson, who had been among the first to realize its significance, made a translation of the two original papers; this, together with his somewhat hasty commentary, is the basis of Professor Morgan's excellent though brief account, and, in part, of Dr. Vernon's less satisfactory one. Since then, Mendel's Law has been found to hold for a considerable number of cases, both among animals and plants, but most unaccountably not to work for a few others; so that, as yet, no one knows how nearly universal it may prove to be, nor how it is to be reconciled with the older Law of Ancestral Heredity of Galton. Its latest important aspect is an ingenious attempt to apply it to the inheritance of that commonest and most obscure of all discontinuities,

sex.

One illustration will serve to make clear the practical workings of Mendel's principle. If a single rough-coated guinea-pig of either sex be introduced

into a colony of normal smooth-coated individuals, all its offspring of the first generation will be rough-coated like itself. In the next generation, if one of the parents is smooth and the other rough, the young will be half of one sort and half of the other, but if both parents are rough, three quarters will take the "dominant" rough coat. In the next, and all subsequent generations, one half of those rough-coated individuals which had one smooth-coated grandparent, and one third of those which had two smoothcoated grandparents, which were not mated, will drop out the "recessive" smooth-coatedness, and become, in all respects, like their original rough-coated progenitor, even to having only roughcoated young, no matter what their mates may have. Thus Mendel's Law, though by no means simple, is very precise. The essential part of his great discovery is that in each generation of plants or animals of mixed ancestry, a definite proportion lose one half of their mingled heritage, and revert, in equal numbers, to one or other of the pure types. As a corollary to this there is also the discovery that there may be, as among our guinea-pigs, two sorts of individuals, alike in outward appearance, but fundamentally different in having or lacking a latent quality which, when it exists, becomes patent again in a fixed proportion of their offspring. Apparently in about one case out of two in which Mendel's Law holds, the new quality or organ of a mutation is "dominant" over the old one, like the rough coat of the guineapig over the smooth, and thereby gets a fair chance to prove its fitness for survival.

If Darwin had only known this, how easily he would have disposed of objections based on the swamping effects of intercrossing!

The reader who follows out at length in the pages of our two authors the case which I have outlined here, and realizes that, in spite of all logic, common varia