LIBRARY UNIVERSTTYOF CALIFORNIA HEREDITY BY J. ARTHUR THOMSON, M.A. Regius Professor of Natural History in the University of Aberdeen THE HER- F.ERT SPENCER," "DARWINISM AND HUMAN LIFE," "THE BIOLOGY OF THE SEASONS," "INTRODUCTION TO SCIENCE," ETC.; JOINT-AUTHOR OF "THK EVOLUTION OF SEX," "EVOLUTION," ETC. SECOND EDITION LONDON JOHN MURRAY, ALBEMARLE STREET, W. 1912 Tt BIOLOGT LIBRARY 6 ALL RIGHTS RESERVED (1907) I DEDICATE THIS BOOK WITH THEIR KIND PERMISSION TO FRANCIS GALTON AND AUGUST WEISMANN WHOSE MAGISTRAL STUDIES ON HEREDITY HAVE MADE US ALL THEIR DEBTORS 260711 PREFACE TO FIRST EDITION THIS book is intended as an introduction to the study of heredity, which every one admits to be a subject of fascinating interest and of great practical importance. In recent years much progress has been made in the scientific study of heredity, and, as the literature is widely scattered, and often very technical, there may be utility in an exposition which aims at being com- prehensive and accurate, without being exhaustive or mathe- matical. Simple the exposition cannot be, if one has any ambition for thoroughness, but it is probably simple enough for those who have got beyond the pottering, platitudinarian stage, which deals in heredity with a capital H. My stacks of unused manuscript remind me sadly of how much I have had to leave out, to keep the volume approximately within the limits of the series to which it belongs; but the bibliography will enable serious students to fill in details, and follow up the clues I have given. It is arranged with a subject-index, so that the literature dealing with particular points can be seen at a glance. I have tried to avoid partisan handling of any theme, though I have been at no pains to conceal my general adherence to what is called Weismannism, or — to take a particular case — my conviction that we do not know of any instance of the trans- mission of an acquired character. I have also tried throughout to keep the practical side of the study in view, but I have re- frained from making many suggestions, in the belief that the viii PREFACE inquiry is not ripe for more than a general recommendation to take thought for the morrow by considering the ideal of Eu- genics. A glance at the book will show that much prominence has been given to three kinds of conclusions — those reached by microscopic study of the germ-cells, those reached by the appli- cation of statistical methods, and those reached through ex- periment. I have equal sympathy with all these ways of attacking the mysterious problems, and since I have not, to my lasting regret, found any opportunity, amid the continuous claims of professional duties, of working along any one of them, I can, without seeming to recommend my own wares, press a consideration of the results which have been achieved on the attention of all thoughtful men and women. The new facts are of especial interest to medical practitioners, to educationists, including clergymen, to social reformers, and to actual or prospective parents. I have, throughout, acknowledged my indebtedness to autho- rities, and the bibliography (which is merely representative) shows how many fields there are from which to glean. In particular, I have been indebted to the works of Galton, Weis- mann, Pearson, Bateson, and De Vries. I have to thank my friends Mr. E. S. Russell and Dr. John Rennie for going over the proofs, and saving the pages from many mistakes. Dr. Leslie Mackenzie was kind enough to read the chapter on Heredity and Disease, and some of his helpful suggestions have been incorporated. I have to thank Professor C. Correns and Professor H. E. Ziegler for generously allowing me to copy four admirable diagrams ; also Mr. Young Pentland and the Walter Scott Publishing Company for allowing PREFACE ix me the use of a number of figures which have done duty in other books of mine. My thanks are also due to Mr. Murray, who has encouraged me in a work which I was often tempted to abandon, whose good-humoured patience over many delays I should long since have exhausted had he been as many men are. J. A. T. THE UNIVERSITY OF ABERDEEN, August 1907. PREFACE TO SECOND EDITION THE demand for a second edition has given me the opportunity of correcting some errors — for a knowledge of which I am in part indebted to my critics — and of inserting references to some of the new discoveries that have been made in the last five years in this rapidly progressive department of Biology. J. A. T. THE UNIVERSITY OF ABERDEEN, May 1912. CONTENTS CHAPTER I PAGE HEREDITY AND INHERITANCE I DEFINED AND ILLUS- TRATED I § i. Importance of the Study of Heredity. § 2. What the Terms Mean. § 3. Heredity and Inheritance in Relation to other Bio- logical Concepts. §4-^4 Question of Words. § 5. The Problems Illustrated. § 6. Denials of Inheritance. CHAPTER II THE PHYSICAL BASIS OF INHERITANCE . . .. . 26 § i. What is true in the Great Majority of Cases. § 2. Diverse Modes of Reproduction. § 3. The Hereditary Relation in Uni- cellular Organisms. § 4. The Hereditary Relation in the Asexual Multiplication of Multicellular Organisms. § 5. Nature and Origin of the Germ-cells. § 6. Maturation of the Germ-cells. § 7. Amphimixis and the Dual Nature of Inheritance in Sexual Reproduction. § 8. Inheritance in Parthenogenesis. § 9. Wherein the Physical Basis precisely consists. CHAPTER III HEREDITY AND VARIATION ... . . .66 § I. Persistence and Novelty. § 2. The Tendency to Breed True. § 3. Different Kinds of Organic Change. § 4. Classifica- tion and Illustration of Variations. § 5. Fluctuating Variations. § 6. Discontinuous Variations. § 7. De Vries on Fluctuations and Mutations. § 8. Causes of Variation. xii CONTENTS CHAPTER IV PAGE COMMON MODES OF INHERITANCE IO6 § i. Though Prediction in Individual Cases is insecure, there ate some Common Modes of Inheritance. § 2. Certain Necessary Saving Clauses. § 3. Blended Inheritance. § 4. Exclusive Inheritance (Unilateral, Absolutely Prepotent, or Preponderant). § 5. Pani- culate Inheritance. § 6, Alternative Inheritance. § 7. Summary of Possibilities. CHAPTER V REVERSION AND ALLIED PHENOMENA . . J?;( .rllQ § i. What is meant by Reversion. § 2. Suggested Definitions. § 3. Theoretical Implications. § 4. Phenomena sometimes con- fused with Reversion. § 5. " Skipping a Generation." § 6. Men- delian Interpretation of Reversion. § 7. Reversion in Crosses. § 8. Reversion of Retrogressive Varieties. § 9. Interpretations in Terms of Reversion. § 10. Further Examples of Reversion. CHAPTER VI TELEGONY AND OTHER DISPUTED QUESTIONS . . . 143 § i. What is meant by Telegony. § 2. The Classic Case of Lord Morton's Mare. § 3. Representative Alleged Cases of Tele-' gony. § 4. Ewart's Penycuik Experiments. § 5. Suggestions which explain away Telegony. § 6. Suggestions as to how Telegonic Influence might be effected. § 7. A Statistical Suggestion. § 8. The Widespread Belief in the Occurrence of Telegony. § 9. An Instructive Family History. § 10. A Note on Xenia. § II. Maternal Impressions. CHAPTER VII THE TRANSMISSION OF ACQUIRED CHARACTERS . . 164 § i. Importance of the Question. § 2. Historical Note. § 3. Definition of the Problem. § 4. Many Misunderstandings as to the Question at Issue. § 5. Various Degrees in which Parental Modifications might affect the Offspring. § 6. Widespread Opinion in favour of Affirmative Answer. § 7. General Argument against the Transmissibility of Modifications. § 8. General Argument CONTENTS xiii PAGB for the Transmissibility of Modifications. § 9. Particular Evi- dence in support of the Affirmative Answer. § 10. As regards Mutilations and the Like. § n. Brown-Sequard's Experiments on Guinea-pigs. § 12. Negative Evidence in favour of the Affirma- tive Answer. § 13. The Logical Position of the Argument. § 14. Indirect Importance of Modifications. § 15. Practical Con- siderations. CHAPTER VIII HEREDITY AND DISEASE . . . . . . . 25O § i. Health and Disease. § 2. Misunderstandings in regard to the " Inheritance " of Disease. § 3. Are Acquired Diseases transmissible ? § 4. Can a Disease be transmitted ? § 5. Pre- dispositions to Disease. § 6. Particular Cases. § 7. Defects, Multiplicities, Malformations, and other Abnormalities. § 8. Some Provisional Propositions. § 9. Immunity. § 10. Note on Chromosomes in Man. § n. Anticipation and Intensification of Disease. § 12. Practical Considerations. CHAPTER IX STATISTICAL STUDY OF INHERITANCE. • . . 309 § i. Statistical and Physiological Inquiries. § 2. Historical Note. § 3. A Hint of the Statistical Mode of Procedure. § 4. Filial Regression. § 5. Law of Ancestral Inheritance. § 6. Criti- cisms of Gallon's Law. § 7. Illustration of Results reached by Statistical Study. CHAPTER X EXPERIMENTAL STUDY OF INHERITANCE .... 336 § i. Mendel's Discoveries. § 2. Theoretical Interpretation. § 3. Corroborations. § 4. Illustrations oj Mendelian Inheritance. § 5. Mendel's Discovery in Relation to Other Conclusions. § 6» Practical Importance of Mendel's Discovery. § 7. Other Experi- ments on Heredity. § 8, Consanguinity. CHAPTER XI - HISTORY OF THEORIES OF HEREDITY AND INHERITANCE . 391 § i. What is required of Theories of Heredity and Inheritance. § 2. The Old Theories of Heredity. § 3. Theories of Pangenesis> § 4. Theory of Genetic or Germinal Continuity. xiv CONTENTS CHAPTER XII PAGE HEREDITY AND DEVELOPMENT '. ' . .- ' „ . . . 412 § i. Theories of Development. § 2. Weismann's Theory of the Germ- Plasm. § 3. Note on Rival Theories. § 4. Weismann's Theory of Germinal Selection. CHAPTER XIII HEREDITY AND SEX . . , . . . . 472 § i. Relations between Sex and Inheritance. § 2. The Deter- mination of Sex. § 3. Different Ways of Attacking the Problem. § 4. Classification of Theories. § 5. First Theory : Environment affects Offspring. § 6. Second Theory : Fertilisation is Decisive. § 7. Third Theory : Two Kinds of Germ-Cells. § 8. Fourth Theory : Maleness and Femaleness are Mendelian Characters. § 9. Fifth Theory : Nurtural Influences operate on the Germ-Cells through the Parents. § 10. Another Way of looking at the Facts. § ii. Conclusion. CHAPTER XIV SOCIAL ASPECTS OF BIOLOGICAL RESULTS . . ... 5IO § i. Relations of Biology and Sociology. § 2. The Chief Value of the Sociological Appeal to Biology. § 3. Originative Factors in Evolution. § 4. Social Aspects of Heredity. § 5. Directive Factors in Evolution. BIBLIOGRAPHY .... . , * 543 SUBJECT-INDEX TO BIBLIOGRAPHY . * , . . 605 INDEX . . . , . . . . V . 619 LIST OF ILLUSTRATIONS FIG. PAGE 1. OVUM OF A THREADWORM (FROM CARNOY) .... 5 2. DIAGRAM OF CELL DIVISION (AFTER BOVERI) .... 31 3. DIAGRAM OF CELL STRUCTURE (AFTER WILSON) 33 4. " COMET-FORM " OF STARFISH (AFTER HAECKEL) 35 5. ASEXUAL REPRODUCTION OF A SEA-WORM (AFTER MC!NTOSH) . 36 6. DIAGRAM OF OVUM AND SOMATIC CELL (AFTER CARNOY) facing 38 7. VOLVOX GLOBATOR t 39 8. FORMS OF SPERMATOZOA 41 9. DIAGRAM OF GERMINAL CONTINUITY (AFTER WILSON) . . 43 10. PARALLELISM OF SPERM ATOGENESIS AND OOGENESIS (AFTER BOVERI) 47 11. DIAGRAM OF REDUCTION AND AMPHIMIXIS (AFTER ZIEGLER) facing 48 12. FERTILISED OVUM OF ASCARIS (AFTER BOVERI) .... 49 13. DIAGRAM OF MATURATION AND FERTILISATION (FROM ZIEGLER). facing 51 14. CHROMATIN ELEMENTS OF NUCLEI (AFTER PFITZNER) . . 59 15. DIAGRAM OF FERTILISATION IN ASCARIS (AFTER BOVERI) . 60, 61 16. A POLLEN GRAIN WITH ITS NUCLEI (FROM CARNOY) ... 63 17. DIAGRAM TO ILLUSTRATE THE DIFFERENCE BETWEEN MODIFICA- TIONS AND VARIATIONS *.*.... 7! 18. VARIETIES OF WALL-LIZARD (AFTER EIMER) .... 74 19. VARIATIONS IN WASP (AFTER KELLOGG AND BELL) . facing 76 20. VARIATIONS IN BEETLE (AFTER TOWER) ... ,,76 21. MUTATION IN MEDUSOIDS ««»..... 89 22. MUTATIONS OF HART'S TONGUE FERN (AFTER LOWE) . facing 98 23. KARYOKINESIS (AFTER FLEMMING) . . 102 24. LEAVES OF WILLOW (AFTER WIESNER) in 25. DEVONSHIRE PONY WITH STRIPES (FROM DARWIN) . . .121 26. VARIETIES OF DOMESTIC PIGEON (AFTER DARWIN) . . 139 •27. BRINE-SHRIMP, ARTEMIA SALINA . . . . . .213 27a. TAIL-LOBES OF ARTEMIA SALINA (AFTER SCHMANKEWITSCH) . 213 a8> HALF-LOP RABBIT (FROM DARWIN) ...... ,289 xv xvi LIST OF ILLUSTRATIONS FIG. PAGE 29. DIAGRAM ILLUSTRATING GALTON'S LAW OF ANCESTRAL INHERIT- ANCE (AFTER GALTON AND MESTON) . . .325 30. DIAGRAM TO ILLUSTRATE THE DIFFERENCE BETWEEN STATIS- TICAL AND PHYSIOLOGICAL FORMULATION (AFTER DARBISHIRE) 330 31. PEAS SHOWING MENDEL'S LAW .... facing 339 32. DIAGRAM OF MENDEL'S LAW ... . . ,, 340 33. DIAGRAM OF MENDELIAN INHERITANCE IN MIRABILIS JALAPA facing 343 34. DIAGRAM ILLUSTRATING MENDEL'S LAW. . ~ « . ' . ,, 347 35. DIAGRAM ILLUSTRATING SEGREGATION OF GERM-CELLS . . 344 36. COMBS OF FOWLS . . . . . .facing 353 37. HYBRIDISATION IN MIRABILIS JALAPA (FROM CORRENS) . „ 355 38. MENDELIAN PHENOMENA IN NETTLES (FROM CORRENS) . 357 39. MENDELIAN PHENOMENA IN WHEAT (AFTER R. H. BIFFEN) facing 358 40. MENDELIAN PHENOMENA IN HELIX HORTENSIS (AFTER LANG) facing 360 4oa. PURE LINES IN PARAMCECIUM (FROM JENNINGS) . . . 378 41. VARIETIES OF WHEAT (AFTER R. H. BIFFEN) , . . 384 42. MODES OF SEGMENTATION . . . . facing 438 43. RELATION BETWEEN REPRODUCTIVE CELLS AND THE " BODY " . 430 44. DIAGRAM OF MATURATION AND FERTILISATION • ., • ' . 436 45. SEXUAL DIMORPHISM IN HUMMING-BIRDS (FROM DARWIN, AFTER BREHM) . ' . . ... . . 479 46. SEXUAL DIMORPHISM IN GRASSHOPPERS (FROM DARWIN) . .481 47. DIAGRAM ILLUSTRATING THE RELATION BETWEEN REPRODUCTION AND INDIVIDUATION . . . . .* • 539 HEREDITY CHAPTER I HEREDITY AND INHERITANCE : DEFINED AND ILLUSTRATED § I. Importance of the Study of Heredity. § 2. What the Terms Mean. § 3. Heredity and Inheritance in Relation to other Bio logical Concepts. § 4. A Question of Words. § 5. The Problems Illustrated. § 6. Denials of Inheritance. § i. Importance of the Study of Heredity Heredity determines the Individual Life. — There are no scientific problems of greater human interest than those of Heredity — that is to say, the genetic relation between successive generations. Since the issues of the individual life are in great part determined by what the living creature is or has to start with, in virtue of its hereditary relation to parents and ancestors, we cannot disregard the facts of heredity in our interpretation of the past, our conduct in the present, or our forecasting of the future. Great importance undoubtedly attaches to Environ- i 2 HEREDITY AND INHERITANCE ment in the widest sense, — food, climate, housing, scenery, and the animate milieu ; and to Function in the widest sense,— exercise, education, occupation, or the lack of these ; but all these potent influences act upon an organism whose fundamental nature is determined, though not rigidly fixed, by its Heredity — that is, we repeat, by its genetic relation to its forebears. As Herbert Spencer said, " Inherited constitution must ever be the chief factor in determining character " ; as Disraeli said, more epigrammatically and less correctly, " Race is everything." Heredity is a Condition of all Organic Evolution. — In the same way, when we consider the race rather than the individual, we must admit that in so far as evolution depends on inborn organic changes, on what is bred in the bone and imbued in the blood, as distinguished from individual efforts and acquirements, external institutions and traditional culture, it is conditioned by the hereditary relation which binds one generation to another. Heredity is a condition of all organic evolution. Innate changes or variations, which form the raw material of constitutional progress or degeneracy, have direct racial importance because they are certainly transmissible ; while, on the other hand, bodily modifications or acquired characters, due to changes in environment or in function, probably have no direct racial importance, since there is little or no evidence that they are ever hereditarily entailed. They are individually important, and in human society they are of much moment, but if they are not transmissible they do not take organic grip, and they cannot afford material for selection to work with. For the human race, the external heritage of tradition, institutions, and law, the permanent products of literature and art, the registrated results of science, and so on, are of paramount importance, but they are outside the immediate problem of organic or natural inheritance. As far as the slow, sure process of constitutional or organic evolution is concerned, everything depends on the heritable resemblances and the heritable variations which form A CONDITION OF ZVOLUTIOtf $ the material on which the many diverse forms of selection and isolation operate. In olden days thoughtful men seemed to see the threads of life within the hands of three sister Fates, — of one who held the distaff, of another who offered flowers, and of a third who bore the abhorred shears of death. So, in Scandinavia, the young child was visited by three sister Norns, who brought characteristic gifts of the past, the present, and the future, which ruled the life to be as surely as did the hands of the three Fates. So, too, in days of scientific enlightenment, we still think of Fates and Norns, though our conceptions and terms are very different. What the living creature is or has to start with in virtue of its hereditary relation ; what it does in the course of its activity ; what surrounding influences play upon it, — these are the three determining factors of life. Heredity, function, and environ- ment— famille, travail, lieu — are the three sides of the bio- logical prism, by which, scientifically, we seek to analyse the light of life, never forgetting that there may be other components which we cannot deal with scientifically, just as there are rays of light which our eyes can never see. In novels like Zola's Dr. Pascal, in plays like Ibsen's Ghosts, in sermons and newspaper articles, in large books and health lectures, in season and out of season, we have all heard in the last few years much about the importance of heredity ; and though it is to be feared that many widespread impressions on the subject are misleading, the awakening of keen interest is in itself a symptom of progress. What is now required is a serious study of what has been securely established. Otherwise we shall continue to think in platitudes and act on guesses. Practical Importance to Breeders and Cultivators.— And what is important in regard to Man's heredity is even more demonstrably important in regard to his domesticated animals and cultivated plants. What has been achieved in the past in regard to horses and cattle, pigeons and poultry, cereals and 4 HEREDITY AND INHERITANCE chrysanthemums, by experimental cleverness and infinite patience, may be surpassed in the future if breeders and cultiva- tors can attain to a better understanding of the more or less obscure laws of inheritance on which all their results depend. Importance in Biological Theory. — The study of heredity is also of fundamental importance in the domain of pure science, in the biologist's attempt to interpret the process of evolution by which the complexities of our present-day fauna and flora have gradually arisen from simpler antecedents. For heredity is obviously one of the conditions of evolution, — of continuance as well as of progress. There would have been heredity even if there had been a monotonous world of Protists without any evolution at all, but there could not have been any evolution in the animate world without heredity as one of its conditions. The study of heredity is inextricably bound up with the problems of development, reproduction, fertilisation, variation, and so on ; in short, it is one of the central themes of Biology. § 2. What the Terms Mean The Terms are tinged with Metaphor. — In the popular, if not also in the biological mind, there often lurks the idea of a hypothetical agent possessing the organism and uniting the congeries of its characters. Expressed in diverse ways, there is a prevalent conception of an organismal unity which gives coherence to the sum of qualities (see Sandeman, 1896). Espe- cially in reference to higher animals with a rich mental life, many find it impossible not to think of a " soul " or " self " to which the body belongs. Naturally enough, therefore, the reappear- ance in the offspring of qualities which characterised its parents or its ancestors has been persistently likened to the inheritance of a legacy. But this is to some extent a metaphorical expres- sion, and not without its dangers. At first the Organism and the Inheritance are Identical.— A WHAT THE TERMS MEAN 5 moment's consideration suffices to show that ideas and phrases borrowed from the inheritance of property — something quite apart from the individual who inherits — are apt to cause ob- scurity and fallacy when applied to the inheritance of characters which literally constitute the organism and are inseparable from it. Therefore, as the biological conception of inheritance seems still to surfer from the irrelevancy of the analogy to which FIG. i. — Ovum of a threadworm (Ascaris), showing (a) the chromosomes of the nucleus, and the reserve products in the surrounding cell- substance. — From Carnoy. the term owes its origin, let us dwell for a little on the fact that, at the start of an individual life, the inheritance and the organism are identical. In other words, the idea of organic inheritance is merely a convenient scientific abstraction, by which we seek to distinguish what the organism is, in virtue of its germinal origin, from what it is as the result of the influence of ensuing circumstances. If we may use Galton's and Shakespeare's terms, the idea of organic inheritance is an abstraction by which 6 HEREDITY AND INHERITANCE we seek to distinguish what is due to " Nature " from what is due to " Nurture." Heredity and Inheritance defined. — In regard to property there is a clear distinction between the heir and the estate which he inherits, but at the beginning of an individual life we cannot biologically draw any such distinction. The organism and its inheritance are, to begin with, one and the same. It is easy to make this clear. Every living creature arises from a parent or from parents more or less like itself ; this reproductive or genetic relation has a visible material basis in the germinal matter (usually egg-cell and sperm-cell) liberated from the parental body or bodies ; by inheritance we mean all the qualities or characters which have their initial seat, their physical basis, in the fertilised egg-cell ; the expression of this inheritance in development results in the organism. Thus, heredity is no entity, no force, no principle, but a convenient term for the genetic relation between successive generations, and inheritance includes all that the organism is or has to start with in virtue of its hereditary relation. Nature and Nurture. — The fertilised egg-cell implicitly con- tains, in some way which we cannot image, the potentiality of a living creature, — a tree, a daisy, a horse, a man. If this rudiment is to be realised there must be an appropriate environment, supplying food and oxygen and liberating-stimuli of many kinds. Surrounding influences — maternal or external — begin to play upon the developing germ, and without these influences the inheritance could not be expressed, the potentiali- ties could not be realised. Thus, the organic inheritance implies an environment, apart from which it means nothing and can achieve nothing. Indeed, it is only by an abstraction that we can separate any living creature from an environment in which it can live. Life implies persistent action and reaction between organism and environment. while the inherited nature and its possibilities of action NATURE AND NURTURE 7 and reaction must be regarded as rigorously determined by the parental and ancestral contributions, the nurture — the en- vironmental influences — must not be thought of as pre- deter- mined. In fact, the surrounding influences are very variable, and the nature of the young organism may be profoundly changed by them. Thus, we soon find it possible to distinguish between the main features, which are the normal realisations of the inheritance in a normal environment, and peculiarities which are due to peculiarities in nurture. The characters of a newly-hatched chick stepping out of the imprisoning egg-shell are in the main strictly hereditary ; but they need not be alto- gether so, for during the three weeks before hatching there has been some opportunity for peculiarities in the environment to leave their mark on the developing creature. Still more is this the case with the typical mammalian embryo, which develops often for many months as a sort of internal parasite within the mother — in a complex and variable environment. And as life goes on, peculiarities due to nurture continue to be superimposed on the hereditary qualities. William of Occam's Razor. — Our preliminary attempt to get rid of capitals, to make the terms heredity and inheritance quite objective, is in line with what has occurred in other departments of science. For one of the distinctive features of the nineteenth century has been a reduction in the number of supposed separate powers or entities — the use of William of Occam's razor, in fact. " Entia non sunt multiplicanda prater necessitate™." " Caloric " was one of the first to be eliminated, yielding to the modern interpretation of heat " as a mode of motion " ; " Light " had to follow, when the undulatory or the electro-magnetic theory of its nature was av^epted ; a specific " Vital Force " is disowned even by the Neo-vitalists ; " Force " itself has become a mere measure of motion ; and even " Matter " tends to be resolved into units of negative electricity, carrying with them a bound portion of the ether in which they are bathed ; and so on. In 8 HEREDITY AND INHERITANCE view of this progress towards greater precision and simplification of phraseology, it cannot be a matter for surprise that a biologist should affirm that to speak of the " Principle of Heredity " in organisms is like speaking of the " Principle of Horologity " in clocks. The sooner we get rid of such verbiage the better for clear thinking, since heredity is certainly no power, or force, or principle, but a convenient term for the relation of organic or genetic continuity which binds generation to generation. Ancestors, grandparents, parents are real enough ; children and children's children are also very real ; heredity is a term for the relation of genetic continuity which binds them together. We study it as a relation of resemblances and differences which can be measured or weighed, or in some way computed ; as a relation which is sustained by a more or less visible material basis — namely, the germinal matter § 3. Heredity and Inheritance in Relation to other Biological Concepts Development. — All living creatures arise from parents more or less like themselves. The reproduction may be asexual, — by fission, fragmentation, budding, and similar processes ; or sexual, — by special germ-cells or gametes, which usually unite in pairs (fertilisation or amphimixis) to start a new individual body. Whatever the mode of reproduction may be — and that is a long story by itself — there is a hereditary relation, a genetic continuity. It is the business of the theory of heredity to inquire into the precise nature of this genetic relation in the diverse modes of reproduction. In what relation, for instance, does a liberated germ-cell or gamete stand to the body which liberates it ? In what relation does a fertilised ovum stand to the germ- cells of the body into which it develops ? What contribution does each parent make to the inheritance ? Do ancestors also make contributions, and if so, how ? To answer this kind of question is the business of (he theory of heredity. RELATION TO OTHER BIOLOGICAL CONCEPTS 9 The separated fragment or the liberated germ-cell has in it the possibility of becoming, in an appropriate environment, a fully-developed organism. Is it possible to form any conception — verifiable or speculative — of the manner in which the in- heritance is thus condensed into a fragment or into a germ-cell ? Is it possible to picture in any way how the potentialities come to be realised in development ; how the obviously complex grows out of the apparently simple ? To answer these and similar questions is the business of the theory of development. The facts of inheritance are those which rise into prominence when we compare the characters of an organism with those of its parents and its offspring, or when we compare the characters of one generation with those of its predecessors and successors. This is a thoroughly concrete study, for the facts observed are quite independent of any theory of the precise organic relation which binds generation to generation (the theory of heredity), and are also quite independent of any theory as to the way in which the germ grows into the adult (the theory of development), It is, in the main, an experimental and statistical study. Before the middle of the nineteenth century considerable attention was given to what may be called the demonstration of the general fact of inheritance — that like tends to beget like. This had, indeed, always been the general opinion of physicians and naturalists, as well as of the laity, but it was a useful task to collect documentary evidence showing that all the inborn characteristics of an organism, whether physical or psychical, normal or abnormal, important or trivial, were transmissible to the offspring, if the possibility of having offspring had not been excluded. This task of demonstrating inheritance was well finished by Prosper Lucas, whose large treatise, published in 1847, gave ample evidence for what we now take for granted,— that the present is the child of the past ; that our start in life is no haphazard affair, but is rigorously determined by our paren- tage and ancestry ; that all kinds of inborn characteristics may io HEREDITY AND INHERITANCE be transmitted from generation to generation. In short, the fundamental importance of inheritance was long ago demon- strated up to the hilt. It remains, however, (i) to make the evidence of transmissibility more precise and systematic ; (2) to inquire into the trans- missibility of subtle characters such as longevity and fecundity ; (3) to discover the different degrees of transmissibility, for some characters are much more heritable than others ; and (4) to classify different modes of hereditary resemblance — e.g. blending of the characters of the two parents, taking after the father in one feature and after the mother in another, apparently resem- bling one parent only, rehabilitating a grandsire's features, harking back to a remoter ancestor, and so on. What happens when there is close in-breeding or pairing within a narrow radius of relationship ? What happens when two hybrids are paired ? In what sense, if any, is a disease heritable ? These and many similar questions will be discussed in our inquiry into the facts of inheritance. Variation — Whenever we begin to compare the characters of an organism with those of its parents, we discover that the familiar saying, " Like begets like," must be modified into, " Like tends to beget like." On the one hand, the child is like its parents, " a chip of the old block," a literal reproduction ; on the other hand, the child is something original, a new pattern, a fresh start — leading the race. We do not gather grapes of thorns, or figs of thistles ; yet two brothers may be very unlike one another or either of their parents, and even the peas in one pod may be different. On the one hand, there is a tendency towards continuity, towards persistence of characters, towards complete hereditary resemblance — in short, a kind of organic inertia in a family or stock or species. On the other hand, there is a tendency towards variation, towards new departures, to- wards incomplete hereditary resemblance, or much more than that. It is necessary to hold the balance between these two VARIATIONS AND MODIFICATIONS u sets of facts, both expressions of the hereditary relation, — inertia, persistence, continuity, resemblances, on the one hand ; deviation, novelty, differences, on the other. Can we hope to discriminate an apparent difference between parent and offspring, which is really due to an incompleteness in the expression of the inheritance, from a real difference, which is due to the dropping out of an old hereditary item or the addition of a new one ? Can we distinguish between inborn peculiarities — germinal variations — and acquired, nurtural pecu- liarities ? Can we distinguish between variations which seem to be simply a little less or a little more of some hereditary character, and variations which involve something new ? These and similar questions must be faced in the study of variation. Modifications. — Furthermore, whenever the study of the facts of inheritance becomes critical, it is necessary to try to dis- criminate between inborn changes, which must have a germinal origin, and are therefore in the strict sense inherited, and are liable to be transmitted, and those theoretically quite different changes which are acquired by the body of the individual off- spring as the result of peculiarities in function and environment. This is the contrast between germinal variations and bodily modifications, a contrast which is of fundamental importance in several ways. It is important to try to distinguish resem- blances and differences due to inherited nature from resemblances and differences due to nurture. A collier may have his collier father's red hair, and he may also resemble him in having " col- lier's lung." But while the first resemblance is a fact of in- heritance, the second is due to the similarity in their life-con- ditions. This distinction remains important whatever conclusion be reached in regard to the transmissibility of modifications, but its importance is enhanced when we discover that practically all variations (except sterility) are transmissible, though not always transmitted, and that the evidence of any modification 12 HEREDITY AND INHERITANCE being transmissible, among multicellular organisms reproducing sexually, is extremely doubtful. Evolution. — Briefly and concretely stated, the general doc- trine of organic evolution suggests, as we all know, that the plants and animals now around us are the results of natural processes of growth and change working throughout unthinkably long ages ; that the forms we see are the lineal descendants of ancestors on the whole somewhat simpler ; that these are de- scended from yet simpler forms, and so on, backwards, till we lose our clue in the unknown, but doubtless momentous, vital events of pre-Cambrian ages, or, in other words, in the thick mist of life's beginnings. The essentially simple idea is that the present is the child of the past, and the parent of the future. It is a way of looking at organic history, a genetic description, a modal formulation. A process of Becoming leads to a new phase of Being ; the study of evolution is a study of Werden und Vergehen und Weiter-werden. But we have to pass from a modal interpretation to a causal one. We have to try to discover the factors in the age-long process, and this leads us into a region where at present uncer- tainties abound. As biologists we start with the postulate of simple living organisms — feeding, working, growing, wasting, reproducing in an appropriate environment. And we try to discover the possible factors in the long evolution-process, the outcome of which is the present-day world of life. Amid all the uncertainties, this is certain, that the fundamental condition of evolution is that genetic relation which we call heredity, — a relation such that it admits, on the one hand, of a continuity of hereditary resemblance from generation to generation ; and, on the other hand, of an organic changefulness which we call variability. Without the hereditary relation there could have been no succession of generations at all. Without hereditary resemblance on the one hand, and hereditary variation on the other, there could have been no evolution. Any discussion of A QUESTION OF WORDS 13 the secondary or directive factors which operate upon the raw materials of progress which variability supplies — notably Selection and Isolation — is not relevant at present. § 4. A Question of Words In every discussion with a serious purpose it is important that there should be clearness as to the terms used. We must, therefore, ask the reader to notice our definition of the chief terms. Thus by " heredity " we do not mean the general fact of observation that like tends to beget like, nor a power making for continuity or persistence of characters — to be opposed to the power of varying — nor anything but the organic or genetic relation between successive generations ; and by " inheritance " we mean " organic inheritance " — all that the organism is or has to start with in virtue of its hereditary relation to parents and ancestors. We do not forget that for man in particular there is an external heritage — a social inheritance — which counts for much. By innate or inborn we mean all that is potentially implied in the fertilised egg-cell ; by the expression of the inheritance we mean the realisation of inborn potentialities in the course of development in an appropriate environment ; by a congenital character (pace many medical writers) we mean one demonstrable at birth, which is not necessarily germinal, being often due to peculiarities — e.g. infection or poisoning or mechanical injury during pre-natal development. Thus, tubercle may be con- genital, but it is never inherited. By modifications or acquired characters we mean structural changes in the body induced by changes in the environment or in the function, and such that they transcend the limit of organic elasticity, and therefore persist after the inducing conditions have ceased to operate. By a variation we mean not any observed difference between offspring and parent, between an individual and the mean of I4 HEREDITY AND INHERITANCE the stock in respect of a given character ; we mean observed differences minus all bodily modifications, we mean changes which have a germinal origin. These definitions will become clearer in the course of our exposition. Our present point is to warn the reader against starting on his journey without reading the conditions on the ticket, and to protest against the slackness with which the terms are so often used. A large part of the energy expended on the long-drawn-out controversy as to the transmission of acquired characters or modifications has been wasted through inattention to the precise significance of the technical terms employed.* To speak of a man " fighting against his heredity " may express a real fact, but it is verbally erroneous. The American's question, " Is my grandfather's environment my heredity ? " is an offence against ordinary English as well as against scientific phrasing ; it should probably read, " Have the structural changes induced by environmental influences on my grand- father's body had any effect on my inheritance ? " Nor can we pardon from an expert such a sentence as this, " I look upon Heredity as an acquired character, the same as form or colour, or sensation is, and not as an original endowment of matter " (Bailey, 1896, p. 23). When the moralist writes : " The only limitations imposed on a man are those which his own nature makes," the biologist asks, " But what is his own nature ? Is * It may be noted that Galton's work on Natural Inheritance is rightly so entitled, for it deals mainly with a statistical comparison of the char- acters of successive generations. Inheritance is also the chief subject of the works of Lucas and Ribot, although these have heredity for their title. Or, to take another example, Weismann's work entitled The Germ- Plasm, a Theory of Heredity, is in great part a theory of heredity, but, naturally enough, it is also in great part a theory of development. The German language has the same word, Vererbung, for both Heredity and Inheritance. As the English language is rich in related terms, laxity of expression is less excusable. Besides " heredity " and " inheritance " we have " heritage," " transmission," and so on. It may be convenient to speak of the parent as transmitting and of the offspring as inheriting. DEFINITIONS 15 it not the expression of a predetermined inheritance in a more or less predetermined environment ? " Definitions of " Heredity." — It may be of interest to give a few samples of definitions : " The word ' Heritage ' has a more limited meaning than ' Nature,' or the sum of inborn qualities. Heritage is confined to that which is inherited, while Nature also includes those individual variations that are due to other causes than heredity, and which act before birth." — Francis Galton, Natural Inheritance, 1898, p. 293. " Heredity is the law which accounts for the change of type between parent and offspring, i.e. the progression from the racial towards the parental type." — Karl Pearson, The Grammar of Science, 1900, p. 474. " Under heredity we understand the transference to the offspring of qualities of the parent or parents." — T. H. Montgomery, Jr., Proc. American Phil. Soc. xliii. 1904, p. 5. [But the line of descent is from germ-cell to germ-ceil. The parent is the custodian or trustee of the germ-cells rather than their producer. It is too metaphorical to speak of the " parent transferring qualities to the offspring." The hereditary relation includes the occurrence of variations as well as the reproduction of likenesses. And what are the offspring apart from their inheritance ?] " ' Heredity ' is most usually defined by biologists as referring generally to all phenomena covered by the aphorism ' like begets like.' In this sense it denotes, inter alia, the phenomenon of the constancy of specific or racial types and of sexual characters ; a character may be said to be inherited when it always, in one genera- tion after another, is one of the characters of the species, of the race, or of the one sex of the race, as distinct from the other. The species, race, or sex, so to speak, ' begets its like ' as a whole. But then a further question remains ; even if the type of the race is constant, do individual types within the race beget their like ? In so far as any individual diverges in character from the mean of the race, do his offspring tend to diverge in the same direction, or not ? It is to this question that statisticians have confined them- selves, and they speak of a character being ' inherited ' or not according as the answer to the question is yes or no — they deal solely with what we may term ' individual heredity.' " — G. Udney Yule, 1902, p. 196. [Biologists are as much concerned with individual 16 HEREDITY AND INHERITANCE heredity as statisticians are, indeed more so ; statistical results are based on individual data, but they do not admit of individual application.] " Living matter has the special property of adding to its bulk by taking up the chemical elements which it requires and building up the food so taken as additional living matter. It further has the power of separating from itself minute particles or germs which feed and grow independently and thus multiply their kind. It is a fundamental character of this process of reproduction that the detached or pullulated germ inherits or carries with it from its parents the peculiarities of form and structure of its parent. This is the property known as Heredity. It is most essentially modified by another property — namely, that though eventually growing to be closely like the parent, the germ (especially when it is formed, as is usual, by the fusion of two germs from two separate parents) is never identical in all respects with the parent. It shows Variation. In virtue of Heredity, the new congenital variations shown by a new generation are transmitted to their offspring when in due time they pullulate or produce germs." — E. Ray Lankester, Kingdom of Man, 1907, p. 10. " By inheritance we mean those methods and processes by which the constitution and characteristics of an animal or plant are handed on to its offspring, this transmission of characters being, of course, associated with the fact that the offspring is developed by the processes of growth out of a small fragment detached from the parent organism." — R. H. Lock, Recent Progress in the Study of Variation, Heredity, and Evolution, 1906, p. i. " Heredity. — The transference of similar characters from one generation of organisms to another, a process effected by means of the germ-cells or gametes." — Lock, op. cit. p. 292. § 5. The Problems Illustrated Even in ancient times men pondered over the resemblances and differences between children and their parents, and wondered as to the nature of the bond which links generation to generation. But although the problems are old, the precise study of them is altogether modern. The foundations of embryology had to be laid, the nature and origin of the physical basis of inheritance THE PROBLEMS ILLUSTRATED 17 — the germ- cells — had to be elucidated, the general idea of evolu- tion had to be realised, before the problems of heredity and inheritance could even be stated with precision. Moreover, it seems to have required the experience of many years of "fumbling" before the main body of biologists became con- vinced that the problems could not be satisfactorily studied in the armchair, nor settled by a priori argument. Now, however, it is unanimously agreed that a satisfactory study of heredity and inheritance demands a minute inquiry into the history of the germ-cells, a statistical study of the characters of successive generations, a careful criticism of the older data and of popular impressions, and a testing of hypotheses by experimental breeding. Let us give a few random illustrations in order to show what some of the problems are : The race-horse Eclipse was the sire of many foals : it is a problem in heredity to compare them with him, and to inquire into the vital arrangements, in virtue of which many of them reproduced his remarkable quality of swiftness. He had also a peculiar, quite useless spot of colour, which reappeared even in the sixth generation of his progeny. In the ancestry of Kaiser Wilhelm II. there have been four grandparents, eight great-grandparents, fourteen (not 16) great- great-grandparents, twenty-four (not 32) great-great-great-grand- parents : it is a problem in heredity to compare the qualities of these successive generations of ancestors, and to inquire if they render more intelligible the illustrious personality whose doings and sayings are familiar to us all. The assassin of the Empress of Austria is said to have been the child of a dissolute mother and a dipsomaniac father : it is a problem in heredity to inquire whether this parentage may render more intelligible an outrage which made Europe shudder. A white man of considerable intellectual ability marries a negro woman of great physical beauty and strength ; the result 2 18 HEREDITY AND INHERITANCE may be — has been — a mulatto who inherits some of his father's intellectual virtue and some of his mother's physical strength, including, for instance, a peculiar insusceptibility to yellow fever. Here are complex problems of inheritance. How is it that certain characteristics of the son are almost wholly of paternal origin, while in other respects he takes after his mother ? An English sheep-dog may show a paternal eye on one side of the head, a maternal eye on the other. A piebald foal may have its mother's hair on some patches, its father's hair on others. Such cases raise the problem of the different modes of hereditary resemblance, of the mosaic-like constitution of an inheritance, and of the various ways in which this may find expression in development. Given in our British population a thousand fathers six feet high, we can predict with great accuracy the average height of their sons. Though we cannot make any prediction as to an individual famity, we know that the average height of all the sons of these tall men will be nearer the average height of the total male population than the height of six feet is. We know, however, that the tall do not always beget the tall, nor the small the small ; that stature in mankind is a character that blends ; and that even among the sons of the thousand fathers we have spoken of, there will be every gradation between the tallest and the smallest. How different this is from stature in pure-bred peas, for if a tall variety of pea be crossed with a dwarf, all the offspring are tall, and among their offspring in turn three-fourths are tall and one-fourth dwarf, but none between the two. V White fowls crossed with black ones often have white off- spring ; black guinea-pigs crossed with white ones have black offspring ; black-eyed white guinea-pigs crossed with albinos have black offspring. It seems at first sight arbitrary, but a rational interpretation of each of these results has been given. THE PROBLEMS ILLUSTRATED 19 A pair of blue Andalusian fowls of selected breed have chickens. But only about half of these are " blue," the rest are blacks or splashed whites. Why is this ? The blacks inbred produce only blacks, the splashed whites produce splashed whites or whites, but if the blacks and splashed whites are paired the progeny is altogether " blue." Why is this ? We read of a mare which, after bearing a foal to a quagga, bore a zebra-striped foal to a horse. Breeders of dogs say that a thoroughbred bitch is spoilt for true breeding if she has once been crossed by a mongrel. Is it possible that a father can influence the subsequent offspring of the same mother by a different father ? This is a problem partly in scientific criticism of evidence, but it raises interesting questions regarding the physiology of reproduction and regarding the hereditary relation. In the sixteenth century Montaigne was puzzled by the fact that, at the age of forty-five, he developed, just like his father, a stone in the bladder. The puzzle of the supposed legacy had its fine point in the fact that his father did not develop his stone till he was sixty-seven years of age, or twenty-five years after Montaigne was born ! It is possible that there was here an interesting problem in inheritance ; but the likelihood is that it merely illustrated the commonest of phenomena, the inheritance of a constitutional tendency and the repetition of more or less similar habits of life. Far too much has been made of homochronous heredity ! — i.e. of the fact that some item in the inheritance may be ex- pressed in the offspring at the same age as in the parents. Thus two brothers, their father, and their maternal grandfather be- came deaf at the age of forty ; blindness occurred in a father and in his four children at the age of twenty-one. But if the constitutions are similar and if the conditions of life are similar, it is not surprising that the expression of an item in the con- stitution should reach its climax at the same age. A case is recorded of abnormalities in the fingers traceable 20 HEREDITY AND INHERITANCE through six generations, and the pathologist Bouchut (cited by Ziegler) refers the origin of the evil to the rage of an ancestor, who terrified his wife during her pregnancy with the wish that the fingers with which she had plucked an apple against his orders might be cut off ! Apart from the story's quaint sugges- tion of a much older episode, it requires but an elementary know- ledge of the facts of heredity and inheritance to convince us that the alleged cause was inadequate to account for the effects. In two hundred families tainted with a predisposition to haemophilia — an excessive and chronic liability to immoderate haemorrhage— Grandidier * found six hundred and nine male " bleeders." It is a problem of inheritance (and partly perhaps of sexual physiology) to discover why the disease should be restricted to males ; and the interest of the problem is enhanced by the fact that the disease rarely passes from father to son, but usually from a male parent, through an apparently unaffected daughter, to a grandson. In short, the female offspring of bleeders hand on the taint to male offspring, without themselves showing the disease, f De Candolle J reported from American statistics that thirty per cent, of the children of congenitally deaf-mute parents were deaf-mute, but that the percentage was fifteen when only one parent was congenitally deaf-mute. It is a problem of heredity to interpret the greater frequency of inheritance when both parents were affected. While there is much and justifiable uncertainty in regard to the origin of what are called instincts, there is no doubt that an organism's inheritance often includes the power of carrying out a complex series of operations without experience and without education when the appropriate stimuli occur. * Grandidier, Die Hamophilie (1876). f Bulloch and Fildes, Haemophilia. Treasury of Inheritance, Pt. xiva. (1911). J De Candolle, Arch. Sci. Phys. Nat. xv. p. 25, cited by Ziegler (1886). THE PROBLEMS ILLUSTRATED 21 Simple illustrations are afforded by instinctive likes and dis- likes, attractions and repulsions. " So old is the feud between the cat and the dog," says Spalding, " that the kitten knows its enemy before it is able to see him, and when its fear can in no way serve it. One day, after fondling my dog, I put my hand into a basket containing four blind kittens three days old. The smell that my hand carried with it set them puffing and spitting in a most comical fashion/' Experiments with young birds hatched from artificially in- cubated eggs and kept away from all contact with their kind show conclusively that certain capacities are truly part of the inheritance, and require no experience or suggestion, while others not more complex require to be learnt. Thus the power j of uttering the characteristic call-note is inborn, but chicks require to learn what is good for eating and what is deleterious. Thus the power of executing the proper swimming and diving movements is inherited, but chicks do not instinctively know that water is drinkable. It is one of the problems of inheritance to distinguish between inborn capacities and those which require education. An even more difficult problem, which Prof. Pearson has successfully tackled by an ingenious indirect method, relates to the inheritance of man's mental and moral qualities. Though very plastic, there is no doubt that they are inherited in rudi- ment, just like physical characters. Just as the Romans dis- tinguished physically the long-nosed Nasones, the thick-lipped Labeones, the swollen-cheeked Buccones, and the big-headed Capitones, so, as Voltaire points out, " the Appii were ever proud and inflexible, and the Catos always austere." The literature of inheritance is crowded with examples of the transmissibility of what we cannot but call trivial peculiarities, though the probability is that they are often the correlates of what is important. A few illustrations may be selected : " A gentleman had a peculiar formation of the right eyebrow. 22 HEREDITY AND INHERITANCE It was strongly arched, and some of the hairs in the centre grew upwards. Three of his sons have the same peculiarity ; one of his grandsons has it also ; so has his great-granddaughter, and, if we are to believe the artists, this gentleman's grandfather and great-grandfather had the same peculiarity " (R. W. Felkin). " There was a family in France, of whom the leading repre- sentative could when a youth pitch several books from his head by the movement of the scalp alone. His father, uncle, grand- father, and his three children possessed the same power to the same unusual degree. This family became divided eight genera- tions ago into two branches, so that the head of the above- named branch was cousin in the seventh degree to the head of the other branch. This distant cousin resided in another part of France, and on being asked whether he possessed the same faculty, immediately exhibited his power." A woman with blonde hair, a birth-mark under the left eye, and a lisp, married a man with dark hair and normal utterance. There were nineteen children, none of whom showed any of the mother's characters. Nor among the numerous grandchildren was there any trace. In the third generation, however, there was a girl with blonde hair, a mark below the left eye, and a lisp. Girou tells of a man who had the peculiar habit of always sleeping on his back with his right leg crossed over his left. His daughter showed the same habit almost from infancy, and per- sisted in it in spite of efforts made to make her sleep in an ortho- dox position. Darwin gives an even better case where a very peculiar gesture reappeared ; and there seems no doubt that trivial peculiarities, e.g. playing with a lock of hair and idio- syncrasies of handwriting, may reappear even in cases where imitation was out of the question (Biichner, 1882, p. 42). And thus the list may be followed till we end with evidence of the inheritance of minutiae often of a most trivial character. Thus : " Schook relates the case of a family nearly all the mem- DENIALS OF INHERITANCE 23 bers of which could not endure the smell of cheese, and some of them were thrown into convulsions by it J> (R. W. Felkin). Here again we are forced back to the general thesis that the germinal organisation is a coherent individualised unity, which may find similar expression in the most detailed peculiarities of the body. § 6. Denials of Inheritance The resemblance between offspring and their parents, both in general and in particular, as to abnormal as well as normal characteristics, cannot be denied as a fact, but it has often been denied as the result of transmission. Although the denials, which have varied greatly in degree and motive, are for the most part due to misunderstanding, they may deserve brief consideration, since even to-day we sometimes hear cultured men declaring that " they do not believe in heredity." The extreme position may be represented by Wollaston, a scientific philosopher of the end of the eighteenth century, who sought to conserve the integrity and sanctity of the human spirit by altogether denying transmission. Each new life was to his mind a fresh start, unrelated in any real sense to parents or ancestors. The speculative naturalist Bonnet and many others admitted the inheritance of generic and specific characters, but denied that of individual characteristics. Buckle is the most illustrious example of those who, while admitting the inheritance of bodily characters, firmly deny that the same is true in regard to the mind. Buckle maintained that the ordinary method of demonstrating the inheritance of talents by collecting examples of similar mental peculiarities in father and son is in the highest degree illogical ; it neglects, for instance, the frequency of coincidence, and yet more the results of similar upbringing and environment. 24 HEREDITY AND INHERITANCE A consideration of these denials, which have ceased to appeal to many, may be of use as affording opportunity for emphasising two facts. i. Reappearance of a character from generation to generation does not of itself prove the inheritance of that character, if it be originally interpretable as the result of nurture (influences of activity and surroundings operative on the body), and if there be from generation to generation a persistence of the conditions which were originally instrumental in evoking the character. It is plain that the reappearance may be the result of similar effects hammered on each successive generation. Alpine plants brought to a lowland garden have been known to become much changed, and their descendants likewise. But there is good reason to believe, as we shall afterwards see, that the novel conditions directly impressed their effects on each successive crop. What impressed Buckle was the power of the environment in the widest sense ; it holds the organism in its grip, and hammers it into shape. This no one will gainsay, but we know that similar nurture has different results on different natures ; the duckling is not known to be less a duckling because hatched and brought up by a hen. Moreover, we know of the reappear- ance from generation to generation of many characteristics which cannot be interpreted as due to nurture — which often emerge, indeed, in the very teeth of nurture. At the same time, it is of great importance to bear in mind that an organism cannot be separated from its environment except at the risk of some fallacy. We may say that along with the organic heritage contained in the germ-cells every organism has what may be called an external heritage of appropriate environmental influences, which supply the stimuli for normal development. Appropriate food is part of the normal environment, and the supply of oxygen and water may be grouped in the same set ; other factors, like the osmotic pressure or the presence of calcium salts in DENIALS OF INHERITANCE 25 the water, are conditions of embryonic coherence ; others, like light and heat, serve to accelerate or to inhibit. It seems, also, that par- ticular combinations of factors are required as the " liberating stimuli " of particular characters in the developing organism. De- velopment is the expression of the inheritance, and the fullness of the expression depends on there being a normal environment. What is called a hereditary defect may be simply a defect in expression due to inadequate environment. How fundamental the germinal nature is may be realised if we think of Heape's experiment of transferring the fertilised ovum of a long- haired white angora rabbit into another variety of rabbit — a short- coated gray Belgian hare. The young were not less long-haired or less white because of the transplantation of the ova. Similarly Castle and Phillips removed the ovaries from a white albino guinea- pig, inserted those of a young black individual, and had the grafted animal mated with a male albino. Normal albinos mated together always have albino young, but the animal experimented on had to the albino male three litters (six young) all black. The foster-body did not count. 2. Beneath the misunderstanding which has led some to deny the facts of inheritance there is, as we have seen, a reasonable though exaggerated recognition of the potency of similar function and environment in producing resemblance ; and there is, per- haps, the recognition of another fact — that of variation. For several reasons — for instance, because the new life usually springs from a fertilised ovum which combines maternal and paternal contributions — the child is never quite like its parents. In other words, we suppose that the germinal material from which a child develops is not quite the same as that from which the parents developed, or not quite the same as that from which its brothers and sisters developed, and the result is variation in the true sense. Each offspring has its individuality and is a new creatidn. Even within a family no two are alike, especially to the care- ful parent's eye, though the impartial onlooker may be struck by the monotony. On the one hand, " Alle Gestalten sind dhnlich " ; on the other, " Keine gleichet der andern." CHAPTER IT THE PHYSICAL BASIS OF INHERITANCE " Gebt mir Materie, und ich will daraus eine Welt schaffen." — KANT. " We may regard the nucleus of the cell as the principal organ of inheritance " (a prophecy proved true). — HAECKEL, Generelle Morpho- logie, 1866, vol. i. p. 288. " The cell is not only the seat of vital activity, but is also the vehicle of hereditary transmission ; and the life of successive generations of living beings shows no breach of continuity, but forms a continuous vital stream in which, as Virchow said, rules an ' eternal law of continuity.' " — WILSON, 1900, p. 76. § i. What is true in the Great Majority of Cases. § 2. Diverse Modes of Reproduction. § 3. The Hereditary Relation in Unicellular Organisms. § 4. The Hereditary Relation in the Asexual Multiplica- tion of Multicellular Organisms. § 5. Nature and Origin of the Germ-cells. § 6. Maturation of the Germ-cells. § 7. Amphimixis and the Dual Nature of Inheritance in Sexual Reproduction. § 8. Inheritance in Parthenogenesis. § 9. Wherein the Physical Basis precisely consists. § i. What is true in the Great Majority of Cases The Inheritance is usually carried by the Germ-cells.— What was for so long quite hidden from inquiring minds, or but dimly discerned by a few, is now one of the most marvellous of biological commonplaces — that the individual life of the great majority of plants and animals begins in the union of two minute elements — the sperm-cell and the egg-cell. These microscopic individualities unite to form a new individuality, a potential offspring, which will by-and-by develop into a creature like to, and yet different from its parents. If we mean by inheritance 26 INHERITANCE CARRIED BY GERM-CELLS 27 to include all that the living creature is or has to start with in virtue of its genetic relation to its parents and ancestors, then it is plain that the physical basis of inheritance is in the fertilised ovum. The fertilised egg-cell is the inheritance, and at the same time the potential inheritor. What might be compared to an inheritance of property as apart from the organism itself is the store of food which may be inside the egg, or round about it. To the general fact stated in the preceding paragraph, a few exceptions must be made — e.g. for bananas which have no longer any seeds, for potatoes which are multiplied by cutting, for the drone-bees and summer green-flies who have mothers but no fathers, and for simple unicellular organisms in which there is no sexual reproduction ; but the exceptions are trivial compared with the vast majority of living creatures, in regard to which it is certain that each life begins in a fertilised egg-cell. An organic inheritance means so much, even when we use the comfortable word potentiality, that, although we are quite sure that the germ-cells constitute the physical basis of inheritance, we may consider for a moment the difficulty which rises in the minds of many when they are told that the egg-cell is often microscopic, and the sperm-cell often only j^^^^th of the ovum's size. Can there be room, so to speak, in these minute elements for the complexity of organisation supposed to be requisite ? And the difficulty will be increased if the current opinion be accepted that only the nuclei within these minute germ-cells are the true bearers of the hereditary qualities. Darwin spoke of the pinhead-like brain of the ant as the most marvellous little piece of matter in the world, but must we not rank as a greater marvel the microscopic germ-cells which contained potentially all the inherited qualities of that ant ? From one microscopic egg of a sea-urchin cut into three, Delage reared three larvae. In another case he reared an embryo from ^Tth of an egg. Twin animals are often developed from one egg. Wilson obtained quadruplets by shaking apart the 28 THE PHYSICAL BASIS OF INHERITANCE four-cell stage in the development of the lancelet. Marchal describes a " legion of embryos " developing from a single ovum of a peculiar Hymenopterous insect Encyrtus. In development, indeed, a half may be as good as a whole. In reference to the difficulty raised in some minds by the minuteness of the physical basis, it may be recalled that the students of physics, who make theories regarding the sizes of the atoms and molecules which they have invented, tell us that the image of an ocean liner filled with framework as intricate as that of the daintiest watches does not exaggerate the possi- bilities of molecular complexity in a spermatozoon, whose actual size is usually very much less than the smallest dot on the watch's face. Secondly, as we learn from embryology that one step conditions the next, and that one structure grows out of another, there is no need to think of the microscopic germ-cells as stocked with more than initiatives. Thirdly, we must re- member that every development implies an interaction between the growing organism and a complex environment without which the inheritance would remain unexpressed, and that the full- grown organism includes much that was not inherited at all, but has been acquired as the result of nurture or external influence. The fact is that size does not count for much in these matters, and the difficulty that some beginners feel in believing that the inheritance of the whale is packed into a pinhead-like egg is mainly due to ignorance of what may be called the fine com- plexity, or from another point of view the " coarse-grainedness," which must form part of our conception of every speck of matter. Nowhere more than in biology are we made to feel that " a little may go a long way." It should be noted that the degree of visible complexity, even in the microscopic nucleus of a germ-cell, is often very consider- able. Thus Eisen observed in the nucleus of a species of sala- mander twelve chromosomes, each of six parts, and in each part six granules — altogether 432 visible units. DIVERSE MODES OF REPRODUCTION 29 § 2. Diverse Modes of Reproduction In the preceding paragraph we have given prominence to what is true of the great majority of living creatures, — that a new life begins as a fertilised egg-cell. It is necessary, however, to refer to the other ways in which a new organism may arise, for some of them help us to understand what the hereditary relation means. The following scheme will probably serve to recall the familiar facts : Multiplication tlari *• \ , • 77 / I By division into two. /* unicellular I By budding> a modjfied form of divisien. I By sporulation, or division into many units. The reproduction may be wholly asexual : (i) in the sense that there is nothing corresponding to fertilisation or amphimixis ; and (2) in the sense that there are no special germ-cells. But in many unicellular organisms there are tlaborate processes of am- phimixis, and in colonial forms, like Volvox, there is a definite beginning of egg-cells and sperm-cells. Among the parasitic Sporozoa or Gregarines in the wide sense there is also a close approximation to the mode of sexual reproduction seen in most multicellular organisms. No hard-and-fast line can be drawn. C I. Without special germ-cells — e.g. by division In multicellnlar] of the body, by giving off buds (and as the organisms. | result of artificial cutting). III. With special germ-cells : 00 Eggs from one parent are fertilised by sperms from another parent — heterogamy, the commonest mode ; (£) Eggs from one parent are fertilised by sperms from the same (hermaphrodite) parent — autogamy, a very rare mode. (0 Eggs may develop without fertilisation — parthenogenesis. [A multicellular organism may also multiply by spore- cells — specialised germ-cells, yet hardly equivalent to eggs — which do not require fertilisation.] * * If we lay emphasis on the presence or absence of special reproductive elements, the classifi- cation of the modes of multiplication would read as follows : I. Without special repro- / Division, budding, etc., in most unicellulars. ductive elements. \Division, budding, etc., in some multicellulars. {More or less distinct specialisation of reproductive elements in some unicellulars. Specialised ova and spermatozoa in most multicellulars. Formation of spore-cells in some multicellulars. If we lay emphasis on the occurrence or non-occurrence of amphimixis (= fication of the modes of reproduction would read as follows : fertilisation) the classi- I. Without any form amphimixis. II. With some form amphimixis. of Without special reproductive-cells: (a) division, budding, etc., in many unicellulars ; and (b) division, budding, etc., in some multicellulars. With special reproductive-cells '. (a) formation of spores in some multicellulars ; (b) parthenogenetic ova. Without specialised reproductive elements, amphimixis occurs in most unicellulars. • With specialised reproductive elements, amphimixis occurs in a few unicellulars and in most multicellulars. 30 THE PHYSICAL BASIS OF INHERITANCE The reasons for lingering over the modes of reproduction — which it is confessedly difficult to arrange in a perfectly clear scheme — are (i) that our general view of the hereditary relation must be one which is applicable to all cases and not merely to the most frequent, and (2) that some of the simplest cases shed light upon the more complex. It is also important that we should make clear that the common phrases, " asexual repro- duction " and " sexual reproduction," are somewhat ambiguous, since attention has to be directed to two distinct points — (a) whether there are specialised reproductive elements, and (b) whether there is any form of amphimixis. § 3. The Hereditary Relation in Unicellular Organisms At what is called " the limit of growth," when the cell has attained to as much volume as its surface can adequately supply with food and oxygen, and so on, a unicellular organism normally divides into two, obviating the difficulties which would ensue if volume increased out of proportion to surface. The halves separate and grow. Two more or less exact replicas of the original unit result. It has been demonstrated that the division is often preceded by that intricate and orderly process of nuclear division, known as karyokinesis, which results in an equal partition of the nuclear constituents between the two daughter- cells. As each of the halves is in the strictest sense half of the organisation of the parent unit, we are not surprised that each should in appropriate environment grow into an almost exact image of the original whole. In most cases we have no methods subtle enough to detect any difference. There is complete here- ditary resemblance, and it would be puzzling if it were otherwise. Even when the unit divides into many units (as in spore-forma- tion), there is no puzzle in the fact that each reproduces the likeness of the original whole, except the puzzle of growth — of HEREDITARY RELATION IN UNICELLULAR^ 31 life, which is at present insoluble. Analogies may be found in methods of treating chemical molecules so that one gets at the end of the operation twice as many molecules as one had to start with ; or in the multiplication of crystals by breaking them into fragments and placing them in solutions of the same substance ; but, at the present time, these analogies are of no particular service, since we do not understand the nature of living matter. That a fragment of a unicellular' s organisation may, in an appropriate environment, reproduce an apparently perfect replica of the original unit, is not in any way explained by pointing out that there may be reproduction of like by like in the case of crystals or chemical molecules. FIG. 2. — Diagram of cell division (after Boveri). ckr. chromosomes, forming an equatorial plate; cs. centrosome. In slightly more complex cases there is a difference between tr^e two units into which the unicellular organism divides. Thus, in the oblique division of the slipper animalcule (Paramcg- cium), the one half goes off with the " mouth," the other has none. In a short time, however, the mouthless half forms a " mouth," and each half grows into a replica of the original. But as the organisation of each half is essentially the same as, and directly continuous with the organisation of the original cell, the development of the halves into similar wholes presents no special difficulty. Similar organisation and similar surround- ings yield similar results. That an injured infusorian should by re-growth repair its loss is an analogous phenomenon. Thus 32 THE PHYSICAL BASIS OF INHERITANCE we are led to see the force of Haeckel's definition of reproduction as discontinuous growth. But in many unicellular elements, what is liberated to begin a new life is not a half of the original nor anything like it, but a minute unit often called a " spore." It also grows into a com- plete reproduction of the original. In such cases, we again try to make the matter more intelligible, by saying that each spore is a representative fragment of the organisation of the original unit, and will therefore, in appropriate surroundings, grow and differentiate as the original did. Exactly the same often occurs when the unicellular organism is artificially divided into several parts ; and the results of these microscopic vivisection experi- ments, to which no one can on any grounds object, show that, if the excised fragment is to survive and develop, it must have a portion of the nuclear substance as well as of the general cell-substance. Without the nuclear constituent it may live for a time, as in Stentor, moving and responding to stimuli, but it cannot assimilate. Therefore, if we are asked what we mean by " organisation," we may say, at this stage, a certain protoplasmic architecture which implies essential relations between nucleoplasm and cytoplasm. The protoplasmic unit is like a firm with many partners of different kinds, each kind having many representatives ; and the retention of vitality, the possibility of regeneration on the part of the fragments, has this for its essential condition, that the integrity of the firm — in which lies its secret — is maintained by each fragment having at least one representative of the different kinds of partners. The reader who is not familiar with the subject should linger over the fact that a fragment or a minute spore, separated from a unicellular organism, may grow into (literally, reproduce) a unit, which to our senses is exactly like the original. This is (within the limits of our senses) complete hereditary resemblance, and we interpret it as due to the fact that the fragment or spore has to start with the essential organisation of the original. This MULTIPLICATION OF UN1CELLULARS 33 is, without complications, the fundamental fact in regard to inheritance. It should also be borne in mind that many of the unicellular organisms (Protozoa, at- the base of the animal series ; Proto- phyta, at the base of the plant series) are highly differentiated — i.e. with great complexity of structure even within the narrow limits of size (where a diameter of T^th of an inch is considered large) — and that many have very definite and interesting modes Gr FIG. 3. — Diagram of cell structure. (After Wilson.) PI. Plastids in cytoplasm or cell-substance ; cc. centrosome ; n. nudeolus ; Chr . chromo- somes ; N. nucleus ; ct. general cytoplasm ; V. vacuole ; Gr. granules. of behaviour, such as swimming in a spiral, seeking light or avoiding it, approaching certain substances and retreating from others, trying one kind of behaviour after another, — functional peculiarities — some of which cannot be described without using psychical terms — which are also included in the inheritance. The case of a fragment of crystal growing into a complete crystal is interesting enough, but that a fragment or spore of apparent simplicity should reproduce the obvious complexity of the unit from which it was separated is relatively more mar- vellous. 34 THE PHYSICAL BASIS OF INHERITANCE A note is needed in regard to the misunderstanding which has led many to cite cases of inheritance in unicellulars as relevant to the discussion on the transmission of " acquired characters." Although we can no longer say that unkellular organisms are without sexual reproduction, since many exhibit the liberation of special reproductive units and the occurrence of amphimixis, we may still say that, apart from transitional forms (like Volvox, which form colonies or " bodies " of one thousand to ten thousand cells), there is among the unicellulars only the beginning of the important distinction between somatic or bodily and germinal or reproductive material which distin- guishes multicellular organisms. This makes a notable differ- ence. § 4. The Hereditary Relation in the Asexual Multiplication of Multicellular Organisms In many of the simpler, but multicellular, plants and animals, a portion of the parent is separated off to form the beginning of a new life. The freshwater sponge multiplies in part by minute gemmules, which float away from the corpse of the parent and develop into new sponges ; many polypes produce buds which may be set adrift, as in the freshwater Hydra, or may remain attached and help to form the great colonies that we see in zoophytes and Anthozoa ; not a few worms also multiply by dividing or by budding, and the examples highest in the scale are found among the Tunicates, which are really vertebrate animals. Moreover, in some cases where asexual multiplication does not normally occur, it may still be a possibility, as is shown by the fact that cut-off portions may, in appropriate conditions, grow into entire individuals. Thus, two earthworms may occasionally be produced by cutting one ; a sponge whicn does not normally liberate buds may be cut into pieces and bedded out successfully j the arms of the starfish, whicn ASEXUAL MULTIPLICATION 35 the fisherman tears asunder, may give rise to several new in- dividuals. From nine excised fragments of a single Planarian worm, Voigt reared nine individuals (see Weismann, 1904, vol. ii. p. 25). Similarly, in regard to plants, many of the simpler multi- cellular forms produce detachable buds, familiar in the case of the liverworts ; and even in the flowering plants the same may occur, as in the bulbils of the tiger-lily. As in animals, great colonies may be formed, consisting of many individuals materially continuous, well seen in strawberries, whose creeping stems root here and there and give rise to independent plants. It is also a FIG. 4. — " Comet-form " of Starfish, showing how one arm regenerates the other four. (After Haeckel.) familiar fact that cut-off portions of a plant may readily give rise to entire individuals ; a little piece of moss, a Begonia leaf, a corner of a potato tuber — and hundreds of instances might be given — will suffice to start a new plant. In many ways the whole vegetable kingdom seems comparable to the sedentary sections of the class Ccelentera among animals (zoophytes, sea-anemones, corals, etc.), e.g. in the various forms of alternation of generations which occur, and in the readiness with which representative fragments will regrow the whole. This capacity of regenerating the whole from a small piece is the more striking when there is considerable differentiation of tissues and organs, as there is in flowering plants and the higher animals. The 36 THE PHYSICAL BASIS OF INHERITANCE fact being that the leaf of a plant, or a quarter of a zoophyte, or an eighth of a sea-anemone, may grow into an entire organism with reproductive cells, we must infer that the characteristic heritable material, usually segregated in the reproductive cells, is present in the cells of the body in these organisms. The feature common to the ordinary forms of asexual multi- plication is, that the reproduction is independent of eggs or sperms, or of any process comparable to fertilisation. What starts the new life, and iorms in this case the material basis of FIG. 5. — Asexual reproduction. A sea-worm (Syllis ramosa), in which budding has produced a branched temporary colony. (After Mclntosh.) inheritance, is a liberated portion of the parent. The heredity- relation is one of obvious material continuity. As regards inheritance, the feature characteristic of asexual multiplication is that the resemblance between parent and offspring tends to be complete. As Sedgwick (1899) expresses it: "The offspring do not merely present resemblances to the parent — they are identical with it ; and this fact does not appear to be astonishing when we consider the real nature of the process. Asexual reproduction consists in the separation of a portion of NATURE AND ORIGIN OF THE GERM-CELLS 37 the parent, which, like the parent, is endowed with the power of growth. In virtue of this property it will assume, if it does not already possess it, and if the conditions are approximately similar, the exact form of the parent. It is a portion of the parent ; it is endowed with the same property of growth ; the wonder would be if it assumed any other form than that of the parent." In asexual reproduction the resemblance of the offspring to the parent tends to be very complete, and the reason for like producing like is no puzzle, when the separated off-portion is a representative sample of the whole organism. § 5. N attire and Origin of the Germ-cells Re-statement of the Central Problem of Heredity. — The central problem of inheritance is to measure the resemblances and differences in the hereditary characters of successive generations, and to arrive, if possible, at formulae which will sum up the facts, such as Galton's Law of Ancestral Inheritance and Mendel's Law. The central problem of heredity is to form some con- ception of what is essential in the relation of genetic continuity, which binds generation to generation. Weismann's theory of the continuity of the germ-plasm is, in the first instance, a theory of heredity, and as important as Galton's law of inheritance. We know that almost every multicellular plant or animal has the beginning of its individual life in the union of two germ-cells (ovum and spermatozoon), and what must be found if the prob- lem of heredity is to be illumined at all is some reason why the germ-cells should have this power of developing, and of developing into organisms which are on the whole like the parents. In what respects are the germ-cells peculiar, and 38 THE PHYSICAL BASIS OF INHERITANCE different from the ordinary cells of the body ? Let us, then, concentrate our attention for a little on the nature and origin of the germ-cells. It is inexpedient to lay on the shoulders of the student of heredity the burden of problems which are not in any special sense his business. It is no doubt interesting to ask how an organisation, supposed to be very complex, may be imagined to find physical basis in a microscopic germ-cell, but the same sort of question may be raised in regard to a ganglion-cell. It is not distinctively a problem of heredity. It is interesting to inquire into the orderly and correlated succession of processes by which the fertilised egg-cell gives rise to an embryo, but this is the unsolved problem of physiological embryology. It raises questions distinct from those of heredity and inheritance, and apparently much less soluble. We shall return in the historical chapter to the various theories of heredity which have been suggested ; in the meantime, we require to refer to them only in dutline. The Typical Ovum. — The germ-cell produced by the maternal parent is usually a relatively large sphere of living matter (cyto- plasm), and various not-living included substances, such as nutritive yolk, pigment, oil-globules, and so forth. In the cytoplasm there lies a central kernel surrounded by a delicate membrane, the nucleus — a microcosm in itself. It contains a network or coil or some arrangement of delicate (linin) threads, carrying minute masses of a readily stainable material, the chromatin. Under high magnification the chromatin is seen to be built up of small corpuscles, sometimes like beads on a string, the microsomes. In certain phases of activity the chromatin forms a definite number of separate masses. They are then called chromosomes or idants, and the same number is always present in all the cells of the body of any particular species. In the nuclear sap which fills the nucleus there is often a rpunded body or vesicle — the nucleolus • or there may be ff-v- FIG. 6a. — Diagram of ovum, showing diffuse yolk-granules, g.v. germinal vesicle or nucleus ; chr. chromosomes. FIG. 66. — Diagram of body-cell, show- ing the nucleus with coil of chromatin filaments and the surrounding cyto- plasm. (After Carnoy.) [Facing p. 38. NATURE AND ORIGIN OF THE GERM-CELLS 39 several nucleoli. As they are very variable and often tran- sient, the nucleoli are not regarded as very important. Often they seem to be aggregations of reserve material or of waste- products. The Typical Spermatozoon. — The germ-cell produced by the FIG. 7. — Volvox globator, an Infusorian forming a colony of cells, showing the ordinary cells (c) that make up the colony or incipient " body " ; a and b, the special reproductive cells, both male and female — the beginning of the distinction between germ-cells and somatic cells. male parent, the spermatozoon, is very different from the ovum in appearance and structure, and is also very much smaller. When the egg is swollen with yolk, which does not count as living material, the spermatozoon may be less than a millionth of its volume. Most of the cytoplasm of the spermatozoon forms a locomotor flagellum or tail, often of intricate structure, which drives the " head" or nucleus before it, always working against 40 THE PHYSICAL BASIS OF INHERITANCE a current if there is one. It is obviously a specialised adaptation which helps the spermatozoon to find the ovum, and it may be absent In cases where no journey or search is required. The so-called head of the spermatozoon contains the stainable material or chromatin, and in many cases it has been shown that the ripe spermatozoon has the same number of chromosomes as the ripe ovum. At the junction of the " head " and the " tail " there is a short " middle piece " or " neck," in which there is often seen a minute " centrosome." There is in animals in most cases a great superficial contrast between the two kinds of germ-cells when fully mature. The typical ovum is relatively large, often laden with yolk, usually passive, and surrounded by some sort of membrane. The typical spermatozoon is relatively very minute, with no reserve material, and adapted to active locomotion. It is significant, however, that both contain the same number of chromosomes. Old Attempts to interpret the Uniqueness of the Germ- cells. — In the preformationist theories, which held sway in the seventeenth and eighteenth centuries — theories which asserted the pre-existence of the organism and all its parts, in miniature, within the germ — there was a kernel of truth well concealed within a thick husk of error. For we may still say, as the preformationists did, that the future organism is implicit in the germ, and that the germ contains not only the rudiment of the adult organism, but the potentiality of successive generations as well. But what baffled the earlier investigators was the question, How the germ-cell comes to have this ready-made organisation, this marvellous potentiality. Discovering no natural way of accounting for this, the majority fell back upon a hypothesis of hyperphysical agencies — that is to say, they abandoned the scientific method, and drew cheques upon that bank where credit is unlimited as long as credulity endures, THE THEORY OF PANGENESIS 41 An attempt to solve the difficulty which confronted the preformationists — the difficulty of accounting for the complex organisation presumed to exist in the germ-cell — is expressed in a theory which seems to have occurred at intervals in the long period between Democritus and Darwin, the theory of pangenesis. On this theory the cells of the body are supposed to give off characteristic and representative gemmules ; these are supposed to find their way to the reproductive elements, which thus come to contain, as it were, concentrated samples of the different components of the body, and are therefore able to develop into FIG. 8. — Forms of spermatozoa, enormously magnified, not drawn to scale. i and 2, Immature and mature spermatozoa of snail ; 3, of bird ; 4, of man — h. head, m, middle portion, t, tail ; 5, of salamander, with vibratile fringe {/) ; 6. of Ascaris, slightly amoeboid, with cap (c) ; 7, of crayfish. an offspring like the parent. The theory is avowedly unverifiable in direct sense-experience, but the same may be said of many other hypotheses, and is not in itself a serious objection. It is more to the point to notice that it involves many hypotheses, some of them difficult to accept even provisionally. Galton long ago tried, by experiments on the transfusion of blood, to test one of these hypotheses, and found no confirmation. But it is still more to the point to notice that there is another theory of 42 THE PHYSICAL BASIS OF INHERITANCE heredity which is, on the whole, simpler — which seems, on the whole, to fit the facts better, for instance the fact that our experience does not warrant the conclusion that the modifica- tions or acquired characters of the body of the parent affect in any specific and representative way the inheritance of the offspring. The Idea of Germinal Continuity. — As is well known, the view which many, if not most, biologists now take of the uniqueness of the germ-cells is rather different from that of pangenesis. It is expressed in the phrase " germinal continuity," and has been independently suggested by several biologists, though Weismann has the credit of working it out into a theory. Let us state its purport. There is a sense, as Galton says, in which the child is as old as the parent, for when the parent's body is developing from the fertilised ovum, a residue of unaltered germinal material is kept apart to form the future reproductive cells, one of which may become the starting-point of a child. In many cases, scattered through the animal kingdom, from worms to fishes, the beginning of the lineage of germ-cells is demonstrable in very early stages before the differentiation of the body-cells has more than begun. In the development of the threadworm of the horse, according to Boveri, the very first cleavage divides the fertilised ovum into two cells, one of which is the ancestor of all the body- cells, and the other the ancestor of all the germ-cells. In other cases, particularly among plants, the segregation of germ-cells is not demonstrable until a relatively late stage. Weismann, generalising from cases where it seems to be visibly demonstrable, maintains that in all cases the germinal material which starts an offspring owes its virtue to being materially continuous with the germinal material from which the parent or parents arose. But it is not on a continuous lineage of recognisable germ-cells that Weismann insists, for this is often unrecognisable, but on the continuity of the germ-plasm — that is, of a specific substance of definite chemical and molecular structure which is the bearer THE IDEA OF GERMINAL CONTINUITY 43 of the hereditary qualities. In development a part of the germ- plasm, " contained in the parent egg- cell, is not used up in the construction of the body of the offspring, but is reserved un- changed for the formation of the germ-cells of the following generation." Thus the parent is rather the trustee of the germ-plasm than the producer of the child. In a new sense, the child is " a chip of the old block." As Sir Michael Foster put it, " The animal body is in reality a vehicle for ova ; and after the life of the parent has become potentially FIG. 9. — Diagram illustrating idea of germinal continuity. (After E. B. Wilson.) Gl, fertilised ovum dividing into lineage of body-cells (B) and lineage of germ-cells— the base line ; B', B", the bodies of two successive generations ; G1, G3, G3, G*, G5, the chain of germ-cells. renewed in the offspring, the body remains as a cast-off envelope whose future is but to die." To use another metaphor, the germ-plasm is the lighted torch handed on from one runner to another. " Et quasi cursores vital lampada tradunt." Early segregation of the germ-cells is in many cases an ob- servable fact — and doubtless the list of such cases will be added to ; but the conception of a germ-plasm is hypothetical, just as the conception of a specific living stuff or protoplasm is hypo- thetical. In the complex microcosm of the cell we cannot point to any one stuff and say, " This is protoplasm " ; it may well be 44 THE PHYSICAL BASIS OF INHERITANCE that vital activity depends upon several complex stuffs which, like the members of a carefully constituted firm, are character- istically powerful only in their inter-relations. In the same way, it must be clearly understood that we cannot demonstrate the germ-plasm, even if we may assume that it has its physical basis in the stainable nuclear bodies or chromosomes. The theory has to be judged, like all conceptual formulae, by its adequacy in fitting facts. Let us suppose that the fertilised ovum has certain qualities, a, b, c . . . x, y, z ; it divides and re-divides, and a body is built up ; the cells of this body exhibit division of labour and dif- ferentiation, losing their likeness to the ovum and to the first results of its cleavage. In some of the body-cells the qualities a, b, find predominant expression, in others the qualities y, z, and so on. But if, meanwhile, there be certain germ-cells which do not differentiate, which retain the qualities a, b, c . . . x, y, z, unaltered, which keep up, as one may say figuratively, " the protoplasmic tradition," these will be in a position by-and- bye to develop into an organism like that which bears them. Similar material to start with, similar conditions in which to develop — therefore, like tends to beget like. . May we think for a moment of a baker who has a very precious kind of leaven ; he uses much of this in baking a large loaf ; but he so arranges matters by a clever contrivance that part of the original leaven is always carried on unaltered, carefully preserved for the next baking. Nature is the baker, the loaf is a body, the leaven is the germ-plasm, and each baking is a generation. MATURATION OF THE GERM-CELLS 45 § 6. Maturation of the Germ-cells We have seen that the germ-cells owe their capacity of develop- ment to the fact that they are the unspecialised descendants of the parental fertilised ovum — the custodians of the characteristic germ-plasm. In some cases the lineage of germ-cells is from the first distinct and apart from the lineage of body-forming cells, and we argue from these clear cases of germinal con- tinuity to the more numerous and less obvious cases where the germ-cells are not recognisable as such until later stages in development. There is no need for our present purpose to follow the genera- tions of the germ-cells within the body, or to trace the stages of growth and differentiation between primitive germ-cells and the fully formed ripe ova and spermatozoa. It is necessary, however, to allude to the process of maturation, which has a direct bearing on the problems of heredity and inheritance. Maturation. — i. It is an elementary fact of histology that the nucleus of each cell in the body of an organism contains a number of readily stainable bodies or chromosomes. In many cases it has been possible to count these, and it has been found that (with a few explicable exceptions) the number is constant for each species. As Prof. E. B. Wilson says (1900, p. 67) : " The remarkable fact has now been established with high probability that every species of plant or animal has a fixed and characteristic number of chromosomes, which regularly recurs in the division of all of its cells, and in all forms arising by sexual reproduction the number is even.* Thus, in some of the sharks the number is 36 ; in certain Gasteropods it is 32 ; in the mouse, the salamander, the trout, the lily, 24 ; in the worm Sagitta. 18 ; in the ox, guinea- * In a few insects the females have in their body-cells one chromo- some in addition to the number possessed by the males, 46 THE PHYSICAL BASIS OF INHERITANCE pig, and in man * the number is said to be 16, and the same number is characteristic of the onion. In the grasshopper it is 12 ; in the hepatic Pallavicinia and some of the nematodes, 8 ; and in Ascaris, another thread- worm, 4 or 2. In the crustacean Artemia it is 168. Under certain circumstances, it is true, the number of chromosomes may be less than the normal in a given species ; but these variations are only apparent exceptions [p. 87, Wilson]. The even number of chromosomes is a most interesting fact, which, as will appear hereafter [p. 205, Wilson], is due to the derivation of one-half the number from each of the parents." 2. About 1883, Van Beneden made the important discovery that the nuclei of the ovum and of the spermatozoon which unite in fertilisation contain each one-half of the number of chromosomes characteristic of the body-cells. This has been confirmed in regard to so many plants and animals that it may now be regarded as a general fact. The student should refer to the partial list given by Wilson (1900, pp. 206-7), where it will be seen that if the somatic nuclei have 12, 16, 18, or 24 chromosomes, the germ-nuclei have 6, 8, 9, or 12 respectively. A striking case is found in the large thread-worm (Ascaris megalo* cephala) of the horse, which occurs in two varieties, — the one, var. univalens, with two chromosomes in its body-cells has one chromosome in its germ-nuclei ; the other, var. bivalens, with four chromosomes in its body-cells, has two chromosomes in its germ-nuclei. 3. If each of the nuclei which unite in fertilisation has only half as many chromosomes as are characteristic of the species, it follows that a reduction of the number must take place in the history of the germ-cells, and this is the outstanding fact in the process of maturation. Alike in the history of the egg (oogenesis) and in the history of the sperm (spermatogenesis), * " Flemming believed the number in man to be considerably greater than 1 6." It is now generally stated to be 24. MATURATION OF THE GERM-CELLS 47 there is a parallel reduction in the number of chromosomes to one-half. P.G.C. OG A .'\ A A l\ IV- L Oc» P.G.C. SG*' \\ I sc P.B fiG > \ /v.: P.L' FIG. 10. — Parallelism of oogenesis (A) and spermatogenesis (after Boveri). P.G.C. in both series (A and B), one of the primitive germ cells. Following the oogenesis (A), there is first of all a period of multiplication (A/), included within the first bracket. The primitive germ -eel 1 gives rise to oogonia (OG). These oogonia give rise to oocytes (OC). Then follows a period of growth (G), included within the second bracket. Then follows the process of ripening or maturatio'n (R), included within the third bracket. Od, the immature ovum, with the normal number of chromcsomes. P.B', the first polar body, usually separated off by a meiotic or reducing division which lessens the number of chromosomes to one half the normal. Ob, the ovum after giving off the first polar body, with half the normal number of chromo- somes. P.B", the second polar body, formed by an ordinary equation division. Oc, the ripe ovum. P.B', the first polar body has divided into two by an equation division. Following the spermatogenesis (B), there are successive periods (or zones in the testis) of multiplication (M), growth (G), and reduction (R). The primitive germ-cell gives rise to spermatogonia (SG). These spermatogonia give rise to spermatocytes (SC). Immature spermatocytes of the first order (Sa) have the normal number of chromosomes. In many cases by a reduction or meiotic division they give rise to spermatocytes of the second order (Sb). with half the normal number of chromcsomes. These give rise by an equation division to spermatozoa (Sc). "The one fact of maturation that stands out with perfect clearness and certainty amid all the controversies surrounding it is a reduction of the number of chromosomes in the ultimate germ- cells to one-half the number characteristic of the somatic cells. It is equally clear that this reduction is a preparation of the germ- 48 THE PHYSICAL BASIS OF INHERITANCE cells for their subsequent union and a means by which the number of chromosomes is held constant in the species. With a few exceptions the first indication of the numerical reduction appears through the segmentation of the spireme-thread, or the resolution of the nuclear reticulum, into a number of masses one-half that of the somatic chromosomes. In nearly all higher animals this process first takes place two cell-generations before the formation of the definitive germ-cells, and the process of reduction is completed by two rapidly succeeding ' maturation- divisions/ giving rise to four cells, all of which become functional in the male, while in the female only one becomes the egg, and the other three — the polar bodies or their analogues — are cast aside. During these two divisions each of the original chromatin masses gives rise to four chromosomes, of which each of the four daughter-cells receives one ; hence, each of the latter receives one-half the somatic number of chromosomes. In the higher plants, however, the two maturation-divisions are fol- lowed by a number of others, in which the reduced number of chromosomes persists, a process most strikingly shown in the pteridophytes, where a separate sexual generation (prothallium) thus arises, all the cells of which show the reduced number " (Wilson, 1900, p. 285). The asexual spore-bearing fern-plant has in its cells twice as many chromosomes (2 ri) as the sexual prothallus has (ri). The spores produced by the fern-plant have n chromosomes ; they develop into a prothallus with n chromosomes ; the prothallus produces sex-cells with n chromosomes ; these undergo no reduction and by their union they restore the number 2 n, which characterises the resulting embryo and the subsequent fern-plant. As Boveri has said : " Thus at some stage or other in the gene- ration-series of the germ-cell there occurs a reduction of the number of chromosomes originally present to one-half, and this numerical reduction is therefore to be regarded, not as a mere theo- retical postulate, but as a fact " (Zellen-Studien, iii. 1890, p. 62). o 6 r~\ < O O o o n oo go L } ^J z: a: o o o o a_ •z. cr G U UJ -« ss QZ hJ cc O O O i t If 5SII-58 • iil!!! * i IKI i Hllll i = -ss. I - 5l^§E i 6 S^-S|t^ 3 5-6B-Ss-S I sllll AMPHIMIXIS 49 § 7. Amphimixis and the Dual Nature of Inheritance in Sexual Reproduction Apart from exceptional cases, the inheritance of a multi- cellular animal or plant is dual — part of it comes from the mother and part of it from- the father ; in other words, the material basis of inheritance is a fertilised egg-cell. The new individuality has its origin in the fusion of two potential individuals, for as such the ovum and spermatozoon must be regarded. The exceptions referred to are cases of asexual multiplication by buds or otherwise, as in the freshwater Hydra ; cases of partheno- genesis, as in the case of the unfertilised eggs which develop FIG. 12. — Fertilised ovnm of Ascaris. (After Boveri.) chr. chromosomes, two from ovum-nucleus and two from sperm-nucleus ; cs. centrosome, from which " archoplasmic " threads radiate, partly to the chromosomes. into green flies (Aphides) in the summer ; and cases like liver- flukes, where an animal is both mother and father to its offspring. Apart from these exceptions the inheritance does at the start consist of maternal and paternal contributions in intimate and orderly union. When a spermatozoon, outstripping its fellows (for there are usually very large numbers), reaches an ovum and bores its way into it, the cytoplasmic flagellum is left behind, having performed its function, and the sperm-nucleus and the ovum-nucleus move towards one another. By a rapid change in the periphery of the ovum, the enveloping membrane becomes firmer, and the ovum becomes non-receptive to other spermatozoa. When 4 50 THE PHYSICAL BASIS OF INHERITANCE several effect entrance at once, abnormalities usually result. In the mature ovum there is no centrosome ; if it was originally present, it disappears. The spermatozoon, however, intro- duces, along with its nucleus, its centrosome, and this divides into two. The two centrosomes appear to take an active part in the approximation and intimate apposition of the maternal and paternal chromosomes, and in their subsequent partition between the first two daughter-cells. Prof. E. B. Wilson states the general opinion of experts some- what as follows. As the ovum is much the larger, it is believed to furnish the initial capital — including, it may be, a legacy of food-yolk — for the early development of the embryo. From both parents alike comes the inherited organisation which has its seat (according to most biologists) in the readily stainable (chromatin) rods of the nuclei. From the father comes a little body (the centrosome) which organises the machinery of division by which the egg splits up, and distributes the dual inheritance equally between the daughter- cells. Let us now proceed to expound four important theorems. i. In Ordinary Sexual Reproduction the Inheritance is Yery precisely Dual or Biparental. — Recent discoveries have shown that the paternal and maternal contributions which come together in fertilisation are, for several divisions at least, exactly divided among the daughter-cells, thus confirming a prophecy which Huxley made in 1878 : " It is conceivable, and indeed probable, that every part of the adult contains molecules derived both from the male and from the female parent ; and that, regarded as a mass of molecules, the entire organism may be compared to a web of which the warp is derived from the female and the woof from the male." " What has since been gained," Prof. Wilson says, "is the knowledge that this web is to be sought in the chromatic substance of the nuclei, and that the centrosome is the weaver at the loom." Alter the paternal and maternal chromosomes have united, 0 0 sf> FIG. 13. — Diagram of maturation and fertilisation and first stages oi cleavage. (From Prof. H. E. Ziegler, with his kind permission.) The colours have been added. i. — The immature ovum, with four double chromosomes, longitudinally cleft ; c, centro- some ; ch, chromosomes ; nl, nucleolus. 2. — First maturation division ; the nnclear spindle has at its equator four groups of tetrads, three of which are visible. 3 and 4.— Formation of the first polar body (P.B.). In fig. 4 a spermatozoon (sp) is entering. The paternal chromatin is shown throughout in red, the maternal in blue, th< centrosome which is brought in by the spermatozoon is shown in yellow. The ovum-centro- some disappears. 5.— The formation of the second polar body and the division of the first (i P.B.). The head of the spermatozoon has formed the male pronudeus (sp). The centrosome introduced by the spermatozoon is surrounded by a clear area and rays. 6. — The second polar body (2 P.B.) has been set adrift. The first has divided into two. The three polar bodies and the now mature ovum have in their nuclei half the normal number of chromosomes. Thus four are seen in the female pronudeus (f.pn). The centrosome has divided into two. 7. — The male and female pronuclci (sp and f.pn) have become like one another, and are near together. The centrosomes (c) have become the centres of two large systems of rays. • 8. — The two pronudei are in contact and are coalescing. 9. — The nudei have lost their membrane, and the first segmentation-spindle or deavage- spindle has been formed, a centrosome lying at each pole. The spindle has the normal number of chromosomes, but each has divided, so that eight pairs are present. 10. — The egg-cell is dividing. The chromosomes are separated into two groups, each group with eight chromosomes. The centrosome at each pole has divided into two. n. — The division or cleavage is complete. The rays have disappeared. The chromosomes are represented by minute vesides or karyomeres. 12. — The new nudd have been constituted by union of the vesicles. The centrosomes lie closely apposed, but will occupy the poles of the spindle at the next division. Facing p. 51. ol ,«smoeomoi/J > sldirab i.'io! jii ?r,<; ^;:c irLJ \J FOUR IMPORTANT THEOREMS 51 but never fused, to form one nucleus — the segmentation-nucleus — the cleavage or segmentation of the fertilised ovum begins. There is a centrosome, derived from the sperm-centrosome, at each pole of the nucleus, and a system of fine rays radiates from each, some of these rays entering into close association with the chromosomes. Each chromosome is halved longitudinally, as a piece of stick might be split up the middle, and after a very complex routine the halves of each split chromosome migrate, either actively or passively, to opposite poles. Thus, near each centrosome there comes to be a group of chromosomes, half of each group being of paternal origin and half of maternal origin. Each group in an orderly fashion rounds itself off into a unified nucleus, the body of the cell (the cytoplasm) constricts across the equatorial plane, and two cells are formed. The gist and import of the whole process is the precisely equal partition of the maternal and paternal contributions, so that each of the daughter-cells has a nucleus half maternal and half paternal. For many successive divisions (e.g. in Cyclops) the duality has been demonstrated,* so that we may fairly say that the maternal and paternal contributions form the warp and woof of the growing organism. 2. Inheritance, though Dual, is strictly Multiple. — Although the whole inheritance which constitutes an offspring usually comes from two parents, and may therefore be called dual, it is obvious that the heritable material of each parent was also dual, being derived from the grandparents, and so on backwards ; so that inheritance is strictly not merely dual, but in an even deeper sense multiple. Amphimixis or fertilisation implies the subtle mingling of two. minute organisations so that they become physiologically one, but each of them was already * According to Haecker's careful observations on the water-flea Cyclops, the paternal and maternal contributions, i.e. chromosomes, are traceable as distinct individualised items throughout the whole of development. 52 'I HE PHYSICAL BASIS OF INHERITANCE the complex product of ancestral lineage. We shall return to the subject when we come to consider Gal ton's Law of Ancestral Inheritance. Though a comparison with the inheritance of property is apt to mislead, it may be of use to think for a moment of a youth inheriting an estate, of which one might accurately say that it had belonged in half t& his father and in half to his mother. Yet a genealogist with a full knowledge of the family might be able to go further back, and might show, with even greater accuracy, how this corner was due to a grandmother and that to a great-grandfather. This conception is so fundamentally important that I cannot refrain from quoting an illustration from Mr. Galton's Natural Inheritance, which puts the matter very clearly. " Many of the modern buildings in Italy are historically known to have been built out of the pillaged structures of older days. Here we may observe a column or a lintel serving the same purpose for a second time, and perhaps bearing an inscription that testifies to its origin ; while as to the other stones, though the mason may have chipped them here and there and altered their shape a little, few if any came direct from the quarry This simile gives a rude though true idea of the exact meaning of Particulate Inheritance — namely, that each piece of the new structure is derived from a corresponding piece of some older one, as a lintel was derived from a lintel, a column from a column, a piece of wall from a piece of wall. . . . We appear to be severally built up out of a host of minute particles of whose nature we know nothing, any one of which may be derived from any one progenitor, but which are usually transmitted in aggregates, con- siderable groups being derived from the same progenitor. It would seem that while the embryo is developing itself, the particles more or less qualified for each post wait, as it were, in competition to obtain it. Also that the particle that succeeds must owe its success partly to accident of position and partly to POUR IMPORTANT THEOREMS 53 being better qualified than any equally well-placed competitor to gain a lodgment. Thus the step-by-step development of the embryo cannot fail to be influenced by an incalculable number of small and mostly unknown circumstances." (Natural Inherit- ance, p. 9.) 3. Duality of Inheritance may be real, though it is not ex- pressed.— It must be carefully observed that the demonstration of the dual nature of inheritance afforded by the facts of amphi- mixis does not necessarily imply that the dual nature of the inheritance will be patent in the full-grown offspring. The offspring is often like both its parents, often particularly like one, often not very like either. The parent of children, the breeder of animals, or the cultivator of plants, has often occasion to remark in the offspring what looks like an entire absence of the characteristics of one of the parents. The foal may seem to take entirely after the sire, as if the maternal inheritance counted for nothing. It is likely that this so-called " exclusive " or " unilateral " inheritance is often more apparent than real, our observation being arrested and preoccupied by a few out- standing features. The certain fact that the resemblance, apparently absent, often reappears in the next generation, shows that the incompleteness was not in these cases in the inheritance, but simply in its expression. We shall return to this subject in connection with the different modes of inheritance. 4. Each Germ-cell has a Complete Equipment of Heredi- tary Qualities. — It is usually assumed that each of the two sex-cells which unite in fertilisation has in it the potentiality of an organism with a full equipment of the essential characters of the species ; but since the spermatozoon always dies unless it enters the ovum, it is difficult to give experimental proof of the assumption. Some recent daring experiments, which demand confirmation, are very suggestive in this connection. Prof. Yves Delage (1898) divided the minute egg of the sea- urchin under the microscope into two parts, one containing the 54 THE PHYSICAL BASIS OF INHERITANCE nucleus and its companion- body the centrosome, the other being necessarily simply half of the living matter of the egg without any nucleus. Beside them he placed an intact ovum, and then let the spermatozoa in. All the three objects showed equal " sexual attraction " in respect to the spermatozoa ; all three were fertilised ; all three segmented, the intact ovum most rapidly, the nucleated fragment more slowly, the non-nucleated fragment more slowly still. In one case the development proceeded for three days ; the intact ovum had become a typical gastrula (two-layered embryo), the nucleated fragment a smaller gastrula, and the non-nucleated fragment also a gastrula but with a very much reduced cavity. All the cells of these embryos showed nuclei. Thus the experimenter was led to the conclusion that fertilisation and some measure of development may occur in a fragment of ovum without nucleus or centrosome. The nucleus of the spermatozoon must have been in this case sufficient in itself, though it will be noticed that in the experiment cited the fragment did not develop far. Delage makes the important suggestion that in fertilisation two things must be distinguished : (a) the stimulus given to the ovum by some specially energetic substance brought in by the spermatozoon, perhaps in its centro- some ; and (6) the mingling of heritable characteristics, Weis- mann's " amphimixis." In subsequent experiments Prof. Delage (1899) reached even more extraordinary results. Non-nucleated fragments of the ovum of Echinus (sea-urchin), Dentalium (elephant's- tooth sheii), and Lanice conchilega (a seashore worm), were effectively fer- tilised and gave rise to the characteristic larval forms — pluteus, veliger, and trochophore respectively. Three larvae were reared from one ovum of a sea-urchin ; a normal blastula embryo (a hollow ball of cells) was reared from -frih of a sea-urchin ovum ; a non-nucleated fragment of a sea-urchin ovum, after fertilisation by a spermatozoon with nine chromo- somes (nuclear rods), gave rise to a larva whose cells had the FOUR IMPORTANT THEOREMS 55 normal number of eighteen chromosomes : such are some of the extraordinary results reached by this clever experimenter. It seems, then, as if fertilisation may, in many cases, be effective without there being any ovum-nucleus present, as if the essential fact were the union of a sperm with a mass of egg- cytoplasm. Delage's experiments cited above seem to prove that the nucleus and centrosome of the ovum are not essential to ferti- lisation. Professor Loeb (1899), of Chicago, has made experi- ments which seem to show that the spermatozoon may be dispensed with. In other words, he has been able to induce parthenogenetic development artificially in cases where it does not normally occur. He has been led to believe that the only reason why the eggs of many marine animals do not develop parthenogenetically is that something in the constitution of the sea-water prevents it. This something is the presence or absence of ions of sodium, calcium, potassium, and magnesium, the two former requiring to be reduced, the two latter to be increased. " The mixture of about 50 per cent, ^-n MgCl2 (magnesium chloride) with about 50 per cent, of sea-water was able to bring about the same effect as the entrance of a sperma- tozoon. The unfertilised eggs [of the sea-urchin Arbacia] were left in such a solution for about two hours. When brought back into normal sea-water they began to segment and form blastuke, gastrulae, and plutei, which were normal in every respect. The only difference was that fewer eggs developed, and that their development was slower than in the case of the normal develop- ment of fertilised eggs. With each experiment a series of control experiments was made to guard against the possible presence of spermatozoa in the sea-water. . . . From these experiments it follows that the unfertilised egg of the sea-urchin contains all the essential elements for the production of a perfect pluteus. The only reason that prevents the sea-urchin from developing par- thenogenetically under normal conditions is the constitution of 56 THE PHYSICAL BASIS OF INHERITANCE the sea-water. The latter either lacks the presence of a sufficient amount of the ions that are necessary for the mechanics of cell division (Mg, K, HO, or others), or it contains too large a quantity of ions that are unfavourable to this process (Ca, Na, or others), or both. All the spermatozoon needs to carry into the egg for the process of fertilisation are ions to supplement the lack of the one or counteract the effects of the other class of ions in the sea-water, or both. The spermatozoon may, however, carry in addition a number of enzymes or other material. The ions and not the nucleins in the spermatozoon are essential to the process of fertilisation." These remarkable experiments are confirmatory of the general assumption that spermatozoon and ovum are completely equipped potential organisms. Further confirmation may be found in cases of partial parthenogenesis — e.g. the development of drone-bees from unfertilised eggs ; from the close similarity in the history of ovum and spermatozoon respectively ; from the exactly equal way in which the paternal and maternal nuclear contributions are distributed to each cell, during the early stages of cleavage at least. Or take the simple experiment of crossing a black guinea-pig with a typical albino. All the offspring are black, although only one of the parents — it does not matter which — has the quality of blackness. It is evident that the germ-cells of either parent are able to carry a complete equipment of blackness. When we consider the ovum and spermatozoon as two fully equipped potential individualities which unite to form the beginning of a new individuality, we see more clearly how, on the one hand, there is a double likelihood of the essential specific characters being sustained, and how, on the other hand, there is every likelihood that the intermingling will lead indirectly, if not directly, to something new. INHERITANCE IN PARTHENOGENESIS 57 § 8. Inheritance in Cases of Parthenogenesis It would be interesting to know with precision what the facts of inheritance are in cases where development proceeds from an unfertilised ovum, particularly in those cases where the parthenogenesis continues uninterruptedly for many generations. On general grounds, from the absence of fertilisation, one would expect to find few new departures or progressive variations ; but rather, on the other hand, hints of degeneracy. The ob- served facts are still very few. Experiments which Prof. Weismann (1893, p. 344) made on a small crustacean (Cypris reptans) showed a very high degree of uniformity between parent and offspring, with occasional exceptions, which he regarded as exhibiting reversions to an ancestral form many generations removed. Dr. Warren's (1899) measurements of successive partheno- genetic generations of Daphnia magna also gave evidence of slight variability (i.e. of incompleteness of hereditary resem- blance). They seemed to favour the view that "inheritance in parthenogenetic generations resembles that from mid-grand- parent to grandchildren." § 9. Wherein the Physical Basis precisely consists The fertilised egg-cell divides into many cells ; these arrange themselves in various ways ; they grow and multiply ; they exhibit division of labour and the structural side of this — which we call differentiation ; they form tissues and organs ; they become integrated into a body ; they reproduce the likeness of the parental type with variations. Meanwhile, some of the cells remain apart from body-making or differentiation, and form the beginnings of the reproductive organs, whence their descendants — the mature germ-cells — are by-and-bye liberated to start another generation. That this next generation is also after the parental type is due to the continuous lineage of cells 58 THE PHYSICAL BASIS OF INHERITANCE containing unspecialised germinal material. In similar con- ditions similar material produces similar results. But, if this has become clear, we have now to inquire into the precise nature of the physical basis which conserves the heritable qualities. Is it the germ-cell as a whole that is essential, or is the cytoplasm most important, or is it the nucleus only ? Importance of the Chromosomes of the Germ-nuclei. — Many observa- tions go to show that the nucleus of a cell plays an important part in nutritive and constructive processes, and it is certain that a cell artificially bereft of its nucleus will soon die if left to itself. The nuclear material (karyoplasm or nucleoplasm) is an essential part of the vital organisation. The view has gained ground that the chromatin bodies or chromosomes are the chief, if not the exclusive, vehicles of the hereditary qualities. Let us consider some of the arguments in support of this view. 1. Argument from cell-division. — Roux, Hertwig, Kolliker, Stras- burger, and many others, have emphasised the fact that, in the ordinary (mitotic) form of cell-division, the chromatin or readily stainable material of the nucleus is divided " with the most scrupu- lous equality " to form the basis of the nuclei of the daughter-cells, while the cytoplasm or general cell-substance " undergoes on the whole a mass-division — a most remarkable contrast." As Prof. Wilson says (1900, p. 351) : " This holds true with such wonderful constancy throughout the series of living forms, from the lowest to the highest, that it must have a deep significance. And while we are not yet in a position to grasp its full meaning, this contrast [between nuclear and cytoplasmic behaviour in division] points unmistakably to the conclusion that the most essential material handed on by the mother-cell to its progeny is the chromatin, and that this substance, therefore, has a special significance in in- heritance." 2. Argument from maturation. — In the changes which lead up to the ripe egg and the fully-formed spermatozcon, there is, as we have seen, an elaborate preparation whereby the germ-nuclei which unite in fertilisation are rendered precisely equal as regards the number of their chromosomes. On the other hand, the cytoplasm of the relatively large, passive, often food-laden and ensheathed ripe ovum BEARERS OF THE HEREDITARY QUALITIES 59 is typically as different as possible from that of the very minute, actively mobile, usually short-lived spermatozoon. The constancy and frequent complexity of the reduction-processes which secure the equivalence of chromosomes suggest that these bodies are of para- mount importance in inheritance. 3. Argument from fertilisation. — In typical cases of fertilisa- tion in animals, and in many plants as well, a spermatozoon enters an ovum, sometimes a hundred thousand times larger \ FIG. 14. — The chromatin elements of the nuclei in coil (a), double star (b), and almost divided stages (c). (After Pfitzner.) than itself. As it enters it may leave behind it the locomotor " tail," which has discharged its function, thus further reducing its infinitely small stock of cytoplasmic material. The " head " of the spermatozoon, which is mostly nucleus, and the little " middle piece " which carries the centrosome, are apparently the important parts, and it is the ovum which furnishes the cytoplasmic basis of further operations. The very gist of fertilisation, so far as we can see it, is the intimate and orderly 60 THE PHYSICAL BASIS OF INHERITANCE combination of the paternal and maternal chromosomes to form one nucleus — the segmentation-nucleus. Moreover, the maternal and paternal contributions are, as we have noted, distributed with scrupulous equality, certainly to the first two cells of the embryo, and probably to all later-formed cells. " The latter conclusion, which long remained a mere surmise, has been rendered nearly a certainty by the remarkable ob- servations of Riickert, Zoja, and Haecker. We must, therefore, accept the high probability of the conclusion that the specific character of the cell is in the last analysis determined by that of the nucleus — that is, by the chromatin ; and that in the equal distribution of paternal and maternal chromatin to all the cells of the offspring, we find the physiological explanation of I I ft. w BEARERS Of THE 'HEREDITARY QUALITIES 61 V FIG. 15. — Diagram of the process of fertilisation in Ascaris. (After Boveri.) a, female pronucleus ; b, polar bodies ; c, sperm pronudeus ; d, sperm-cap ; ac, chromosomes of united female and male pronuclei (a and c) • e, centrosomes ; fine (archoplasmic) threads radiating from the centrosomes. I-V show union of paternal and maternal chromosomes ; VI shows equatorial plate of segmentation nucleus ; VII-X show the division into the two first cleavage-cells or blastomeres. 62 THE PHYSICAL BASIS 'OF INHERITANCE the fact that every part of the latter may show the character- istics of either or both parents " (Wilson, 1900, p. 352). 4. Argument from Boveri's ingenious experiment. — Taking a hint from the experiments of the brothers Hertwig, who showed that non- nucleated fragments of unfertilised sea-urchin ova (broken by shaking) might be successfully fertilised and might segment, Boveri (1889, 1895) showed that such fertilised fragments developed into dwarf, but normal, larvae. In these, as T. H. Morgan (1895) after- wards showed, the nuclei contain only half the normal number of chromosomes, having had only a sperm-nucleus to start with. Interesting as this was, Boveri's further experiment was yet more striking. He fertilised the enucleated egg-fragments of one species of sea-urchin (Splicer echinus granularis) with spermatozoa of another species (Echinus micyotuberculatus) , and obtained in a few cases dwarf larvae (plutei), which showed, except as regards size, the paternal characters only. Therefore he concluded that the nucleus is the exclusive bearer of the hereditary qualities, for it seemed from the experiment that the enucleated maternal cytoplasm had remained without specific influence. It is admitted by Boveri himself that further experiments are necessary, and it must be granted also, as has been pointed out by Seeliger, Morgan, and Driesch, that in cases of hybridism, as in Boveri's experiment, there may be a marked illustration of what is called unilateral or preponderant inheritance. Most hybrid Echino- derm larvae show maternal characters only, some show paternal characters only, some show both. There is also much individual variability. Thus Boveri's famous experiment affords no secure basis for argument. In further support of the importance of the chromosomes reference may be made to the fact that the number of chro- mosomes in any given organism is always the same, except in the reduced gametes which have half the normal number. Another argument may be found in the fact that in some insects the sex of the offspring seems to depend on whether the egg is fertilised by a spermatozoon with an extra " accessory chromosome " or by a spermatozoon without this. Generally accepted Conclusion. — The general conclusion BEARERS OF THE HEREDITARY QUALITIES 63 from the foregoing and other arguments may be illustrated by two or three quotations from recognised authorities. Prof. O. Hertwig says : " The female nuclear material transmits the characters of the mother, the male nucleus those of the father, to the offspring." Prof. Strasburger says for higher plants : " The process of fertilisation depends upon the union of the sperm-nucleus with the nucleus of the egg-cell ; the cell- substance (cytoplasm) does not share in the process; the cell-substance of the pollen -grain is only the vehicle to conduct the generative nucleus to its destination." Prof. Weismann says : " We can hardly ascribe to the body of the ovum a higher import than FIG. 16. — A pollen grain, a, the two nuclei, with their chromosomes; b, the general protoplasm ; c, the outer wall. (From Carnoy.) that of being the common nutritive basis for the two conjugating nuclei." Criticism. — i. "The life of a complex multicellular organism certainly depends upon the inter-relations and interactions of many parts ; the life of a cell apparently depends upon the inter-relations and interactions of different parts of the cellular organisation, especially on the give-and-take between nucleo- plasm and cytoplasm ; and it is not unlikely that life itself — i.e. vital activity or function — may depend upon the inter- relations and inter-actions of a number of complex substances, none of which could by itself be called alive. Just as the secret 64 THE PHYSICAL BASIS OP INHERITANCE of a firm's success may depend upon a particularly fortunate association of partners, so it may be with vitality." * " We are compelled by the most stringent evidence to admit that the ultimate basis of living matter is not a single chemical substance, but a mixture of many substances that are self- propagating without loss of their specific character." f Holding firmly to this view, which we have elsewhere expressed, that life is a function of inter-relations, we confess to hesitation in accepting without saving clauses any attempt to call this or that part of the germinal matter the exclusive vehicle of the hereditary qualities. 2. The sperm-nucleus brings with it into the ovum a little cytoplasm, and it is also accompanied by the minute central- corpuscle or centrosome, which seems to play an important part in regulating the mechanism of cleavage. It may be that the minimal quantity of cytoplasm is also important, though we cannot trace its behaviour as we do that of the centrosome. Strasburger says that if it were important there would be more of it, but in these matters size and mass seem of small moment ; the little cytoplasm there is may act like the little leaven which leavens the whole lump. It seems in this connection very desirable that the experiments which have been begun (Fieri and Winkler) of extracting a ferment (" ovulase ") from seminal matter and using it as a fertilising agent, should be confirmed or confuted. 3. In Loeb's experiments unfertilised sea-urchin's eggs developed into complete and normal larvae ; the sperm-nucleus was dispensed with. In Delage's experiments non-nucleated fragments of the ova of sea-urchin, worm, and mollusc were fertilised and developed into normal larvae ; the ovum-nucleus was dispensed with. But it must be noted carefully that in both cases there was a nucleus present. * J. Arthur Thomson, Science of Life, p. 115 (London, 1899). t E. B. Wilson, The Cell in Development and Inheritance (ist ed., 1896). CRITICISM 65 4. Hickson (1907) has argued forcibly in support of the view that " for the present at any rate we can only say that the germ-cells as a whole, and not any special part, are responsible for the transmission of heritable characters from, generation to generation." He suggests speculatively that the more plastic characters may be transmitted mainly by the cytoplasm and the rigid characters by the nucleus. In his criticism he refers to cases where chromosomes are quite indistinct in the gametes, to the importance of cytoplasm-fusion in the conjugation of some Protozoa, to the experiments of Herbst and Fischel on hybridisation in Echinoderms, which indicate the im- portance of the cytoplasm of the ovum in transmitting characters, and to other sets of facts which indicate the danger of exaggerating the importance of the chromosomes. The observations oi Godlewski are also strongly suggestive of the importance of the cytoplasm, as well as the nucleus, in inheritance. 5. Bateson (1907) has pointed out that if the chromosomes were the bearers of hereditary characters, we should expect some degree of correspondence between the differences distinguishing the types and the visible differences of number or shape distinguishing the chro- mosomes. Moreover, if the chromosomes were the chief governors of structure we should expect to find greater differences between them in different tissues of the same body. 6. No one has protested more clearly and vigorously than Guyer (1909, 1911) against "the inordinate importance which has been attributed to the chromosomes as vehicles of heredity." He points out, for instance, that there is definite experimental evidence of the great importance of the ovum-cytoplasm, and argues that " the number and arrangement of the chromosomes in a given species are the effects of the fundamental constitution of a given kind of living matter, rather than that they stand in a specifically causal relation to such constitution." " Heredity is the problem of the handing-on of metabolic energies already established, rather than of the transmis- sion of a series of determinative units which create a wholly new organism." " This much is* certain : no chemical, physiological, or morphological evidence is yet extant which places the hereditary factors wholly within the chromosomes." It seems highly probable that the chromosomes " control the velocities in cell-chemistry " by supplying the proper amounts and kinds of ferments which act on a series of fundamental cell-constituents that are largely common to both egg and sperm. Perhaps then the safest conclusion at present is that the chromo- somes, along with other germ-cell constituents, " stand in some definite causal relation to adult characters/' CHAPTER III HEREDITY AND VARIATION " The organic world as a whole is a perpetual flux of changing types." — FRANCIS GALTON. " Inheritance and variation are not two things, but two imperfect views of a single process." — W. K. BROOKS. " Variation and inheritance are, at present, one fundamental mystery of the vital unit." — KARL PEARSON. § i. Persistence and Novelty. § 2. The Tendency to Breed True. § 3. Different Kinds of Organic Change. § 4. Classification and Illustration of Variations. § 5. Fluctuating Variations. § 6. Discontinuous Variations. § 7. De Vries on Fluctuations and Mutations. § 8. Causes of Variation. § i. Persistence and Novelty CLOSE observers of the relation between successive generations in mankind, or among plants and animals, are at one in record- ing two distinct impressions, — on the one hand, of persistent hereditary resemblance, on the other hand, of variability. Oftenest we are first impressed by the remarkable homogeneity which obtains from generation to generation, but as we get to know the organisms better we become aware of individual traits standing out against the background of general similarity. Or it may be that, with the partiality of parents, our first 66 LIKE TENDS TO BEGET LIKE 67 impression is of the novelty and individuality of our children, and only later do we recognise in those, who seemed so original, a re-incarnation of our average selves. Oftener, perhaps, it will be discovered that the resemblance in habits of mind and body is purely mimetic, and that the idiosyncrasies which were really present, as buds at least, have been pruned off both for good and for ill by the hook of criticism, or driven into latency — like " sleeping-buds " — by mis-education or lack of appro- priate stimulus. Like Tends to Beget Like. — The hereditary relation is such that offspring are on the whole like their parents, but the degree of this likeness varies within wide limits. Indeed, the discre- pancies are often very conspicuous, and we can understand how Prosper Lucas, one of the early students of inheritance (1847) — careful and scholarly according to his lights — imagined a meta- physical entity, which he called " Vinneite " and opposed to " I'heredite," the former originating^ what is new, the latter con- serving what is old. In modern phraseology, the occurrence of variations is a fact of life so general that we must replace the adage " Like begets like " with the more cautious statement " Like tends to beget like." The popular adage " Like begets like " is often true as a general statement. Offspring are often so like their parents that even the scientific observer cannot tell one from the other. In other words, the species " breeds true." But the more intimate our acquaintance with organisms becomes, the more plainly do we detect individual peculiarities, and we have to change the adage to " Like tends to beget like." On the whole it is true that average parents have average offspring, that exceptional parents have exceptional offspring. Like tends to beget like. Yet it is well known that, for instance as regards stature, the tall do not always beget the tall, or the small the small, so that we have to broaden the most general "fact of inheritance" still further, and say that the average character 68 HEREDITY AND VARIATION attained by the individuals of one generation tends to be very nearly the same as the average character of the preceding generation. This is the broad fact of specific inertia. A False Antithesis between Heredity and Variation.— Much obscurity of thought has been due to the false antithesis between heredity and variation. When we say that like tends to beget like, that offspring tend to resemble their parents and ancestors, we are stating a fact of life. But when we speak of an opposition between a force or principle of heredity, securing resemblance between offspring and their parents, and a tendency to variability which makes offspring different from their parents, we are indulging in verbiage. Heredity, as we have repeatedly said, is the relation of genetic continuity between successive generations, and it is such that while many characters seen in parents persist in their offspring, there is also in most cases a distinct individuality in these offspring. Heredity is a condition of evolution, a condition of inborn variations ; it is just a name for the reproductive or genetic relation between parents and offspring. The inheritance which was expressed in the development of the parent may be almost identical with the inheritance which is expressed in the development of the offspring, but in most cases the inheritance does not persist in this intact way from generation to generation, and then we speak of variation. The contrast is not between heredity and variation, but between inertia and change, between continuity or persistence and novelty or mutation, between completeness of hereditary resemblance and incompleteness of hereditary resemblance. As Prof. W. K. Brooks says (1906, p. 71) : " Living beings do not exhibit unity and diversity, but unity in diversity. These are not two facts, but one. The fact is the individuality in kinship of living beings. Inheritance and variation are not two things, but two imperfect views of a single process/' TENDENCY TO BREED TRUE 69 § 2. The Tendency to Breed True Relative Stability of Specific Characters. — Belonging as we do to a race which seems to have varied very slowly within historic times, we have not far to seek for good examples of what is the biggest fact of inheritance — the stability of specific characters throughout a long series of generations. If we exclude monstrosities due to arrested development and the like, if we set aside the numerous malformations and deforma- tions induced on the bodies of individuals by peculiarities of function and environment, the stability of the essential human characteristics for many millennia is obvious. This racial inertia, which holds in some measure at least for mental charac- teristics, is at once the hope and the despair of the social reformer. If we pass from general specific characters to those of par- ticular races, we read the same story. Not only do the salient characteristics of the skull persist within a narrow radius of variability, but the same is true of minor features : the oblique eyes of the Japanese, the oval face of the Esquimaux, the woolly hair of the Negro and the Jewish nose. Conservative Types of Organisation.— But the persist- ence of structural and mental characters as illustrated in man- kind is but a tale of yesterday when compared with the persist- ence of type exhibited by many animals which have lived on apparently unchanged for many millions of years. Whatever may be true in regard to the soft parts, of which no record remains, there seem to be no differences in hard parts dis- tinguishing the Lingula of to-day from those of the Silurian ages ; and there are other instances of what are sometimes called " living fossils." The reasons for such remarkable per- sistence do not now concern us, but the fact that structural characters established millions of years ago are reproduced with exactness at the present moment does. 70 HEREDITY AND VARIATION Persistent Peculiarities in Families. — Not less striking than the long persistence of specific and stock characters is the fact that offspring frequently reproduce the individual peculiarities — both normal and abnormal — of their parents or ancestors. A slight structural peculiarity, such as a lock of white hair or an extra digit, may persist for several generations. A slight functional peculiarity, such as left-handedness, has been recorded for at least four generations, and colour-blindness for five. The strong under-lip of the Hapsburgs persisted for six centuries. There are endless illustrations of the fact that a pathological diathesis- rheumatic, gouty, neurotic, or the like — may persist and express itself similarly, even in spite of altered conditions of life, through- out many generations. And what is true of bodily characteristics is not less true of mental peculiarities : as to this, popular im- pressions and the careful investigations of Galton and others are in agreement. We think at once of cases like the Bachs, the Bernouillis, the Darwins ! § 3. Different Kinds of Organic Change It may conduce to clearness if we think over the different kinds of changes which occur in organisms. i. Metabolism. — All living creatures are, as it were, whirl- pools in the universal ocean of matter and energy. They are continually changing as they live. Streams of matter and energy pass in and out. Organisms are animate systems which transform matter and energy in a characteristic way which we call living. Their physical basis is continually undergoing disruption and reconstruction ; it breaks down and is built up again, it wastes and is repaired, it runs down and is ever being wound up again — until the arrears of imperfect recuperation become so serious that the organism dies, or until some fatal accident occurs. The chemical and physical changes involved in living are summed up ORGANIC CHANGES 71 in the term metabolism, the two aspects of which — constructive and disruptive — are called anabolism and katabolism. 2. Cyclic Changes. — An equally familiar fact is that organisms pass through a series of changes. The fertilised egg undergoes cleavage, the resulting cells grow and differentiate, an embryo is formed, and gradually — often by circuitous paths — a minia- ture form of the adult creature is attained. Out of apparent simplicity an obvious complexity results. Growth still con- tinues, often punctuated by resting periods, often rhythmic and FIG. 17. — Diagram illustrative of variation and modification. S, the soma or bod}' ; G. the germinal material ; E, an environmental change. A, an environmental change acting on the body directly evokes a modification (M). B, an environmental change, without modifying the body directly, acts as a stimulus on the germ-plasm, and is followed by a variation (V). C, a variation (V) arises from some germinal change which cannot be causally connected with any particular environmental change. expressible in complex curves, often interrupted by peculiar crises. Quickly or slowly the organism passes from youth through adolescence to maturity, to its limit of growth and its reproductive maturity. Quickly or slowly thereafter it sinks on a down-grade towards death. As the old naturalists said, from one period of vita minima the creature rises to a period of vita maxima, and sinks back again into a vita minima which 72 HEREDITY AND VARIATION dwindles to a vanishing point. It is characteristic of organisms to pass through a series of cyclic changes. 3. Changes involved in Functioning. — As contrasted with inanimate systems, organisms are characterised by their power of effective response to environmental stimuli. A living creature's responses tend towards self-preservation or species-preservation. Though they may fail, the reactions are primarily and funda- mentally effective. And these functionings or effective responses necessarily involve changes in the system. They involve wear and tear, and leave more or less discernible results. Normally, however, the results, known as fatigue-effects and the like, are obliterated by nutrition, rest, and other forms of recuperation. In the study of an intricate structure, like a bee's brain, it is possible to arrange on an inclined plane the changes which are normally obliterated by a night's rest, the changes which require prolonged recuperation before they disappear, and the changes which cannot be recovered from — which accumulate until the bee dies a natural death. 4. Temporary and Individual Adjustments. — In addition to the inherent primary power of effective response, organisms have different degrees of plasticity. They can adjust their reactions to novel conditions. They can " try " first one mode of reaction and then another, finally persisting in that which is most effective. Even the unicellular Infusorians do this. How much of this plasticity is primary, or inherent in the very nature of living matter, how much of it is secondary and wrought out by Natural Selection in the course of ages, must remain in great measure a matter of uncertainty. Each case must be judged on its own merits. It is certain that many unicellular organisms are very plastic, and it seems reasonable to suppose that as differentiation increased, restrictions were placed on the primary plasticity, while a more specialised secondary plasticity was gained in many cases, where the organisms lived in environments liable to frequent vicissitudes. It is convenient to use the MODIFICATIONS AND INBORN VARIATIONS 73 term " accommodations " for the frequently occurring indi- vidual adjustments which many organisms are able to make to new conditions. 5. Modifications. — Besides being plastic, organisms are modifiable: that is to say, in the course of their individual life they are liable to be so impressed by changes in surrounding influences and by consequent changes in function that, as a direct result, modifications of bodily structure or habit are acquired. Modifiability is the capacity of registering the direct results of changed function or of changed environment. " Modifications " may be defined as structural changes in the body of an individual organism, directly induced by changes in function or in environment, which transcend the limit of organic elasticity and persist after the inducing conditions have ceased to operate. They are often inconveniently called " acquired characters." They are not proved to be trans- missible as such or in any representative degree, but they are often adaptive and individually very valuable. They are dis- tinguishable from temporary adjustments or accommoda- tions on the one hand, and from inborn variations on the other. 6. Inborn Variations. — Finally, when we subtract from a total of " observed differences " between members of the same species all that can be described as accommodations and modi- fications, we find a large remainder which we must sharply define off as variations. We cannot causally relate them to peculiarities in habit or in surroundings ; they are often distinct at birth or hinted at before birth; and they are rarely alike even among forms whose conditions of life seem absolutely uniform. They may be large or small in amount, fluctuations or freaks, progressive or retrogressive — that is a matter for further analysis — but they agree in having a germinal origin. They are endogenous, not exogenous ; they are born, not made ; and they are more or less transmissible, though they are not CLASSIFICATION OF VARIATIONS 75 always transmitted. They form — at least some of them form — the raw material of organic evolution. § 4. Classification and Illustration of Variations. "Variation." — It is a common confession of naturalists that a label is a necessary evil. A collection without labels is a contradiction in terms, and yet the label is often a full-stop to investigation. This is true in regard to the concrete ; it is more lamentably true in regard to the abstract. Thus the label " Variation " has been a great hindrance to progress. As Mr. Bateson says (1905, p. 575) : " The indiscriminate confounding of all divergences from type into one heterogeneous heap under the name ' Variation ' effectually concealed those features of order which the phenomena severally present, creating an enduring obstacle to the progress of evolutionary science. Specific normality and distinctness being regarded as an accidental product of exigency, it was thought safe to treat departures from such normality as comparable differences : all were ' variations ' alike." All organic changes imply some incompleteness in the heredi tary resemblance — a little more of one character, a little less of another, or the occurrence of some feature which deserves to be called distinctly " new." Both variations and modifications may cause this incompleteness in the hereditary resemblance ; an apparently similar condition may result from two different processes of change. But the variation has a germinal origin, is blastogenic, is not directly dependent on the external con- ditions of life, is endogenous, and is transmissible ; while the modification has a somatic origin, is the direct result of functional or environmental influence, is exogenous, and, so far as we know at present, is not as such transmissible. Classification.— There are many different ways of classifying these variations which form the raw material of evolutionary change, 76 HEREDITY AND VARIATION a. If we attend to the nature of the change, we may distinguish " meristic " variations — e.g. in the number and pro- portions of parts, from " substantive " variations of a qualitative sort — e.g. change in colour. (3. If we attend to the direction of the change in successive generations, we may distinguish " definite " variations, which occur along one line (like stages in normal development), from "indefinite" variations, which "fluctuate hither and thither with no uniformity in the course of generations." Many evolutionists have maintained that there is good reason for believing in definite or determinate variation along particular lines, as if certain organisms had an inherent bias to change in certain parts and not in others, in certain directions and not in others, just as certain inorganic substances can crystallise in different forms but only within strict limits. It is possible to arrange a series of species A, B, C, D, E, F, in such a way that they suggest progressive definite variation along a particular line, and it seems not unlikely that this kind of evolution may sometimes occur. Moreover, along quite different lines of evolution we find evidence that the same kind of step has been taken independently, over and over again. This suggests that the possibilities of variations may be limited and defined by deep-rooted constitutional conditions or physio- logical alternatives. But the weakness of the argument lies in the almost insuperable difficulty of deciding whether the apparent definiteness is not the result of the primary action of selection which eliminates divergent variants at early stages — nipping idiosyncrasies in the bud — or which may have estab- lished a bias in previous generations. In conditions of rigid elimination the lines of variation will naturally tend to become more and more restricted. y. If we attend to the amount of the change from one generation to the next, we may distinguish minute fluctuations about a mean, which are connected by intergrades, from sudden 2 3 VAR IATION Fig.19 FIG. 19.— Some of the numerous variations in the pattern of the abdomen in the yellow jacket Wasp. [After Kellog and Bell. ] VAR IATION Fig.20. FIG. 20.—" Mutations " or rapidly developing large inheritable variations in Leptinotarsa multitceniata. The type of the species (2; and its extreme mutants ruhiatnda (1) and melanothorax (8). [After W. L. Tower.] [Facing p. 76 VARIATIONS 77 " sports " which reach a new position of organic equilibrium as if by a leap. This is the contrast between " continuous " variations small in amount, and " discontinuous " or " tran- silient " variations in which a considerable step is taken with apparent suddenness, without the occurrence of intermediates. The term variation, used concretely to denote an organic pecu- liarity or idiosyncrasy, is obviously a relative term, implying some standard of comparison. It is a deviation from the parental type, a divergence from the mean of the stock. Thus there are different degrees, or perhaps even different kinds of discontinuity. In many cases, a variation may be described as simply an in- completeness in the inheritance or in the expression of the inherit- ance. The divergence from the norm is due to the suppression or inhibition of some character. This may be illustrated by a per- fectly white (albino) baby, born to almost coal-black parents.* If such a form became the founder of an albino race, as in the case of rats and mice, we should be justified in concluding that the particular material organisation which eventually leads to the deposition of pigment in the body had somehow dropped out of the inheritance. If the albinism was in no respect transmitted to the next generation, we should be justified in concluding that the structural arrangements which lead on to pigmentation had simply been hindered from finding their normal expression in develop- ment. A minus variation like albinism may be described as due to an incompleteness in the inheritance or in the expression of the inheritance, but there are other variations which must, so to speak, bear the plus sign, for they involve the augmentation or exaggeration of a character. Plus variations of this sort have " Its father and mother were horrified ; their fi lends and relations, in fact all the villagers, were called to examine and criticise it. Why such surprise ? Wky such commotion ? The answer is self-evident : the law of heredity had been broken." — R. W. Felkin. The vulgar mind is always impressed by size and quantity ; big deviations strike the imagination, and the normal occurrence of small deviations is forgottent 78 HEREDITY AND VARIATION been taken advantage of in breeding sheep with long fleece, Japanese cocks with tails ten feet long, " wonder horses " with manes reaching the ground, and so on. But the offspring is sometimes so different from the parent that we cannot describe its peculiarity as an incompleteness in the expression of the normal inheritance, or as an exaggeration of parental or ancestral traits. It is sometimes a new pattern, a fresh departure, with what one might call organic originality. It is more than a discontinuous variation. It seems to have passed suddenly into a new position of organic equilibrium, where it has not only individuality, but a distinctively novel individuality. These distinctive novelties, which arise brusquely, are often included in the conception of " mutations." § 5. Fluctuating Variations When we examine a number of individuals of the same species we usually find that they differ from one another in detail. Some of the observed differences may be modificational or due to differences of nurture, but it is often possible to abstract these from differences due to hereditary nature. Thus, when we collect a large number of specimens of the same age from the same place at the same time, we often find that no two are exactly alike. They have peculiarities of germinal origin — or, in other words, they show fluctuating variations. The characteristic feature of these fluctuations is that they are continuous, i.e. con- nected by intergrades, and that they can be arranged in a gradual series (a curve of frequency) on each side of a mode. To construct such a curve (let us say of variation in stature), take a base line, and divide it into equal parts, each to represent a unit of measurement, say an inch. From a middle division of this base line erect an ordinate to represent by its length the number of those individuals whose stature is found to be the most frequent, On each side of this, from their appropriate FLUCTUATING VARIATIONS 79 divisions on the base line, erect ordinates representing by their length the number of individuals of each stature, the lower statures to the left, the greater to the right. Now a line joining the tops of the ordinates will form a polygon or (if the divisions in the base line be quarters of an inch) a curve, which will show graphically the distribution of variation in stature in the population measiired. If the curve is symmetrical on each side of the highest ordinate, the mode, it is called the " normal curve " ; the average or mean coinciding with the " mode." If there are more varia- tions on one side of the mode, the curve is " skew " ; if there are • two maxima or modes, the curve is " dimorphic " ; and so on. In various ways, which are of great practical convenience, a measure of variability can be deduced from the steepness or flat- ness of the curve, and thus we can readily compare the variability of different characters, or of the same character in different groups and at different times. The curves, especially if made year after year, may show the direction in which the species is moving, perhaps the way in which selection is working, perhaps even that the species is splitting up into two subspecies. One of the results of measuring large numbers of variations is to show that there is a relation between the amount of a deviation and the frequency of its occurrence. The greater the divergence from the average, the fewer instances are there. Measurements of a large number of soldiers gave Quetelet the following result» in which the upper line indicates the heights in inches, and the lower line the number of soldiers of each of these heights. 60, 61, 62, 63. 64. 65, 66,- 67, [ 68. 69, 70, 71, 72, 73. 74, 75, \4r 2, 2, 20, 48, 75, 117, 134, isyJH0' I2I» 8°. 57, 26, 13, 5, 3. ^ ^ J^ The general symmetry is plain, on each side of the most frequent condition, 67 inches, which is called the "mode." Registration of Variations. — " The modern methods of statistics deal comprehensively with entire species, and with entire groups of influences, just as if they were single entities, and express thg 80 HEREDITY AND VARIATION relations between them in an equally compendious manner. They commence by marshalling the values in order of magnitude from the smallest up to the largest, thereby converting a mob into an orderly array, which, like a regiment, thenceforth becomes a tactical unit. Conceive each value to be represented by an ex- tremely slender rod of proportionate length, and the rods to be erected side by side, touching one another, upon a horizontal base. The array of closely-packed rods will then form a plane area, bounded by straight lines at its sides and along its base, but by a flowing curve above, which takes note of every one of the values on which it is founded, however immense their multitude may be. The shape of the curve is characteristic of the particular group of values to which it refers, but all arrays have a family resemblance due to similarity of origin ; they all drop steeply at one end, rise steeply at the other, and have a sloping back. An array that has been drilled into some such formation as this, is the tactical unit of the new statistics " (Biometrika, vol. i., 1901, p. 7). Theory of Evolution by Selection of Fluctuating Varia- tions.— It is certain that most offspring differ from their parents in many quantitative details. It is certain that when measure- ments are taken of a large number of individuals of the same species in reference to a particular character, the results, when plotted out, conform approximately to the normal curve of frequency. If measurements be taken in a subsequent genera- tion there is a similar result, but the curve need not be precisely the same. The mode of the curve — i.e. the most frequently occurring dimension of the measured character, may change from one generation to another. It is usually believed that one of the ways in which this change can be effected is by natural selection. But to think of new species arising by slow changes of this sort is in many ways difficult, apart altogether from the fact that definite demonstration of the operation of selection has been rarely attempted. (i) Such a character as a Roman nose is certainly heritable, though it is not always inherited. But we cannot speak so FLUCTUATING VARIATIONS 81 definitely in regard to small quantitative variations. A tall father does not necessarily have tall children. Where the characters in which the two parents differ are such as readily blend, regression towards the mean of the stock will occur. (2) Even with very thorough isolation — segregation of like individuals — and very consistent selection, it is doubtful whether a new race could be evolved from the cumulative increase of small quantitative variations, e.g. in stature or colour of hair. It is doubtful whether any domestic races have so arisen. It is not in this way that dwarf-races and giant-races have been formed. They arise from sudden discontinuous variations or mutations, which are often peculiarly heritable, which are any- thing but liable to be swamped by inter-crossing, and which sometimes exhibit Mendelian inheritance. (3) The result of the gradual accumulation of small quantitative variations may be very important in a long time, just as a small sum may become large from interest accumulated for centuries ; but it is difficult to believe that minute fluctuations in quantity would always have sufficient selective value to ensure their persistence. There are several reasons why selectionists have restricted themselves so much to continuous variations as the raw material of evolution, (i) Until lately we have known comparatively little in regard to discontinuous variations or mutations. (2) It was hastily concluded that these changes were not likely to be transmitted — a generalisation in part due to preoccupation with teratological non- viable freaks. (3) In many cases related species can be arranged in a gradual series with intermediate forms linking the extremes. Now, there is no need to hamper the Evolution Theory by restricting selection to minute variations. We know that sports, mutations, or discontinuous variations are frequent, and that they are remarkably stable in their hereditary transmission. 6 82 HEREDITY AND VARIATION We know also that many domestic races have, as a matter of fact, arisen by sudden mutation. As to the series of related species which may be often arranged as if on an inclined plane, two points should be noted : (i) that it is likely enough that some kinds of species, e.g. vege- tative forms like Alcyonarians and Corals, may have evolved by minute steps, and (2) that although species are often connected by intermediate links it does not follow that these links are stages in the evolution. They may have been formed after the species to which they are theoretically supposed to give rise. We should remember Galton's warning, " If all the variations-of any machine that had ever been invented were selected and arranged in a museum, each would differ so little from its neigh- bours as to suggest the fallacious inference that the successive inventions of that machine had progressed by means of a very large number of hardly discernible steps." Many facts now lead us to conclude that the Proteus leaps as well as creeps. § 6. Discontinuous Variations One of the steps of progress in Evolution- lore since Darwin's day is the recognition of the frequency and importance of dis- continuous variations — i.e. of organic changes which arise abruptly and not by a gradual series of steps. If dwarfs arise suddenly in a tall race, and are not mere modifications, they illustrate discontinuous variation of a quantitative sort. A hornless calf, a tail-less kitten, a short-legged lamb, a thornless rose, illustrate discontinuous quantitative variations of a negative kind. Giants, " wonder-horses," long-tailed Japanese cocks, merino-fleeced sheep, spine-covered holly leaves illustrate dis- continuous quantitative variations of a positive kind. Sometimes the novelty cannot be readily expressed in quantitative terms — an entirely new colour turns up, the variant is immune to certain diseases to which the stock is susceptible, leaves become fasciated, DISCONTINUOUS VARIATIONS 83 a tree becomes " weeping," a genius is born. When a new pattern of organisation or a new constitutional property turns up, we may speak of a discontinuous qualitative variation. Historical Note. — The idea that organic changes might come about by leaps and bounds is not novel, though the evidence substantiating it is quite modern. Some of the older evolutionists, such as Etienne Geoffroy St. Hilaire, believed in saltatory evolution, and were far from agreeing with Lamarck that Nature is never brusque. Darwin also recognised that big steps may be taken suddenly — e.g. in the origin of large-crested Polish fowls, black-shouldered peacocks, short-legged Ancon sheep, but he thought that these discontinuous variations occurred rarely, and would be liable to be swamped by intercrossing. He relied rather on the action of natural selection on the small, continuous variations which are always forthcoming. But the modern appreciation of the importance and frequency of discontinuous variations is mainly due to Bateson, who, in his Materials for the Study of Variation (1894), gave many instances of the sudden appearance of offspring which in some particular diverge widely and abruptly from their parents ; and to De Vries, who has observed the occurrence of " mutations " in many plants, and has also followed them through generations, showing that they tend to breed true ; and to Johannsen, who recognised the import- ance of individual new departures in starting stable " pure lines." A Change of View. — Darwin and orthodox Darwinians relied in the main on the operation of selection on small individual variations — many of which are nothing more than quantitative fluctuations. If new adaptations and new discontinuous species arise in this way, the small variations must be heritable, the new character must be capable of cumulative increase by the per- sistent outcrop of similar variations generation after generation, the selection must be persistent and consistent, and a long time must be allowed. 84 HEREDITY AND VARIATION Even when this theory is strengthened by subsidiary theories, e.g. as to the efficacy of isolation and germinal selection, it is more theoretically than practically convincing. It places such a heavy burden on the shoulders of Natural Selection that the idea of a leaping instead of a creeping Proteus has always been welcome. But why are evolutionists now entertaining an idea — the importance of discontinuous variations — which Darwin con- sidered and then rejected ? The answer is that we now know of many instances of discontinuous variation in animals, and even more among plants, that we have some good evidence of these discontinuous variations or mutations " breeding true," and that we have in the theory of Mendelian inheritance a reason why a mutation which has once arrived should persist. Some modern authorities go the length of saying that " mutations " form the sole raw material of evolution, and that "individual fluctuations" do not count at all. This seems an illustration of the common tendency to take 'up an extreme position in the enthusiasm of a new discovery. Because dis- continuous variations are common and important it does not follow that continuous fluctuations are of no moment. Those " whose humour is nothing but mutation " confess that it is very difficult to distinguish between a small mutation and a large fluctuation. If the large fluctuation be heritable — which we may assume until it has been disproved — we confess that we do not see what is gained by trying to distinguish it from a small mutation. The New View. — Dominated by the idea that " organisms are mere conglomerates of adaptative devices," and that these patents cannot but be the outcome of slow accumulation of minute fluctuations under the directive agency of selection, naturalists have paid little heed to the open secret that the living creature is inherently a Proteus suddenly and discon- DISCONTINUOUS VARIATIONS 85 tinuously passing from one guise to another by transilient variation. Mr. Bateson (1905, p. 577) notes that Marchant in 1719 was the earliest to comment on the suggestiveness of sudden changes, such as he saw in plants of Mercurialis with laciniated and hair- like leaves which for a time established themselves in his garden. He suggested that species may arise in like manner. " Though the same conclusion has appeared inevitable to many, including authorities of very diverse experience, such as Huxley, Virchow, F. Galton, it has been strenuously resisted by the bulk of scientific opinion, especially in England. " Upon whatever character the attention be fixed, whether size, number, form of the whole or of the parts, proportion, distribution of differentiation, sexual characters, fertility, pre- cocity or lateness, colour, susceptibility to cold or to disease — in short, all the kinds of characters which we think of as best exemplifying specific difference, we are certain to find illustrations of the occurrence of departures from normality, presenting ex- actly the same definiteness elsewhere characteristic of normality itself. Again and again the circumstances of their occurrence render it impossible to suppose that these striking differences are the product of continued selection, or, indeed, that they represent the results of a gradual transformation of any kind. Whenever by any collocation, of favouring circumstances such definite novelties possess a superior viability, supplanting their ' normal ' relatives, it is obvious that new types will be created/' Heredity and Evolution.— Mr. Bateson has done good service in exposing to ridicule the prevalent misconception that domesticated races are " so many incarnations of the breeder's prophetic fancy." " Except in recombinations of pre-existing characters — now a comprehensible process — and in such intensi- fications and such finishing touches as involve variations which analogy makes probable, the part played by prophecy is small 86 HEREDITY AND VARIATION Variation leads ; the breeder follows. The breeder's method is to notice a desirable novelty, and to work up a stock of it, picking up other novelties in his course — for these genetic disturbances often spread — and we may rest assured the method of nature is not very different " (1905, p. 578). This is obviously a very important change of view, though it is also in a way a return to what Darwin himself taught. " Variation leads ; the breeder follows." But more than that : Variation leads by leaps and bounds. As Mr. Bateson says, let the believer in the efficacy of selection operating on continuous fluctuations try to breed a white or a black rat from a pure strain of black-and-white rats by choosing for breeding the whitest or the blackest ; or to raise a dwarf (" Cupid ") sweet pea from a tall race by choosing the shortest. It will not work. Variation leads and selection follows. Illustrations of Discontinuous Variation Wonder Horses. — The so-called wonder-horse " Linus I." had a mane eighteen feet long and a tail twenty-one feet long. The parents and grandparents had unusually long hair. This seems a good illustration of a " sport " or discontinuous variation which not only persisted for several generations, but increased very rapidly. Shirley Poppies. — The well-known Shirley Poppies arose from a single discontinuous variation, which may have occurred often before Mr. Wilks saved it from elimination and made it the ancestor of a prolific and distinctive stock. Star Primrose. — The graceful star primrose (Primula stellata) arose as a sport from the conventional Chinese primrose, and was raised by Messrs. Sutton into a favourite stock. It had been thrown off before as a sporadic variety over and over again, but was " promptly extirpated because repugnant to mid- Victorian primness." DISCONTINUOUS VARIATIONS 87 The Moth Amphidasys. — Some sixty years ago in the urban conditions of Manchester the black variety doubledayaria of the moth Amphidasys betularia found its chance, and soon practically superseded the type in its place of origin, extended over England, and appeared even in Belgium and Germany (Bateson, 1905, p. 577). The Common Jelly Fish. — A good case of abundant discon- tinuity in variation is furnished by the common jelly-fish Aurelia aurita, whose sports have been studied by eight or more ob- servers, from Ehrenberg (1835) onwards. Its parts are normally in multiples of 4 (4 equal areas in the radially symmetrical disc, 4 oral lips, 4 genital organs, 16 radial canals, 8 marginal sense- organs or tentaculocysts) ; but numerical sports are very common. These are sometimes irregular, e.g. when the radial symmetry of the disc is lost ; but they are oftener quite sym- metrical, e.g. when the animal has 2 genital organs, 2 oral lobes, 8 radial canals, and 2 marginal sense-organs. In studying Aurelia aurita at Plymouth, Browne (1895) found that out of 1515 young forms (ephyrae) 21^4 per cent, had more or fewer than 8 marginal sense-organs, and that out of 383 adults 22 '8 per cent, were similarly affected. The figures seem to show that the abnormal forms survive quite as well as the normal forms, yet there is no evidence that the sports were more numerous in 1895 than when Ehrenberg studied them sixty years before. In other words, although a plentiful crop of brusque variations is being continually supplied by this plastic form, there is no hint of the origin of a new race. (Bateson, 1894, p. 428.) The Case of Pseudoclytia. — Although the numerous discon- tinuous variations of Aurelia aurita do not suggest that any new race is at present arising, it is possible to find an analogous case where it does seem that we have to do with a species newly arisen, or still in process of being established. A. G. Mayer found at the Tortugas, Florida, large numbers of a medusoid 88 HEREDITY AND VARIATION or swimming bell — Pseudoclytia pentata — a leptomedusan belonging to the family Eucopidae. " It differs from all other Hydromedusae in that it normally possesses 5 radial canals, 5 lips, and 5 gonads, all 72° apart, instead of 4 of these various organs 90° apart, as in other Eucopidae." In the structure of its tentacles, otocysts, gonads, and manubrium, in the general shape of its bell, and the arrangement of its tentacles and otocysts, it is so closely similar to Epenthesis folleata, that it seems safe to conclude that the former has been derived from the latter or from some closely allied species. The two forms are somewhat different in colour and slightly different as to the position of the gonads, but the resemblance is exceedingly close, and no one can suppose that a medusoid with 5 radial canals is a primitive form. As there are pentamerous variants of Epenthesis folleata and tetramerous variants of Pseudoclytia pentata, we are not aware of any case which more cogently suggests the evolutionary interpretation. As Mayer says, " P. pentata may be called ' a new race ' in the sense that it is evidently derived from Epenthesis, and departs from the quadratic arrangement of organs, which is almost uni- versal among Hydromedusae. It is remarkably variable, and its great commonness attests to its successfulness in the struggle for existence" (Mayer, 1901, p. 20)." To obviate misunderstanding, it may be observed that by the term " newly arisen " which Mayer uses in reference to Pseudoclytia pentata, he means simply that " it has departed widely from the fundamental type of all other Hydromedusae, and that it is appar- ently derived from a genus (Epenthesis] which is itself quite highly differentiated. It is, therefore, ' new ' in the sense that it cannot be a primitive form, although we have no means of determining how long a time it may have been in existence " (Mayer, 1901, p. 8). While we cannot exactly demonstrate that Pseudoclytia pentata arose by discontinuous variation from Epenthesis folleata, or some closely allied form, the evidence in favour of that interpretation DISCONTINUOUS VARIATIONS 89 is exceedingly strong. It is interesting further to notice that " the newly-arisen species " is very successful as regards numbers, and that its variations have a strong family resemblance to those of its supposed ancestor, and are yet more abundant. In regard to its more abnormal variants, Mayer observes that they are handicapped by their loss of symmetry, for some are neither radial nor bilateral, and by a reduction of fertility even in cases where the number of gonads has been increased to six or seven. The evidence from Medusae and Medusoids is sufficient to show FIG. 21. — Mutation in Medusoids (after A. G. Mayer). The figure to the left is an oral view of Epenthesis folleata. The figure to the right is an oral view of Pseudoclytia pentata. that discontinuous variations may occur in large numbers, that similar brusque changes may occur year after year, that there is sometimes a strong family resemblance in the variations of related forms. In some cases (e.g. in regard to Aurelia aurita) we are not in a position to say that anything has come of the abundant crop of discontinuous variations ; in other cases (e.g. the very abnormal forms of Pseudoclytia pentata} the discontinuity has gone too far, as shown by the reduction of fertility and the entire loss of sym- metry ; while, thirdly, from the relationship of Pseudoclytia pentata to Epenthesis folleata, we are led to conclude that one species may arise from the discontinuous variation of another. 90 HEREDITY AND VARIATION § 7. De Vries on Fluctuations and Mutations. Professor Hugo de Vries is one of the foremost of Darwin's intellectual heirs, with a rich endowment of his insight and patience. Long- continued and carefully controlled observations and experiments with generations of plants have led him to conclusions which have given the Evolution Theory a fresh start. His " Mutation Theory" is certainly one of the greatest advances since Darwin's day. The General Idea. — The origin of species and varieties is an object for experimental inquiry. " Comparative studies have contributed all the evidence hitherto adduced for the support of the Darwinian theory of descent, and given us some general ideas about the main lines of the pedigree of the vegetable kingdom, but the way in which one species originates from another has not been adequately explained. The current belief assumes that species are slowly changed into new types. In contradiction to this conception the theory of mutation assumes that new species and varieties are produced from existing forms by sudden leaps. The parent-type itself remains unchanged throughout this process, and may repeatedly give birth to new forms. These may arise simul- taneously and in groups, or separately at more or less widely distant periods. .... My work claims to be in full accord with the principles laid down by Darwin, and to give a thorough and sharp analysis of some of the ideas of variability, inheritance, selection, and mutation, which were necessarily vague at his time" (From preface to Species and Varieties, their Origin by Mutation" Chicago and London, 1905). A Theoretical Implication. — De Vries's Mutation Theory involves the theoretical conception that " the characters of the organism are made up of elements that are sharply separated from each other. These elements can be combined in groups, and in related species the same combinations of elements recur. Transitional forms like those that are so common in the external MUTATION THEORY OF DE VRIES 91 features of animals and plants do not exist between the elements themselves, any more than they do between the elements of the chemist." The Case of the Evening Primrose. — In 1886, De Vries began hunting about around Amsterdam for a plant which would show hints of being in what we may call a changeful mood. He tried over a hundred species, bringing them under cultivation, but almost all were disappointingly conservative. It seemed as if most of the species around Amsterdam were in a non-mutable state. It is possible, as Weismann suggested in one of his first evolutionary essays (1872), that in the life of species periods of constancy alternate with periods of changefulness. The human historian has often made a similar remark. In the course of his wanderings around Amsterdam, De Vries came across a deserted potato-field at Hilversum — a field of treasure for him. For there he found his long-looked-for mutable plant, an evening primrose ((Enothera lamarckiana) . Like its nearest relatives, (Enothera biennis and (Enothera muricata, which it excels in size and beauty of flowers, it probably came from America, where it is a native. It had probably " escaped " at Hilversum about 1875, and in the following ten years it had spread in hundreds over the field. It had been extremely prolific in its freedom, but that was not its chief interest. Its chief interest was its changefulness. It had, so to speak, frolicked in its freedom. Almost all its organs were varying — as if swayed by a restless tide of life. It showed minute fluctuations from generation to generation ; it showed extraordinary freaks like fasciation and pitcher-forming ; it showed hesitancy as to how long it meant to live, for while the majority were biennial, many were annual, and a few were triennial ; it showed what can hardly be otherwise described than as new species in the making. It is possible that the prolific multiplication in a new environ- ment may have had something to do with the awakening of the impulsive mutability. 92 HEREDITY AND VARIATION In 1887, a year after his discovery of the potato-field, De Vries found two well-defined new forms — a short-styled 0. brevistylis and a beautiful smooth-leaved 0. Icevifolia — distinguishable from the parent 0. lamarckiana in many details. He hailed these as two new " elementary species," * and he applied one of the crucial tests of specific or subspecific rank : Did they breed true ? He found that this was so ; from their self-fertilised seeds similar forms arose. Neither of the two new forms was represented in the herbaria at Leyden, Paris, or Kew ; neither had been described in the literature of Onagraceae. They seemed to be distinctively new. It is interesting to note that in 1887 there were few ex- amples of these two new elementary species, and that each occurred on a single plot in the field. The impression conveyed was that each had arisen — by a sudden mutation — from the seed of an individual parent. The next chapter in the famous investigation began with a transference of samples of the new forms and the parent stock — partly as plants and partly as seeds — from the potato-field at Hilversum to the botanic garden at Amsterdam. The three stocks gave rise under cultivation to many thousands of individuals, which bred true along certain lines, and yet gave rise to other new forms. In short, De Vries had found a plant in process of evolution. The predisposition to mutability — which remains a mystery — was present, De Vries gave it scope, and like the primeval gardener he had the pleasure of giving names to a crop of new creations which emerged before him. From each of his three samples there arose several distinctive groups — which if they had been found in nature would have been reckoned as distinct species of evening primrose. But the most interesting feature was the apparent abruptness in the origin of the ne\v * By an " elementary species " is meant simply a group of individuals which agree with one another and differ from other groups in a certain number of characters, normally constant through successive generations. MUTATION THEORY OF DE VRIES 93 forms. They seemed to arise by leaps and bounds, by organic jerks ; they illustrated what De Vries has called " Mutation." Besides the smooth-leaved O. lavifolia and the short-styled 0. brevistylis, both of which appeared in the potato-field, the cultiva- tion of O. lamarckiana resulted in the emergence of sej^en^constant elementary species — O. gigas (rare), O. rubrinervis, O. oblongata, 0. albida, O. leptocarpa, O. lata, and a dwarf O. nanella. Besides these there were a few inconstant variants and a few which were sterile. One form, O. scintillans, that only appeared eight times, was not constant like the others. When self-fertilised it produced O. ob- longata, 0. lamarckiana, and others like itself. It is interesting to notice that some of the forms — e.g. O. oblongata — were produced over and over again ; that five of the new forms appeared afterwards in the field or from seeds collected in the field, which shows that the cause of their origin was not to be found in the cultivation. As De Vries says, the new elementary species arise suddenly without transitional links ; for the most part they are quite con- stant ; within the limits of their essential constancy they exhibit similar minor fluctuations ; they are usually represented by nu- merous individuals within the same period of time ; the observed changes affect many organs and parts, and in no definite direction ; and the mutability seems to be periodic, not continuous. If cases like that of 0. lamarckiana are indicative of what often occurs and has occurred in nature, then our view of the evolution- process must be in several respects modified. It will be necessary to distinguish more sharply between fluc- tuating variations and discontinuous mutations. If a new ele- mentary species may arise as it were ready-made, " at a single advance," it is not necessary to hold to the formula that species have arisen by the gradual accumulation, (under selection) of minute individual variations. As mutations occur in large numbers and occur repeatedly and are very constant, the familiar difficulties in regard to the swamping of novelties, the inappre- ciable value of incipient stages, the apparent non-utilitarian character of some specific differences, and so on, will be greatly 94 HEREDITY AND VARIATION lessened. The reader may be referred to Prof. T. H. Morgan's Evolution and Adaptation (1903) for a valuable discussion of the advantages of the Mutation Theory. De Vries's Analysis of Variation. — In order to appreciate more thoroughly the importance of the changes which De Vries has necessitated in our evolutionary conceptions, we must briefly refer to his analysis of the distinct phenomena which have been too often unfortunately slumped under the title " Variations." " Elementary Species." — In many groups of organisms which are usually called Linnaean species, there are several or numerous " subspecies," or " varieties." They remain more or less constant in their characters from generation to generation, they breed true in artificial conditions, they are not local races with similar modi- fications ; De Vries calls them " elementary species." Thus there are about two hundred " elementary species " of the com- mon Crucifer, Draba verna, and a few " elementary species " of the common European heartsease (Viola tricolor), and so on. " The systematic species," De Vries says, " are the practical units of the systematists and florists, and all friends of wild nature should do their utmost to preserve them as Linnaeus has proposed them. These units, however, are not really existing entities ; they have as little claim to be regarded as such as the genera and families have. The real units are the elementary species ; their limits often apparently overlap, and can only in rare cases be determined on the sole ground of field-observations. Pedigree- culture is the method required, and any form which remains constant and distinct from its allies in the garden is to be con- sidered as an elementary species " (1905, p. 12). Elementary species are considered to have originated from their parent form in a progressive way ; they have succeeded in attaining something quite new for themselves. Retrograde Varieties. — De Vries applies this term to those numerous forms which have thrown off some peculiarity charac- teristic of their ancestors. Like elementary species they may arise DE VRIES'S ANALYSIS OF VARIATION 95 suddenly, but while "progressive steps are the marks of ele- mentary species, retrograde varieties are distinguished by appar- ent losses." Retrograde varieties usually differ from their parent species by a single sharp character only,— they have lost pigment, or hairs, or spines, and so on ; while elementary species are dis- tinguished from their nearest allies in almost all organs. More- over, the same kind of retrograde variety occurs repeatedly in different series of species, hence the long lists of unrelated varieties called by the same varietal title— e.g. alba, inermis, canescens, or glabra. " Varieties differ from elementary species in that they do not possess anything really new. They originate for the greater part in a negative way, by the apparent loss of some quality, and rarely in a positive manner by acquiring a character already seen in allied species " (1905, p. 152). Ever-sporting Varieties.— De Vries uses this term to describe cases like the striped larkspur," which for centuries has gone on producing unstriped as well as striped flowers. " Its changes are limited to a rather narrow circle, and this circle is as constant as the peculiarities of any other constant species or variety. But within this circle it is always changing, from small stripes to broad streaks, and from them to pure colours. Here the vari- ability is a thing of absolute constancy, while the constancy con- sists in eternal changes ! " Plants with variegated leaves, with double flowers, with fasciated branches, with peloric flowers, and so on, often illustrate the " ever-sporting " tendency. The common snapdragon (Antirrhinum majus) is a very good case, the striped variety, for instance, cannot be fixed. There is some inherent instability in the combination of unit-characters in these ever-sporting varieties. Fluctuations.— De Vries applies this term to the continually occurring individual variations. "It is normal for organisms to fluctuate to and fro, oscillating around an average type. Fluctuations are linear, amplifying or lessening the existing 96 HEREDITY AND VARIATION qualities, but not really changing their nature. They are not observed to produce anything quite new ; they always oscillate around an average, and if removed from this for a time, they show a tendency to return to it." They are inadequate ever to make a single step along the great lines of evolution, whether progressively or retrogressively. They do not form the raw material of evolution, as* has often been supposed. But, we submit, it is difficult with our present knowledge to discriminate between a fairly large fluctuation and a small mutation. Mutations. — " In contrast to the ever-recurring variability, never absent in any large group of individuals, and determining the differences which are always to be seen between parents and their children, or between the children themselves, we have to rank the so-called sports or single varieties, not rarely denomin- ated spontaneous variations, for which I propose to use the term ' mutations.' They are of very rare occurrence, and are to be considered as sudden and definite steps " (1905, pp. 190-1). " De Vries recalls Galton's apt comparison between variability and a polyhedron which can roll from one face to another. When it comes to rest on any particular face, it is in stable equilibrium. Small vibrations or disturbances may make it oscillate, but it returns always to the same face. These oscillations are like the fluctuating variations. A greater disturbance may cause the polyhedron to roll over on to a new face, where it comes to rest again, only showing the ever-present fluctuations around its new centre. The new positio'n corresponds to a mutation " (T. H. Morgan, 1903, p. 289). According to De Vries, mutations have furnished the material for the process of evolution. The Oldest Known Mutation. — A few years before the close of the sixteenth century (1590), 'rprenger, an apothecary of Heidelberg, found in his garden a pecufiar form of Chelidonium majus or greater celandine. It was marked by having its leaves cut into narrow lobes with almost linear tips, and by having the THE OLDEST KNOWN MUTATION 97 petals also cut up. This sharply denned new form suddenly appeared among the plants of Chelidonium majus which the apothecary had cultivated for many years. It was recognised by botanists as something quite new, and eventually it got the name Chelidonium laciniatum ; it was not to be found wild, or anywhere except in the Heidelberg garden. But from the first this new cut-leaved celandine proved constant from seed. It has been naturalised in England and other countries, and is sometimes now found as an " escape." Its origin by mutation seems as certain as its constancy. It is further of interest to note that in crosses with C. majus it follows the law of Mendel. Summary. — De Vries has done great service in analysing the complex concept of variation ; in sharply contrasting individual fluctuations and mutations; in defining "elementary species," "retrograde varieties," and "ever-sporting varieties"; in ob- serving the actual origin by mutation of stable new varieties or subspecies of (Enothera lamarckiana and some other plants ; in showing by historical research combined with experiment that many stable stocks of cultivated plants have arisen by mutation ; and by corroborating throughout the fundamental idea that " the characters of organisms are composed of units sharply distin- guished from one another." The contrast between fluctuations and mutations is so impor- tant that we may state it once more, (i) Fluctuations are continually occurring generation after generation : mutations are rare and occur intermittently. (2) Fluctuations give rise to a series of minute differences which may be arranged on a frequency curve, according to the laws of chance : mutations may be large or small, and their occurrences do not illustrate any ascertained law of frequency. (3) Fluctuations do not lead to a permanent change in the mean of the species unless there be very rigorous selection, and even then, if the selection be slackened, there is regression to the old mean : mutations lead per saltum to a new specific position, and there is no regression to the old mean, 7 98 HEREDITY AND VARIATION (4) Fluctuations do not yield anything really new, they imply a little more or a little less of characters already present : mutations are novelties, they imply some new pattern, some new position of organic equilibrium. According to De Vries's theory, no new species can be established without mutation. "When a muta- tion has occurred a new species is already in existence, and will remain in existence, unless all the progeny of the mutation are destroyed." . . . The phrase " survival of the fittest," as de- scribing a process of evolution, ought to be replaced by " survival of the fittest species." According to De Vries, species originate by mutation instead of by the continuous selection of fluctuations. " Natural Selection may explain the survival of the fittest, but it cannot explain the arrival of the fittest." In regard to these far-reaching conclusions it should be noted that while De Vries has given much convincing evidence in regard to plants, we have as yet very slight evidence of the origin of species of animals by mutation. We know of many discontinuous variations among animals, but the subsequent history of these is not known except in a few cases. It must be remembered that, morphologically regarded, the whole vegetable kingdom does not correspond to more than the first three or four phyla in the animal kingdom— to the Protozoa, Porifera, and Ccelentera, where, as in plants, the contrast between germ-plasm andsomatoplasm has not been accentuated, as it is in higher animals. It is quite conceiv- able that a mode of evolution common among plants may be rare among animals. It is difficult at present to apply the mutation concept with security to the animal kingdom. The idea of mutation is very welcome because it lessens the burden which it has been found theoretically necessary to lay on the shoulders of the selection hypothesis, and because it fits in well with the a priori convictions which some naturalists have as to the autonomy of the organism, that it is as much a self -changing insurgent Proteus as a pawn in a game which the Environment plays. But because it is so welcome, it is to be entertained VARIATION IN HARTS TONGUE FERN FIG. 22. — Mutations of Hart's Tongue Fern ( Scofapendrium imlgare) After LOWE. i. Typical ; 2, variety sagittato-cristatum ; 8, reniforme ; 4, cristatum ; 5, contractum ; 6, stansfieldii. [Facing />. 98 FLUCTUATIONS AND MUTATIONS 99 the more cautiously. An authority on domesticated animals, Prof. Keller of Zurich, finds but little evidence of it in the history of the well-known stocks. It seems to us that in emphasising the importance of mutations De Vries has swung to the extreme of greatly depreciating the importance of fluctuations. Until we know more about animal mutations, it does not seem to us legitimate to deny that fluc- tuations may form, as Darwin believed, an important part of the raw material on which selection operates. We cannot but regard with suspicion the distinction between large fluctuations and small mutations. It seems to us a verbal distinction. Finally, it must be remembered that, as De Vries frankly points out, we are ignorant in regard to the conditions in which mutations occur. The Mutation Theory does not as yet give us a theory of mutations. " Pure Lines." — The position held by De Vries has been strength- ened by the work of Johannsen and Jennings on " pure lines." If we succeed in starting a " pure line " — " the progeny of a single self- f ertilised homozygous plant" — say an innately exceptional bean-plant with very large seeds, we shall find slight individual differences in the size of the beans from generation to generation ; if we take the biggest and the smallest of these and start afresh, we find that their progeny are neither larger nor smaller than the average. The original bigness was a -fixed mutation ; the other differences were probably mere modifications and non-transmissible. If we take a considerable number of the largest beans and the smallest beans from a field and sow them, we are likely to get in the progeny of the former a larger average size than in the progeny of the latter, for we are almost sure to have started with a number of beans which are innately (not modi- ficationally) large-sized and small-sized. What Johannsen did for the bean and some other plants, Jennings has done for the slipper- animalcule, Paramascium. He isolated eight pure lines differing in average size, and found that he made no progress by selecting the largest in an established large pure line, the exceptional largeness being probably the accidental result of peculiar nurture. Selection from a mixed population, however, resulted, as in the case of the beans, in a distinctly altered average size. ioo HEREDITY AND VARIATION The experiments were made with consummate carefulness, but it is difficult to accept the idea of the rigid fixity of the hereditary char- acters in a pure line. It may be that in some cases, such as beans, the viable limit of size has been reached. It may also be that the variational steps that count do not occur often. Perhaps some time must elapse before the organism takes another step. Prof. Castle asks : " Is it not possible that along with the striking size differences due to nutrition there may occur also slight size differences due to germinal variation within the pure line, that is owing to variations in the potency of the same unit-character or com- bination of unit-characters ? " And he points to Woltereck's success- ful selecting-out of a variation in a parthenogenetic pure line of the water-flea, Hyalodaphnia. He selected forms which showed the ex- ceptional occurrence of a rudimentary eye, and definitely increased the degree of development of that organ and the frequency of its occurrence (up to 90 per cent.). In short, it is premature to abandon belief in the efficacy of selection even in pure lines. § 8. Causes of Variation In regard to the causes of variation it is too soon to speak, except in tentative whispers. What Darwin said must still be said : " Our ignorance of the laws of variation is profound. Not in one case out of a hundred can we pretend to assign any reason why this or that part has varied." Variability. — The difficulty which every naturalist has felt in trying to define the concepts of variability and variation is due to the fact that living creatures are individualities — in some degree, personalities. In the ocean of matter and energy organisms are, as it were, whirlpools, each one with a particular character of its own. They are animate systems, each with a unity or individu- ality which we cannot fully interpret. They have the power — again an ultimate prerogative — of giving rise to other whirlpools, to other animate systems, which tend to be like themselves. But because each organism is a very complex whirlpool in a very com- plex environment, and because a living individuality cannot reproduce others without subtle molecular manoeuvres which we CAUSES OF VARIATION , ici know only in a far-off sort of way, one individuality is very un- likely to reproduce an absolute facsimile of itself. It is of the very essence of a living thing to change, and an individuality cannot be halved. From this point of view, variation is a primarily normal occurrence, and breeding true has secondarily come about as the result of restriction. In short, variability is a primeval character of organisms. We cannot explain variability ; it is a datum in the world of life. We may, however, try to show in certain cases how it operates and what conditions help or hinder it. The unending problem of life is to establish some sort of modus vivendi between an extremely complex and changeful animate system and the extremely complex and changeful environment in which it lives and moves and has its being. In all viable organ- isms this equilibration has been established, and it is plain that those organisms which could secure an entailment of this equili- bration would be the organisms to survive. The producers of survivable descendants survive in them — an obvious economy of successful experiment, if such a point of view can be entertained. We have seen that during the early stages of development there is often a visible segregation of a lineage of germ-cells which do not share in body-making, but continue like the fertilised ovum. This distinction between somatic cells which undergo differentia- tion and germ-cells which retain the heritable qualities intact is obviously an advantageous method of entailing on successive generations that valuable asset which we have called organic equilibration. It also economises and facilitates the process of reproduction. But in spite of this almost universal device, the general tend- ency of which is to secure persistence, continuity, and complete hereditary resemblance, there is abundant opportunity left for the assertion of that variability which we believe to be a primary quality of vital units. Thus an inquiry into the causes of varia- tion seems to us to be in the main an inquiry into the oppor- tunities for the reassertion of a pristine tendency which the 102 HEREDITY AND VARIATION continuity of the germ-plasm has to some extent restricted. The stream of life passing through a continuous lineage of germ-cells is, so to speak, hemmed in, but it continually tends to deviate from this course, and there are not a few opportunities — some normally recurrent, some more accidental — which allow of this or even prompt it. In some cases, as we have said, it is impossible to distinguish offspring from parent, or brother from brother, or cousin from cousin. On what does this completeness of heredi- tary resemblance (i.e. the absence of variation) depend ? It means, in the case of unicellular organisms, that the sepa- rated parts are identical in substance and carry on the complete organisation of the parent cell in absolute integrity. In the case of multicellular organisms it depends on the same thing. The cell which in the embryo begins the germ-cell lineage may be identical with the fertilised ovum, and the complete heritage may be con- tinued intact through successive cell-divisions until the next generation is started, and the process begins anew. The com- pleteness of hereditary resemblances depends, in Bateson's phrase, on " that qualitative symmetry characteristic of all non- differentiating cell-divisions." It seems, therefore, useful to say that variation is " the expres- sion of a qualitative asymmetry beginning in gametogenesis. Variation is a novel cell-division." But to tell what specific cause induces this novelty is still beyond our power. Yet we can point to certain conditions which may induce novelty or qualitative asymmetry in gametogenesis. Thus, there is the complex change- ful environment of the developing germ-cells, there is the possible struggle of analogous hereditary units or determinants for sus- tenance, there is the complex process of reduction which occurs during the maturation of the germ-cells, and there are the chances of new combinations and permutations in fertilisation. Results of A mphimixis. — That amphimixis is one of the provocatives of variations is strongly suggested by what results when two breeds FIG. 23. — Karyokinesis. (After Flemming.) i, Coil stage of nucleus ; cc centrosome ; 2, Division of chromatin into U-shaped loops, and longitudinal splitting of these (astroid stage) ; 3, 4, Recession of chromosomes from the equator of the cell (diastroid) ; 5, nuclear spindle with chromosomes at each pole, and achromatiu threads between ; 6, Division of the cell completed. [Facing p. 102. CAUSES OF VARIATION 103 are interbred. As Prof. Cossar Ewart says * : " Domestic animals reproduce themselves with great uniformity if kept apart ; but the moment one mixed up two different races, strains, or breeds, one did something that was difficult to put in words, but the result was what has been best described as an ' epidemic ' of variations." On the other hand, Hatschek and others have pointed out that amphimixis acts as a check on variability, obviating heterogeneous idiosyncrasies. This was suggested even by Lamarck : "In repro- ductive unions the crossings between the individuals which have different qualities or forms are necessarily opposed to the continuous propagation of these qualities and these forms." Similarly Darwin said : " When species are rendered highly variable by changed con- ditions of life, the free intercrossing of the varying individuals tends to keep each form fitted for its proper place in nature." . Combinations of Chromosomes. — Prof. H. E. Ziegler has given much attention to the number of possible combinations of parental chro- mosomes in the offspring, supposing the distribution to be fortuitous. If the normal number of chromosomes in a species is n, the number of tetrad groups is - the number of possible combinations in th& mature germ-cells is — |- i, and the number of possible combinations (n \ a n* -+lj = — + w + !. If the normal number of chromosomes be 8 (as in the fluke often found parasitic in frogs, Polystomum integerrimum), the number of tetrad groups is 4, the number of possible combinations in the mature germ-cells is 5, and the number of theoretically different offspring is 25, i.e. on the assumption that the chromosomes are heterogeneous. But according to the laws of chance certain combinations are much more frequent than others ; the larger the number of tetrad groups the more frequent is the occurrence of an approximately equal number of paternal and maternal chromosomes in the germ-cell. Sutton puts the matter as follows. An individual receives from his father 4 chromosomes, A, B, C, D, and from his mother (an equal number) a, b, c, d. The immature germ-cell has A, B, C, D ; a, b, c, d. These group themselves in four tetrads, each composed of two double chromosomes, two maternal and two paternal, Aa, Bb, Cc, Dd. The mature germ-cell receives one chromosome from each * Discussion on Heredity in Disease, Scottish Med. and Surg. Journal, vi 1900, p. 308. 104 HEREDITY AND VARIATION tetrad, and there are 16 possible combinations — viz. a, B, C; D ; A, b, C, D ; A, B, c, D ; A, B, C, d ; a, b, C, D ; a, B, c, D ; a, B, C, d ; a, b, c, d ; and eight others which may be got by replacing small letters by capital letters and vice versa. The number of possibly different offspring would be i62. Sutton gives the following table, which is of some interest as suggesting the possibilities of variation. Normal Number Number of combina- Number of possi- number of of Tetrad- tions in the mature bilities in the chromosomes groups . germ -cells offspring 8 .. 4 .. 16 .. 256 12 6 .. 64 .. 4,096 16 .. 8 .. .. 256 .. 65,536 24 .. .. 12 .. . . 4096 16,777,216 Summary. — In certain moods biologists are accustomed to say that they do not know anything in regard to the causes of varia- tion. They imply that it is of the essence of living creatures to vary, that variability is a primary property of organisms. The sequence of generations is a life stream, changing as it flows. In other moods, however, biologists often point out how natural it is that organisms should vary. When the body of the parent is a-making, a lineage of germ-cells is started and the unspecialised descendants of these develop into offspring, which are on the whole like the parent because they are made of the same stuff. "True" twins developed from one ovum are usually almost facsimiles of one another. Why should not the offspring be a facsimile of the parent ? Sometimes, to our eyes, it is quite con- f usable with the parent, but this is not common. Why not ? i. It is common to point out that the germ-cell which is liber- ated to become an offspring is not likely to be identical with the germ-cell which developed into the parent. It has been sojourn- ing in the parent's body, exposed to a variable food stream and often to a variable complex environment, partly somatic and partly external. Is it likely to be exactly the same as the original germ-cell from which it is descended by continuous cell-division ? SUMMARY OF CAUSES 105 The experiments of Prof. W. L. Tower, in particular, suggest that important external changes may provoke changes in the germ-cells without necessarily affecting the parental body. He subjected full- grown potato-beetles (Leptinotarsa) to peculiar conditions of tem- perature and humidity during the time when the eggs were maturing, and found that " mutations " occurred in a certain proportion of the offspring. The parents were not affected, having passed the plastic stage; and some of ih^ eggs were not affected at all. Moreover, the same environmental peculiarity did not always produce the same mutation in the offspring. But what Tower's experiments forcibly suggest is this : that deeply saturating environmental changes may serve to pull the trigger of germinal variability. 2. It is also to be remembered that if the chromosomes stand in some definite causal relation to heritable qualities, as seems practically certain, then the maturation reduction of the chromo- somes to one half their original number offers an opportunity for variation. 3. It is likely that fertilisation or amphimixis — the intimate and orderly union of two sets of hereditary contributions which have often had very different histories — will promote variation. It is difficult to believe that it does not bring about new permuta- tions and combinations. 4. It is possible that variations may also arise in a less con- ceivable fashion — " bathmically," as the phrase goes — for un- known internal reasons. It is not absurd to suppose that the germ-plasm grows from generation to generation, and, in growing, changes — because it is its nature so to do. Apart from variation of internal origin and positive modifica- tion of external origin, we must remember that the offspring may differ from its parents through non-expression of certain items of its inheritance, the non-expression being due to the absence of the appropriate liberating stimulus. This kind of deviation may of course be obliterated next generation, when the full en- vironment allows the latent character to re-express itself. CHAPTER IV COMMON MODES OF INHERITANCE " Lord, I find the genealogy of my Saviour strangely checkered with four remarkable changes in four immediate generations. 1. Roboam begat Abia ; that is, a bad father begat a bad son. 2. Abia begat Asa ; that is, a bad father a good son. 3. Asa begat Josaphat ; that is, a good father a good son. 4. Josaphat begat Joram ; that is, a good father a bad son. I see, Lord, from hence, that my father's piety cannot be entailed ; that is bad news for me. But I see also, that actual impiety is not always hereditary ; that is good news for my son." — THOMAS FULLER, Scripture Observations, No. viii. § I. Though Prediction in Individual Cases is insecure, there are some Common Modes of Inheritance. § 2. Certain Necessary Saving Clauses. § 3. Blended Inheritance. § 4. Exclusive Inheritance (Unilateral, Absolutely Pre- potent, or Preponderant). § 5. P articulate Inheritance. § 6. Alternative Inheritance. § 7. Summary of Possibilities. ESPECIALLY among the lower animals, the offspring sometimes appear to us as if they were perfect reproductions of the parents, and we venture to speak of complete hereditary resemblance. Thus, in a crowd of Myriapods collected from one place at the same time, no individual peculiarities could be detected. A daughter-Hydra may be easily obtained which seems identical with the parent. A series of generations of green-flies or Aphides may be studied and no individual peculiarities discovered. 106 MATERIALS FOR STUDY 107 In other words, there seem to be cases in which generation succeeds generation without any variation. But there is every reason to suspect that in most cases the apparent absence of variation is illusory, and due to a lack of sufficiently intimate acquaintance with the individual organisms. The sheep which seem " all the same " to the careless eye are often known individually by the shepherd, and it is easy to demonstrate that the peas in one pod are often far from being alike. Similarly, the members of a group of individuals may seem " all the same " even to the naturalist's eye, but minute differences are soon detected by the expert who has devoted years to becoming intimately acquainted with that particular type. There are observable differences between sister-bees or ants, between the rooks from one clutch or the pigs from one litter. Even when there is only one parent — e.g. a self-fertilising liver-fluke or a parthenogenetic water-flea — there may be variations among the descendants. There is no doubt, however, that the range of variability differs greatly in different types, and it is obviously in cases where individual peculiarities are frequent and well marked that we can most hopefully study the relations of resemblance and difference between parents and offspring, or between the members of a series of generations. In horses and dogs, in sheep and cattle, in rats and mice, in rabbits and guinea-pigs, in pigeons and fowls, in butterflies and small, rapidly breeding crustaceans, in wheat and barley and maize, in peas and stocks, and in man himself, there is ample opportunity for studying the modes of inheritance. § I. Though Prediction in Individual Cases is insecure, there are some Common Modes of Inheritance When we are dealing with the generations ol d.n animal or plant in regard to which previous observation has shown us that the members of the species are strikingly uniform in their characters, we may venture with some security to predict that io8 COMMON MODES OF INHERITANCE the offspring of a pair will as usual exhibit more or less complete hereditary resemblance to their parents and ancestors. And yet this prediction may be falsified, for variations may suddenly crop up without known cause. Similarly, when we are dealing with the generations of a so-called " pure-bred " race of animals or plants, we may venture with some security to predict that the offspring of a pair will exhibit, as regards their more essential features, a large measure of complete hereditary resemblance to their parents and ancestors. And yet in individual cases this prediction also may be falsified; for no known reason a " freak " or " sport " may unexpectedly appear. When we consider the variable nutritive conditions of the germ-cells, the subtle processes of maturation and fertilisation, and the intricate nature of the environment appropriate to each development, we cannot be surprised that the result may often belie individual prediction. The possibly anecdotal instance, cited by Lucas, of the twin children of an Antillean negress — one white with long hair, the other black with woolly hair — may serve as a diagrammatic illustration. On the other hand, experience shows that, in spite of uncer- tainty in regard to individual cases, there is often perfect certainty as to the average results where we have to do with large numbers ; that the degree of resemblance to parents and ancestors is sometimes capable of precise prediction ; that in (particular sets of cases (Mendelian phenomena, see Chapter X.) I we can definitely predict how many of the offspring will be like I the parents, how many like one grandparent, how many like another ; and that, apart from such statistical generalisations, there are what we may call alternatives of expectation with varying degrees of probability. In other words, there are certain more or less well-defined modes of hereditary resemblance which occur very frequently. To explain and illustrate three of these is the pbject of this chapter. AN EXUBERANT TERMINOLOGY 109 A discussion of the different modes of hereditary resemblance is somewhat hampered by an exuberant terminology, and by the fact that different authors have sometimes used the same term in different ways. We read of inheritance being unilateral and bilateral, unisexual and bisexual, blended and conspired, neutralised and combined, direct and collateral, atavistic and progressive, and so on. We have tried to reduce this complex terminology to a minimum. This is the more justifiable since we cannot doubt that all the ordinary phenomena are of a piece, that many of the ordinary modes will be embraced eventually in one general formula — probably some modification of Galton's Law of Ancestral Inheritance, and that others will be embraced in Mendelian formulae. We propose, then, to restrict attention to three frequently occurring modes of hereditary resemblance, which are called blended, exclusive, and

" They are undoubtedly to be looked upon as reversions to extremely remote characters possessed by our lower mammalian forefathers." But it seems simpler to regard them as independent variations, comparable to many other ab- normal multiplications of parts. They happen to suggest bygone conditions, but that is probably all that we are warranted in saying. Polydactylism in man has been interpreted as a reversion to an 9 130 REVERSION AND ALLIED PHENOMENA ancestor with more than five digits ; but this is illegitimate, for the so-called " heptadactylous ancestor " is a pure myth. Polydacty- lism in man can only be called a reversion when there is in the family history a previous occurrence of the same abnormality some generations back. It occasionally happens that a particular part of the skin in man exhibits a mouse-like covering of close-set hair. To interpret this — a mere random variation — as a reversion is credulous in the ex- treme. It may also be noted, incidentally, that to call the wool-like covering of small hairs (the " lanugo ") on the human foetus a re- version to a hairy ancestor is quite absurd ; it is a normal stage in development quite outside the rubric of reversion. It may be an inheritance from a distant past, but it is no more a reversion than the occurrence of a notochord as a constant antecedent to the development of its substitute, the backbone. Similarly the dog's habit of turning round and round before it settles down to sleep may be interpretable in the light of past history, but it has nothing to do with reversion. " When horses are occasionally born at the present day in which one or two accessory toes are present on two or even all four feet, we are perfectly right in considering the development of these toes to be due to reversion to an ancestor of the Miocene period." That the modern horse which steps daintily on the tip of a single (third) toe for each limb, and has merely hidden rudiments of the second and fourth, has been evolved from a many-toed ancestor, is one of the most certain of evolutionist inferences, but are we " perfectly right " in interpreting the occasional development of supernumerary toes, as on Julius Caesar's horse, to the reassertion of latent ancestral items in the inheritance ? Is it not simpler to regard this as an independent variation, comparable to multiplications of other parts to which reversionary interpretations are inapplicable ? We must remember, also, that vestigial organs are in many cases peculiarly liable to vary. It ought not to be necessary to remark that the ancestor to whom the organism is supposed to revert must be real, not hypo- thetical. Some enthusiastic exponents of the reversion theory have not scrupled to name or even invent the ancestor to whom the IMPROBABLE CASES 131 peculiarity in question is supposed to be a reversion, although evidence of the pedigree is wanting. And the terribly vicious circle is not unknown of arguing to a supposed ancestor from the supposed reversion, and then justifying the term " reversion " by its resemblance to the supposed ancestor. Playing with biology can hardly go further than this! Moreover, the postulate of characters remaining latent (save for occasional more or less hypothetical reawakenings) for millions of years, is made as glibly as if it were just as conceivable as a throw-back to a great-grandfather. There are many reasons why it is absurd to describe a Cyclopean one-eyed human monster as a reversion to the one-eyed larval ascidian. One is that there is no warrant for believing that the ascidian type was in the direct line of our long pedigree. One of the diagnostic features of gout is the presence of uric acid in the blood, and its deposition in various tissues of the body (doubt- less helped by the frequently associated degeneration of the kidney, which is normally competent to filter out the normal nitrogenous waste-product, which is mostly in the form of urea). It is known, however, that reptiles, for instance, like many backboneless animals, normally excrete most or a large part of their nitrogenous waste in the form of uric acid. This has led even such an eminent pathologist as Prof. Hamilton (1900, p. 297) to say, "May we not entertain, as a possibility, that the gouty constitution, so-called, is in part a reversion to some far-back ancestor, in which uric acid was excreted normally to a much larger extent than it is at present in an average member of the human race ? " That is to say, the gouty person reverts to the physiological habit of a far-back ancestral organism (not even any known mammalian type), which had uric acid as a characteristic waste-product, but he does not, unfortunately, revert to the associated condition of having kidneys able to excrete the uric acid adequately. But our simple point is that the supposition of gouty tendencies lying latent in some form or other through literally millions of years taxes our imagination too severely. Such instances are almost sufficient to damn the reversion hypothesis altogether. 132 REVERSION AND ALLIED PHENOMENA § 5. " Skipping a Generation " It is often remarked in human inheritance that a child re-exhibits the peculiarity of a grandfather or grandmother, which the parents did not show, A Mendelian interpretation of this is in some cases possible. " If the two grandfathers have blue eyes and both grand- mothers brown eyes, then the parents may both have simplex brown eyes * ; they will both form germ-cells of which 50 per cent, have and 50 per cent, lack the determiner to form brown iris pigment. From such brown-eyed parents one child in four will have blue eyes like the grandfathers. This is atavism. Cases of atavism can, in general, be explained on the same ground as atavism to blue-eyed grandparents" (Davenport, 1910, p. 292). In case of sex-limited characters, such as bleeding or haemophilia, the phenomenon of " skipping a generation " may be illustrated. For the haemophilia is usually transmitted through unaffected daughters to grandsons. This may be comparable to other cases of sex-limited inheritance, e.g. in certain strains of sheep where the horns are confined to the males. It is likely that skipping a generation is less frequent than it is supposed to be ; thus features which the parent thinks he never had may have been plain enough when he was of the same age as his son now is. Moreover, in the case of characters that blend it is an obvious possibility that a grandson should sometimes show his grandfather's pattern. Finally, some cases of the disappearance of exceptional ability and the return to mediocrity come within the rubric of " filial regression." But our present point is that there seems little utility in calling " skipping a generation " a " reversion," or even an atavism. A drone-bee arises from an unfertilised egg ; it has a mother and two grandparents, but no father. But it seems rather absurd to call its resemblance to its grandfather either atavistic or rever- sionary. This is a reductio ad absurdum, for the drone-bee would resemble its father if it had one 1 * " Ordinarily when parents are similar, each unit character of the offspring develops from two similar determiners — one paternal and one maternal. Thus in its origin any unit character is duplex. When, however, the determiner is found in only one of the parents the character is simplex." This will be clearer after the chapter on Mendelism has been read, NEW VIEW OF REVERSION 133 § 6. Mendelian Interpretation of Reversion As we have already indicated, the number of alleged rever- sions has been greatly reduced by the results of the study of Mendelian inheritance. An interesting re-interpretation of " re- versions " has been supplied. Some red guinea-pigs, as Castle has shown, produce in crosses with a black race the " agouti " type of coat found in all wild guinea-pigs, and various experiments prove that this is due to the coming to- gether of three colour-factors — simple red, simple black, and a third which is carried by the red but can become visible only in the presence of both black and red. In certain instances, which are quite well denned by the Mendelian experimenters, a cross between a black and an albino mouse, or be- tween a black and an albino rabbit, results in a complete reversion to the wild grey form. Crosses between the tall, upright, bush-like " Bush " sweet-pea and the dwarf prostrate " Cupid " variety resulted in a procumbent plant with long internodes, like the wild type that is found growing in Sicily. In these and in similar cases it has been possible by various experimental tests to give convincing proof that the reversion is a re-synthesis of characters that had been analysed apart. As Prof. R. C. Punnett concludes : " Reversion, therefore, in such cases we may regard as the bringing together of complementary factors which had somehow in the course of evolution become separated from one another " (1911, p. 54). § 7. Reversion in Crosses False Reversion or Yicinism.— In his criticism of cases which have been labelled " reversions," De Vries draws a sharp dis- tinction between "true re version/ 'due to unknown internal causes which induce long-lost latent ancestral characters to assert themselves, and " false atavism or vicinism," which is due to crossing. His investigation of a large number of cases led him 134 REVERSION AND ALLIED PHENOMENA to conclude that " true atavism, or reversion caused by an innate latent tendency, seems to be very rare," and that most of the botanical instances are due to crossing. He calls this false reversion " vicinism," as indicating the sporting of a variety under the influence of others in its vicinity. " Crossing and pure variability are wholly distinct groups of phenomena, which should never be treated under the same head, or under the same name." He does not deny in any way the numerous " rever- sions " which gardeners describe ; he simply points out (with much circumstantial evidence to warrant his contention) that nearly all these ordinary " reversions " are due to crosses. He shows, for instance, how a famous case, the reversion of the " Tuscarora " variety of American corn cultivated by Metzger in Baden, may be readily interpreted as a typical instance of vicinism. Why .the offspring of hybrids should revert to the parental type is another question, to which we shall return in the chapter on Mendelism (Chapter X.). Two white-flowered sweet-peas are crossed, and the result is a progeny with the wild, purple flowers. Two smooth stocks are crossed, and the result is a progeny with the original hoary, ancestral type. These cases are what Darwin called " re- version on crossing." But, as Mr. Bateson says, " such reversion is nothing but the meeting of two parted complementary elements, which have somehow been separated by variation." Thus it is possible that many so-called reversions may be simply Mendelian phenomena in disguise. § 8. Reversion of Retrogressive Varieties Within a species it is often possible to distinguish several subspecies or " elementary species " (De Vries), which differ from one another in many characters affecting many organs. Thus in the species called Draba verna, or whitlow grass, there are two hundred or so minor groups, like constellations within RETROGRESSIVE VARIETIES 135 constellations. But the species may also include " varieties," more or less sharply distinguished from the rest of the species by the apparent absence of some notable specific feature, or, more rarely, by the acquisition of some peculiarity already seen in closely allied species. They stand aside, as it were, like far out- lying parts of the constellation. " Varieties," thus defined, usually differ from their parent species in a single sharp character only, or in several correlated characters ; they usually arise in a negative way by the apparent loss of some quality ; and they have great stability. They are comparable to the familiar colour- varieties in rabbits, guinea-pigs, mice, etc., which seem to arise by the dropping out of part of the ancestral equipment of char- acters. They are in no sense reversions. Illustrations. White " varieties " of red and blue flowers — e.g. of red-flowering currant. Smooth " varieties " of hairy plants — e.g. nectarine (from peach). Smooth "varieties " of prickly plants — e.g. holly and gooseberry. Rayless "varieties" of many composites normally with ray florets — e.g. white marigold, camomile, daisy. Radiate " varieties " of many composites, normally with no ray-florets — e.g. tansy and groundsel. Red "varieties " of white flowers — e.g. hawthorn. Red " varieties " of green trees and shrubs — e.g. beech and birch. Weeping " varieties " of ash, willow, etc. Starchless seeds — e.g. sugar-corn. Seedless fruits — e.g. banana and mandarin orange. Mr. Burbank's stoneless plum. As these varieties are most frequently in a negative direction having apparently lost some character which their parent-species possesses, De Vries includes most of them in the term "retrograde varieties." Perhaps "retrogressive varieties " would be a clearer term. They usually breed true, but some of them are perpetuated asexually — e.g. of course, the seedless fruits. Sometimes, however, the apparently lost ancestral character re-appears, as when the smooth nectarine, a " variety " of peach, becomes 136 REVERSION AND ALLIED PHENOMENA downy, or when the white-flowering currant puts forth red flowers. Such cases may be described as reversions to the specific type, and they can be intefpreted only in two ways. Either we have to do with new variations which happen to hit the old mark, or, as seems more probable, latent ancestral characters have re-asserted themselves. It is a current belief that these " varieties " have a strong tendency to " revert " to the parent species, but, according to De Vries, this is, as regards pure varieties, not of hybrid origin, and ordinarily propagated by seeds, a popular delusion. " In the present state of our knowledge it is very difficult to decide whether or not true reversion occurs in constant varieties. If it does occur it surely does so very rarely, and only under unusual circumstances, or in particular individuals " (1905, p. 155). It must be noticed, however, that De Vries distinguishes true reversion (due to a spontaneous germinal change) from false reversion which is induced by hybridising. In illustration of the constancy of varieties he cites the wide- spread rayless form of the wild camomile (Matricaria chamomilla discoidea), which is so constant that many botanists have made a species of it. De Vries raised in successive years between 1,000 and 2,000 seedlings, but observed no trace of reversion. Similarly, the rayless " variety " of the common tansy ragwort (Senecio jacobcea) is quite as stable as the radiate species. De Vries also refers to the stability of white strawberries, green grapes, white currants, crisped lettuce, crisped parsley, smooth spinach, white flax, sugar-corn, and strawberries without runners. Seed- re version very Rare. — Excluding cases where it is doubtful whether the variety has not a hybrid origin, and is therefore liable to the peculiar phenomenon known as the splitting up of hybrids, excluding also all cases of "sporting varieties," where an apparent reversion might be a mere coincidence in the crowd of variations, De Vries concludes that "seed-reversions must be said to be extremely rare. , ? , INSECURE INTERPRETATIONS 137 It would be bold indeed to give instances of seed-atavism, and I believe that it will be better to refrain wholly from doing so. ... It is by far safer in the present state of our know- ledge to accept bud- variations only as direct proofs of true atavism. And even these may not always be relied on, as some hybrids are liable to split up in a vegetative way, and in doing so to give rise to bud- variations that are in many respects apparently similar to cases of atavism " (1905, p. 176). § 9. Interpretations in Terms of Reversion As in many other cases, one of the difficulties in regard to the reversion theory is that in terms of it much can be interpreted and relatively little demonstrated. In regard to the origins of domesticated animals and cultivated plants, we remain in great obscurity. In regard to the actual pedigree of wild species our ignorance is even greater. Thus, while it is often easy to interpret an unexpected variation as a reversion to a plausible ancestral type, we have little security in so doing. Thus De Vries distinguishes between experimentally demon- strable reversion and what he calls " systematic atavism," where the ancestral type is merely presumed to be so-and-so on the basis of taxonomic considerations. It is probable that the common ancestors of the " elementary species " (Primula officinalis, P. elatior, and P. acaulis), which make up the systematic species of primrose, Primula vera, were " perennial plants with a rootstock bearing their flowers in umbels or whorls on scapes. Lacking in Primula vera> these scapes must obviously have been lost at the time of the evolution of this form." But in the common acaulescent " elementary species," P. acaulis, a scape sometimes develops. It may be reasonably interpreted as due to the re-vitalising of a dormant scape -character inherited from the presumed ancestor. "Simi- larly with the appearance of bracts in the usually bractless 138 REVERSION AND ALLIED PHENOMENA Crucifers, and with the unexpected appearance of upright tomatoes. Similarly, the twisted teasels lose their decussation, but in doing so the leaves are not left in a disorderly dispersion, but a distinct new arrangement takes its place, which is to be assumed as the normal one for the ancestors of the teasel family." § 10. Further Examples of Reversion In one of Prof. Cossar Ewart's experiments a pure white fantail cock pigeon, of old-established breed, which in colour had proved itself prepotent over a blue pouter, was mated with a cross previously made between an owl and an archangel, which was far more of an owl than an archangel. The result was a couple of what were, theoretically, fantail-owl-archangel crosses, but the one resembled the Shetland rock-pigeon, and the other the blue rock of India. Not only in colour (slaty-blue), but in shape, attitude, and movements there was an almost complete reversion to the form which is believed to be ancestral to all the domestic pigeons. The only marked difference was a slight arching of the tail, but there were only twelve tail- feathers, as in the rock- dove, whereas the father fantail had thirty. A dark bantam hen, crossed with an Indian game Dorking cock, produced amongst others a cockerel almost identical with a jungle fowl (Callus bankiva) — i.e. with the original wild stock (Ewart). Similarly, in his horse-zebra hybridisations, Ewart obtained forms whose stripings were at least plausibly interpreted as reversions to an extremely old type of horse, such as is hinted at in the striped ponies of Tibet. A smooth-coated white rabbit, derived from an Angora and a smooth-coated white buck, was mated with a smooth- coated, almost white doe (grand- daughter of a Himalaya doe), with very interesting results, significant qf the complexity of FURTHER EXAMPLES 139 the conditions. In the litter of three, one was the image of the mother, one was an Angora like the paternal grand- FIG. 26. — Varieties of domestic pigeon arranged around the ancestral rock-dove (Columba livia). (Based on Darwin's figures.) mother, and the third was a Himalaya like the maternal great- grandmother, 140 REVERSION AND ALLIED PHENOMENA For all these cases, except that of the horse-stripings, as also for similar cases given by Darwin, Mendelian interpreta- tions are now forthcoming, and the hypothesis of the re-assertion of long latent ancestral characters is unnecessary. When the swimming-bell or medusoid Epenthesis folleata appears with pentamerous symmetry instead of the usual arrangement of its organs in fours or multiples of four, no one would dream of calling this discontinuous .variation an instance of reversion, for we only know of one medusoid (Pseudoclytia pentata) where five is normally the ruling number (Mayer, 1901). But when the last-named medusoid occurs with four oral lips, as it occasionally does, it may be said that this variation is reversionary, since there is good reason to believe that Pseudo- clytia pentata is a pentamerous derivative of the Epenthesis stock. Even in this case the interpretation of the four lips as reversionary may not be correct, since, as a matter of fact, the number of lips in Pseudoclytia varies from one to seven. Reversion in Parthenogenesis. — Weismann (1893, p. 344) reports a very interesting case which he observed in varieties of a small Ostracod crustacean (Cypris reptans) which multiplies parthenogenetically. In the course of observations extending over eight years he found that, amidst the expected uniformity of resemblance between parent and offspring, exceptions occa- sionally occurred. These were of such a nature that he could only interpret them " as exhibiting reversions to an ancestral form many generations removed." OTHER INSTANCES OF REVERSION White- flowering Currant. — The white-flowering variety of the red-flowering currant (Ribes sanguineuni) is said to have originated many years ago from seed in Scotland. " Occasionally this white- flowered currant reverts back to the original red type, and the reversion takes place in the bud. . . . Once reverted, the branches remain for ever atavistic. It is a very curious sight, these small OTHER INSTANCES OF REVERSION 141 groups of red branches among the many white ones " (De Vries, 1905, p. 167). This case is peculiar, however, because the white variety is propagated only by cuttings or grafting. "If this is true, all specimens must be considered as constituting together only one individual, notwithstanding their wide distribution in the gardens and parks of so many countries. This induces me to sup- pose that the tendency to reversion is not a character of the variety as such, but rather a peculiarity of this one individual " (p. 168). Wheat-ear Carnations. — Large beds of carnations sometimes show peculiar anomalous forms known as " Wheat-ears,** with small green ears instead of flowers. There has been a loss of flowers and a multiplication of bracts. On a specimen of this De Vries observed that some branches reverted wholly or partially to the production of normal flowers. " The proof that this retrograde modification was due to the existence of a character in the latent state, was given by the colour of the flowers. If the reverted buds had only lost the power of producing spikes, they would evidently have returned to the characteristics of the ordinary species, and their colour would have been a pale pink. Instead of this, all flowers displayed corollas of a deep brown. They obviously reverted to their special progenitor, the chance variety from which they had sprung, and not to the common prototype of the species " (1905, p. 229). A Picturesque Case. — The long-headed green dahlia originated twice from two different double-flowered varieties — a deep carmine with white tops on the rays, and a pale orange known as " Sunrise." They were quite sterile and were progagated asexually, one in Prof. De Vries's garden, the other in the nursery at Haarlem, where both arose. " In the earlier cultures both remained true to their types, never producing true florets. No mark of the original differ- ence was to be seen between them." But in 1903 both reverted to their prototypes, and bore ordinary double flower-heads. " Thus far we have an ordinary case of reversion. But the important side of the phenomenon was, that each plant exactly ' recollected ' from which parent it had sprung. All of those in my garden re- verted to the carmine florets with white tips, and all of those in the nursery to the pale orange colour and the other characteristics of the ' Sunrise ' variety " (1905, p. 231). It seems impossible not to admit that characters of the parent-varieties had lain for a time latent and had eventually reasserted themselves. 142 REVERSION AND ALLIED PHENOMENA CONCLUSION. — In his Locksley Hall Sixty Years After Tennyson spoke of — Evolution ever climbing after some ideal good, And Reversion ever dragging" Evolution in the mud ; but this is making a bogey of reversion. Many of the phenomena commonly labelled as " reversions " are wrongly labelled, and true Reversion does not seem to be of frequent occurrence. Moreover, when it does occur, it may mean, not a deterioration, but a return to a position of greater organic stability. What acts as a drag or brake — often advantageously— on progressive variation is not so much reversion as filial regression. But the great step of progress that has been made of recent years is due to the Mendelian experimenters who have shown that many of the reversions which follow crossing are due to the re-combination of complementary factors which had become separated in the course of domestication and cultivation. Wherever this can be shown there is, of course, no warrant for the hypothesis that reversion is due to the sudden activation of a long latent ancestral character. But this hypothesis may be in the meantime retained for any cases that appear to demand it, CHAPTER VI TELEGONY AND OTHER DISPUTED QUESTIONS "The mysterious wireless telegraphy of ante-natal life." — J. W. BALLANTYNE. § I. What is meant by Telegony. § 2. The Classic Case of Lord Morton's Mare. § 3. Representative Alleged Cases of Telegony. § 4. E wart's Penycuik Experiments. § 5. Suggestions which explain away Telegony. § 6. Suggestions as to how Telegonic Influence might be effected. § 7. A Statistical Suggestion. § 8. The Widespread Belief in the Occurrence of Telegony. § 9. An Instructive Family History § 10. A Note on Xenia. § ii. Maternal Impressions. § i. What is meant by Telegony THE term "telegony" is applied to doubtful, certainly rare, but, if true, very remarkable cases where an offspring resembles a sire which, though not its father, had previously paired with its mother. More theoretically expressed, telegony is the supposed influence of a previous sire on offspring subsequently borne by the same female to a different sire. The ovum or the embryo is supposed to be influenced by the mother's previous impregna- tion or by the consequences thereof 143 144 T&LEGONY To take a simple instance, the racehorse Blair-Athol had a very characteristic blaze or white bald face, and it is said that mares which had once borne foals to Blair-Athol subsequently produced to quite different stallions foals which exhibited the Blair-Athol blaze. It is very generally asserted by dog-breeders that if a thorough-bred bitch has had pups to a mongrel, her value is greatly decreased, for she will not afterwards breed true. The alleged phenomena are of much interest, but the evidence of their actual occurrence is far from satisfactory, and their theoretical interpretation in terms of telegony is beset with physiological difficulties. But as a belief in telegony is still widespread, it will not be unprofitable to consider (a) the alleged facts, and (b) the interpretations suggested. § 2. The Classic Case of Lord Morton's Mare The classic case, given by Lord Morton (1821), is thus sum- marised by Darwin : " A nearly purely bred, Arabian, chestnut mare bore a hybrid to a quagga ; she was subsequently sent to Sir Gore Ouseley, and produced two colts by a black Arabian horse. These colts were partially dun-coloured, and were striped on the legs more plainly than the real hybrid, or even than the quagga. One of the two colts had its neck and some other parts of its body plainly marked with stripes. Stripes on the body, not to mention those on the legs, and the dun-colour, are extremely rare — I speak after having long attended to the subject — with horses of all kinds in Europe, and are unknown in the case of Arabians. But what makes the case still more striking is that the hair of the mane in these colts resemWed that of the quagga, being short, stiff, and upright. Hence there can be no doubt that the quagga affected the character of the offspring subsequently begot by the bUck Arabian horse " (Darwin, 1868, vol. i. pp. 403-4). THE CASE OF LORD MORTONS MARE 145 In 1823 the mare had again a foal by an Arab stallion, and this also showed some quagga characters. It may well be asked : If this was not telegony, what was it ? But the case is not quite so satisfactory as it seems. Settegast * remarks that the drawing made of the foal with the alleged quagga characters merely shows indistinct dark stripes on the neck, withers, and legs, and that similar stripes not uncommonly occur on pure-bred foals. A stiff mane may also occur as a variation in horses. It is possible that the alleged quagga-like characters had nothing to do with the original quagga sire, but were reappearances of latent ancestral characters. Sanson (1893) sets another case against Lord Morton's. A bay mare had by two different stallions seven foals of a uniform colour, and then by a third stallion a foal more zebra-like than Lord Morton's. To which Delage adds that this eighth foal was pommelled grey — a colour with which zebra-like stripes are not infrequently associated. Cornevin cites a breeder from the Pyrenees to the effect that a mare served by an ass and producing a mule was thereafter served by a horse and cast a foal which had hoofs more mule- like than horse-like. But this is too vague to be of much use, and besides, " asinine " variations sometimes occur in horses where there has been no hybridising (Sanson, 1893). Moreover, the opposite result has been often obtained. Sette- gast (1888) gives the case of four stud mares which were served by asses and bore mules. They were subsequently served by horses, and the foals showed no asinine traits. § 3. Representative Alleged Cases of Telegony Man. — Herbert Spencer cites from Flint's Human Physiology (1888) the case of a white woman who had intercourse with a negro and afterwards with a white man. There were some negro-pecu* * Thierzucht, Berslau, Bd. i. 1878, pp. 223-34. id i46 TELEGONY liarities in the children by the second male. But it is perhaps enough to say that it is difficult to get at the truth in such cases. Cornevin (1891, p. 356) gives the following case. The widow ot a hypospadic man had by a second and normal husband four hypo- spadic sons, two of whom transmitted the abnormality (Lancet, 1884). But in a case like this we require further particulars — e.g. as to the normality of the mother, and as to any tendency to hypospadism both in her ancestry and in that of her second husband. Cornevin also cites the case of a woman married to a deaf-mute, by whom she had one deaf-mute child. By a second normal husband she had a deaf-mute child, and then others who were normal (Ladreit de Lacharriere, in preface to Goguillot's Comment on fait parler les sourds-muets, Paris, 1889). But here again it is necessary to know whether there was any tendency to deaf-mutism on the mother's side or in the ancestry of her second husband. Dogs. — It is the deeply rooted opinion of dog-breeders — doubtless resting on a basis of experience, though it may be misinterpreted experience — that a bitch of good stock once lined by a mongrel is spoilt for further prize-breeding. It is said that many valuable bitches have been sacrificed because of this deeply rooted opinion. The following case is cited by Cornevin (1891, pp. 356-7), from Kiener (1890). An Artesian bitch was first lined by a wall-eyed mastiff, and afterwards by an Artesian dog. Among the pups born to the latter one was wall-eyed. One requires to know how fre- quently a wall-eyed variation crops up, and whether there was any occurrence of it in the ancestry of the mother or of the second male. Darwin (1868) gives the case of a hairless Turkish bitch which was lined by a spaniel, and had some hairless pups and sonic with short hair. She was subsequently paired with a hairless Turkish dog, but the offspring were as before. It must again be asked whether there may not have been some spaniel strain in the previous ancestry. Spencer (1893) tells of a Dachshund bitch which was paired with a collie and had a hybrid litter. The following year she bore to a Dachshund a similar hybrid litter. But we require to know how thoroughly pure-bred the Dachshund mother and father were. Perhaps the most useful comment on the cases of reported telegony in dogs is that made by Prof. Cossar Ewart (1901): "When it is remembered that we are surprisingly ignorant of the origin of the various breeds of dogs, and that, however pure the breed, reversion ALLEGED CASES 147 to a former ancestor may at any moment occur, it will, I think, be admitted that, for the purpose of testing the ' infection ' doctrine, the dog, of all our domestic animals, is the least satisfactory." Mr. C. H. Lane, discussing toy spaniels in his book, All about Dogs, says, " I have been told by breeders that they have had in one litter a specimen of all four breeds [i.e. of King Charles, Prince Charles, Blenheim, and Ruby spaniels]. In the same way rough and smooth terriers often occur in the same litter, not because of infection, but because of reversion." Cats. — Dr. H. de Varigny tells of a normal cat which, after pro- ducing kittens to a Manx cat, had several tail-less kittens to an or- dinary cat ( Journal desDebats, September 9th, 1 897 ; cited by Ewart, 1901 ). But the mother, or the second father, or both, may have had a tail-less ancestor, to which some of the kittens happened to revert. Or even if there were no such ancestor, the tail-lessness may have been merely a variation that happened to coincide with the pecu- liarity of the first sire, but was not in any way due to him. For tail-lessness is not a very rare " sport." As a counter-case, Prof. Ewart refers to " a pair of young cats, of a somewhat peculiar variety, obtained from Japan. These cats belonged to a small breed, bluish in colour, with the exception of the ears and extremities, which were black. When the female grew up she first had kittens to a common tabby cat. These kittens showed the characteristic tabby markings. Her next kittens were by her Japanese mate, but in no respect did they suggest the previous tabby- coloured mate. No better experiment than this could be made with cats. The imported breed was quite distinct, and yet not sufficiently prepotent to swamp the common domestic English cat. Yet, though the first litter was sired by a common tabby, there was no indication whatever of the previous tabby mate in her second and pure-bred litter." (Case cited by Sydney Villar, F.R.C.V.S., Proc. Nat. Vet. Assoc. 1900, p. 130.) Sheep. — Dr. Alexander Harvey, in a paper " On a Curious Effect of Cross-breeding " (1851), gives on the authority of W. McCombie of Tilliefour, Aberdeenshire, the following case : Six pure-bred black-faced horned ewes were put, in the autumn of 1844, some to a Leicester ram (white-faced and polled), and others to a Southdown ram (dun-faced and polled), and produced cross- bred lambs. In the autumn of 1845 the same ewes were put to a pure black- 148 TELEGONY faced horned ram of their own breed. The lambs were all polled and brownish in the face. In the autumn of 1846 the ewes were again put to another fine ram of their own breed. Again the lambs were mongrels, but not so markedly as before. Two were polled and dun-faced, with very small horns ; while the other three were white-faced, with small round horns. At length the owner parted with his ewes without getting from them a single pure-bred lamb. Perhaps, however, the ewes were not so pure-bred as was supposed. Cornevin cites from Magne the statement that white ewes, first crossed by black rams and then by white rams, bear to the latter, lambs which are piebald or which have blackish eyelids, lips, and limbs (Magne, J. H., Hygiene veterinaire appliquee, p. 206). But black variations are common even when no black rams have been used for several generations. Cattle. — Weismann (1893, p. 385) refers to a case reported by Carneri. A cow of a dark grey Miirzthal herd was put to a " light- coloured Pinzgau bull " ; it bore a calf with the characteristic brown and white patches of the Pinzgau breed, as well as with dis- tinct traces of the dark grey Miirzthal cross. It was subsequently served by a Miirzthal bull, and the second calf, while for the most part grey, showed " large brown spots like those of the Pinzgau breed." But this case is also inconclusive, since it is possible, as Carneri admitted, that " a drop of Pinzgau blood " may have pre- viously got into the Miirzthal herd without his being aware of it. Pigs. — Another circumstantial case cited by Darwin is that of a sow of Lord Western's black-and-white Essex breed, which Mr. Giles put first to a deep chestnut wild boar and after a time to a boar of the black-and-white breed. The offspring of the first union showed the characters of both parents, but in some the chestnut colour of the boar prevailed. From the second union the sow produced some young plainly marked with the chestnut tint, which is never shown by the Essex breed (Darwin, 1868, vol. i. p. 404). Rodents. — Breeders of rabbits, rats, and mice have sometimes reported phenomena which suggest telegony ; but the great varia- bility of these rodents makes them very unsuitable subjects of ex- periment. Prof. Cossar Ewart refers to two cases. Mr. C. J. Pound, bac- teriologist to the Queensland Government, " crossed a grey rabbit with a grey-and-white buck, and then mated her with a black buck E WART'S PENYCUIR EXPERIMENTS 149 with the result that in the second litter there were grey-and-white as well as grey-and-black young. Again, a female black rat after breeding with a pure white rat produced, to a brown rat, white, brown, and piebald offspring. . . . Had Mr. Pound made a number of control experiments he would doubtless have discovered that black female rats sometimes yield to a brown rat white, brown, and piebald offspring, without having been first mated with a white rat, and that grey doe rabbits often produce to a black buck grey- and-white as well as grey-and-black young." Experiments on rats and rabbits made by Dr. Bond (Trans. Leicester Literary and Philosophical Society, vol. v. October, 1899) yielded no results which could not be readily interpreted as due to reversion and other forms of variation. Birds. — A case of supposed telegony in birds is referred to by Darwin (1868, vol. i. p. 405) : "A careful observer, Dr. Chapuis, states (Le Pigeon Voyageur Beige, 1865, p. 59) that with pigeons the influence of a first male sometimes makes itself perceived in the succeeding broods ; but this statement, before it can be fully trusted, requires confirmation." Mr. Frank Finn, in a paper entitled " Some Facts of Telegony " (Natural Science, iii., 1893, pp. 436-40), cites a number of cases which seem to him to afford evidence of telegenic phenomena in birds, but they are not convincing. From the above citations it appears that the evidence of the occurrence of telegony is in great part, at least, of the same un- satisfactory character as that adduced in favour of use-inheritance — largely anecdotal, impressionist, and uncriticised. The need for careful experiments like those begun by Prof. Ewart (1896) is obvious. § 4. Ewart's Penycuik Experiments. The position of affairs being that a number of great authorities — e.g. Darwin and Spencer — had expressed their belief in the occurrence of telegony, and that a number of equally competent authorities had expressed themselves extremely sceptical on the subject, Prof. Ewart resolved on definite experiment — the only secure path. In general terms, he made a number of experiments likely 150 TELEGONY to give telegony the best possible chance of declaring itself, and although he has displayed his scientific mood in abstaining from dogmatic conclusion, and in suggesting many other experiments which should be made, there is no ambiguity in his verdict that the evidence of any undoubted telegony is very unsatisfactory. The Penycuik experiments proved this, at least — that telegony does not generally occur, even when what were considered to be favourable conditions were secured ; indeed, anything sug- gestive of telegony occurred only in a very small percentage of cases. Moreover, where peculiar phenomena of inheritance were observed, they seemed to be readily explicable on the reversion hypothesis. The general nature of the experiments may be understood by taking one of the best cases, which loses much, however, when summarised apart from the beautiful pictures illustrating the book (Ewart, 1899). A Rum pony mare, Mulatto, of remarkably pure breed, was served by a Burchell zebra stallion, Matopo, and the result in August, 1896, was Romulus, whose markings were quite different from those of his sire, being suggestive rather of the Somaliland zebra. In 1897 Mulatto had a bay colt foal to a grey Arab stallion, and this foal — unfortunately short- lived— gave no proof of telegony. The stripes which most frequently occur in horses were absent ; there were others which are not uncommon in horses ; but the most distinct markings (not that any were strongly developed) — namely, those across the croup — were of a sort extremely rare in both foals and horses. In short, the markings of Mulatto's second foal were puzzling, but in no definite way suggestive of the influence of the previous zebra sire. In this, as in the other cases, the verdict as to the occurrence of telegony was (C non-proven." In regard to experiments it should be remembered, however, that if telegony (supposing it to be a fact) be due to some strange persistence or unusual influence of the spermatozoa of a previous sire, then many isolated cases with negative results do not prove SUGGESTED EXPLANATIONS 151 much. As Pearson observes (1900, p. 462), " should it occur once in a hundred trials we are hardly likely just to hit upon the successful instance." § 5. Suggestions which explain away Telegony (a) It has been repeatedly suggested, by those who do not believe in the reality of telegonic influence, that the phenomena are simply illustrations of reversion. A normal cat has kittens to a Manx cat, and afterwards to a normal cat. In the second litter some are tail-less. " It does not follow, however, that some of the subsequent kittens were tail-less because their dam had been previously mated with a cat of the Manx breed. . . . The most likely explanation is that tail-less individuals occurred in the ancestry of one or both of the parents ; in other words, the absence of the tail is due to reversion to an ancestor " (J. Cossar Ewart, Trans. Highland and Agricultural Society of Scotland, 1901). This view amounts to denying telegony in the strict sense. We are asked to believe that there is no causal nexus between the previous sire and the subsequent offspring who resemble him. They happen to resemble him because he resembled one of their ancestors. This seems to us easier than believing in telegony. The plausibility of this explanation will vary in different cases. Thus Finn points out that the occurrence of feather-legged fowls in a pure Dorking breed, or of polled lambs from black-faced horned ewes, cannot be set down to reversion, " feather-legged fowls and polled sheep not being ancestral types." (b) It has also been suggested that the subsequent offspring have accidentally varied in the direction of resemblance to the previous sire. The resemblance is a mere coincidence. As the reliable facts are few and far between, there is much to be said for this view. 152 TELEGONY (c) Another suggestion explains away the alleged facts of tele- gony by referring them to maternal impression, the supposition being that the mental image, etc., produced in the mother by the first sire exerts an influence on subsequent germs or on their development after fertilisation by another sire. There is little to be said in favour of this interpretation ! § 6. Suggestions as to how a Telegonic Influence might be effected (a) It is well known that in most European bats sexual union usually occurs in autumn, but the spermatozoa are simply stored in the uterus, ovulation and fertilisation taking place in spring after the winter sleep. A somewhat similar retention of stored spermatozoa, which become operative long after impregnation, is familiar in insects : thus, in some queen bees the store has been known to last for two or three years, and Sir John Lubbock gives the remarkable instance of an aged queen ant which laid fertile eggs thirteen years after the last union with a male. From a consideration of such facts the suggestion has emerged that the second offspring are really fertilised by persistent spermatozoa derived from the first sire. Weismann (1893, p. 385) suggests the possibility that " sper- matozoa had reached the ovary after the first sexual union had occurred, and had penetrated into certain ova which were still immature." When these ova mature amphimixis might occur, and coincide in time with a second coitus to which the subse- quent offspring would be ascribed. But were this the explanation, we should sometimes find, as Weismann remarks, that offspring were produced without any second sire at all. No such phenomenon is known among higher animals. Moreover, there is no warrant for supposing that spermatozoa can persist as such through a period of gestation. " There is abundant evidence," Prof. Cossar Ewart says, " that in the SUGGESTED EXPLANATIONS 153 rabbit, as in other mammals, unused sperms lose their fertilising power and disintegrate long before the period of gestation comes to an end." For these two reasons the above interpretation may be rejected. (b) Somewhat subtler is the suggestion — often also called the " infection hypothesis " — that although the sperms of the first sire cannot be supposed to persist and fertilise ova discharged long afterwards, yet it is conceivable that the disintegrated substance of the sperms may persist and influence the ovaries and the ova, or that the sperms may exert an influence which does not amount to fertilisation. So great a physiologist as Claude Bernard seems to have believed in the possibility of such an influence, though it is somewhat suggestive of the " aura seminalis " of the ancients. In this connection, however, Cornevin recalls the facts that a turkey-cock's impregnation of the female suffices for the score or so of fertile eggs which are laid during the season, and that the common cock's act suffices for seven or eight eggs. .In both cases the fertile eggs are succeeded by other " clear " eggs, which are incapable of developing, and Cornevin asks whether we can believe that there is a brusque separation between the two sets, or whether the first at least of the " clear " set may not illustrate this supposed partial fertilisation. Romanes also suggested that the supposed effect was due to an absorption by the eggs of surplus sperm-material. (c) Another slightly different suggestion is that the surplus sperms derived from the first sire exert a physiological influence on the constitution of the mother, such that subsequent gestations are affected. Perhaps no one will deny that the male may in this way affect the constitution of the female, and Brown- Sequard's experiments on injections of spermine or testicular extract may be recalled in this connection ; but it is difficult to conceive that the influence should be of so precise a nature as to 154 TELEGONY evoke, for instance, the alleged quagga mane and quagga stripes in the second foal of Lord Morton's mare. Baron compares this supposed influence to the influence of pollen upon fruit (see § 10), and Darwin says that this analogy " strongly supports the belief that the male element acts directly on the reproductive organs of the female " (Darwin, 1868, p. 405). But no specific effect on the female animal has ever been demonstrated. (d) Perhaps the most plausible theory is that the mother is influenced through the foetus during pregnancy, and that the influence re-acts on subsequent offspring. On this so-called " saturation hypothesis " the suggestion is that the characters of the sire, while expressing themselves in the unborn embryo, also saturate into the dam and affect her constitution in such a precise way that her offspring by subsequent sires may through maternal influence acquire (or inherit ?) some of the character- istics of the first. Thus Sir William Turner (1889), in dis- cussing Lord Morton's case, says, " I believe that the mother had acquired, during her prolonged gestation with the hybrid, the power of transmitting quagga-like characters from it, owing to the interchange of material which had taken place between them in connection with the nutrition of the young one. . . . In this way the germ-plasm of the mother, belonging to ova which had not yet matured, had become modified whilst still lodged in the ovary. This acquired modification had influenced her future offspring, derived from that germ-plasm, so that they in turn, though in a more diluted form, exhibited zebra-like markings." Similarly, Cornevin (1891) asks, may not the foetus have in its blood special properties derived from the father, and may not these act like a vaccine on the blood of the mother ? The blood of the mother, thus affected, will act on the ova subsequently fertilised by another sire (Cornevin, 1891, p. 359). So also Harvey, 1851. A similar hypothesis has been suggested to explain A STATISTICAL SUGGESTION 155 certain facts connected with the so-called transmission of syphilis. This view did not, however, commend itself to Darwin, for he says (1868, vol. i. p. 405) : " It is a most improbable hypothesis that the mere blood of one individual should affect the repro- ductive organs of another individual in such a manner as to modify the subsequent offspring." He also points out that this hypothesis would not apply to telegony in birds, which has been alleged, though denied by Harvey and still requiring confirma- tion (Darwin, 1868, vol. i. p. 405). It is conceivable that something like the " saturation " above indicated may occur in a case of a poison or protective anti- toxin, which might diffuse in and out. We can imagine that a sire in- fected with some virulent disease, and showing certain structural disturbances associated therewith, may have offspring which are similarly affected, and that the influence from them may pass before their birth into the constitution of the mother, and so affect her that subsequent offspring by a healthy sire are diseased after the manner of the first. But while we have some facts to go upon in regard to the diffusion of toxins and anti-toxins, we have none as yet which justify us in supposing the diffusion of structural characteristics or of representatives of these. § 7. A Statistical Suggestion Prof. Karl Pearson (1900, p. 461) has approached the problem from the statistical side. If the female can be influenced at later reproductions by a male who has been associated with her in earlier ones, and if the alleged telegony is not due to some abnormal persistence of the spermatozoa of earlier unions, then in the permanent union of a pair we ought to find an increasing influence of the paternal type. But there seems to be, as regards stature, no evidence of any increase in the " hereditary influence " 156 TELEGONY of the father, therefore " no evidence of any steady telegonic influence.'* But an increasing hereditary influence of the same father seems to us rather different from the precise point at issue in the controversy over the occurrence or non-occurrence of telegony. It must be remembered that the bias of the child this way or that depends on the relative potency of the various items in the paternal and maternal contributions to the fertilised egg-cell, and that this relative potency may be affected by a variety of circumstances — e.g. the relative age or vigour of the gametes at the time of amphimixis. Careful comparisons of the families of the same mother by two successive husbands would be interesting — especially if there be anything in the suggestion that the telegonic influence is an influence exerted on the mother during gestation by the previous offspring, rather than directly through the previous father. § 8. The Widespread Belief in the Occurrence of Telegony The belief that offspring sometimes resemble not so much " the father, but an earlier mate of the mother," is widespread among experienced breeders, and, like the belief in the influence of maternal impressions upon the offspring, is probably very ancient. Apart from stock, the belief is often expressed in regard to man himself. " We certainly know that what used to be spoken of as the ' infection of the germ,' but which, following Weismann, we nowadays call ' telegony,' was considered possible by physiologists at the end of the seventeenth century ; we know the infection tradition has long influenced Arab breeders, and that believers in this hypothesis may now be found in every part of the world, more especially wherever an overlapping of distinct races occurs — as e.g. in the southern states of America and in certain Turkish provinces. Further, until quite recently many biologists considered that what is commonly and conveni- WIDESPREAD BELIEF IN ITS OCCURRENCE 157 ently known as Lord Morton's experiment has proved ' infection of the germ' to at least occasionally take place " (Ewart, 1899, P- 57)- It is psychologically interesting, therefore, to ask for some explanation of the widespread belief in the occurrence of a phenomenon the scientific evidence of which seems so slender. There is no doubt, we are told, that the value of a pure-bred bitch at once goes down if she has been accidentally lined by a mongrel, and it is possible that there may be good reason for this apart from the fact that the episode is not one which figures well in the record. It is possible that the constitution of the bitch may be subtly affected by a crossing — especially a fertile crossing — with a dog of inferior strain ; and that the deteriorated constitution may react upon future offspring although real telegony does not ensue. One must remember, however, that the statements one hears are often fairly precise. " If a pointer bitch gets accidentally served by a collie dog and produces a litter, the pups will be of various types, some like the pointer, some like the collie, and some a blend. And let that pointer bitch be afterwards served by a pure pointer dog, the result will be a litter among which the collie type can be unmistakably observed." It is desirable that some effort should be made to secure absolutely definite state- ments, supplemented by photographs. It is hardly sufficient to remind ourselves that people are indescribably careless about their beliefs, and that breeders are notoriously superstitious ; for considerations of money value have a potent effect in evolving carefulness, and breeding is gradually becoming an art based on scientific conclusions. There must be some basis for the widespread belief, and the answer given by the practical men themselves is that they have had abundant experience of the occurrence of telegony. This asser- tion leads us to look for phenomena which might be readily mistaken as telegonic, and there can be little doubt that Prof. 158 TELEGONY Ewart is right in maintaining that the mistake is in the mis- interpretation of reversions. A glance at the chapter on reversion (Chapter V.) will remind the reader that the crossing of different strains often results in apparent " throw-backs." A dark bantam hen paired with an Indian game Dorking produced, amongst others, a cockerel almost identical with a jungle fowl (Callus bankiva) — that is, with the original wild stock. What occurs when different breeds are crossed may occur on a smaller scale when individuals of the same breed, but of different strains, are crossed. When reversionary phenomena occur they usually spell disappointment to the practical breeder. In search of an explanation, he some- times thinks that he finds one in telegony ; that is to say, gives the blame of the reversion not to the immediately preceding crossing, which was theoretically correct, and should have turned out well, but to some remoter, less careful, or perhaps accidental crossing. In this way the remoter sire is made the scapegoat for the reversion, and the belief in telegony has grown. § 9. An Instructive Family History A good instance of the way in which cases of alleged telegony evaporate when analysed has been given by Dr. O. vom Rath. It concerns the somewhat intricate family history of certain cats. A family who had lived for many years in Tunis migrated in 1888 to Baden, taking with them a beautiful pair of kittens. These were none the worse for the change, except that they grew up very unwilling to leave the house, and more or less vicious. The female cat (F) was grey-brown with black stripes ; the torn (M) was pitch-black, except a large white spot on the right breast, and had a naturally half-sized left ear. In each litter which they cast there were some abnormal kittens, with rudimentary ear and tail. All these and all the males were destroyed ; the normal females were given away. But as the torn (M) became more and more vicious he was castrated, and became peaceful and lazy. A NOTE ON XENIA 159 The she-cat (F) was then crossed with an unblemished German torn, but she still produced abnormal kittens in each litter. Thus a strong suggestion of telegony arose. Further inquiries showed, however, that a normal daughter of F, crossed with a normal German torn, had borne a red male with rudimentary left ear and rudimentary tail. Inquiries as to the pedigree of F and M showed that /, the mother of F, had a rudimentary tail, but no rudimentary ear, and was like F in colour. This / had been crossed with a red torn (R), who had a rudimentary ear and tail ; there was but one litter, which was destroyed, and R soon afterwards died. Then / was paired with a normal black younger brother (S) of the deceased (R). From this normal S and from this / with a rudimentary tail, F sprang. But the two parents of / and the two parents of R and S were relatives, belonging to a family in which a rudi- mentary ear and tail were common — all springing from a pair which the owner of F and M had found in a hollow tree near Tunis. Dr. vom Rath has more to tell, but enough has been quoted to show the correctness of his conclusion that there was no tele- gony at all. There was a strong family tendency to having a rudimentary ear and tail. But it is evident that if Vom Rath had not had patience to search out the family history, the case for the occurrence of telegony would have been fairly good — at least as good as many others. § 10. A Note on Xenia The mysterious name " xenia," which seems to mean " guest- gifts," was applied by the botanist Focke to cases where the pollen from the " male " parent seemed to affect the tissue of the maternal ovary — the substance of the seed, or even the fruit, as distinguished from the embryo itself. Correns has made careful experiments with maize and estab- 160 TELEGONV lished that there at least xenia occurs. When the white-grained variety (Zea alba) is pollinated from the blue-grained variety (Zea cyanea), the majority of the seeds have white endosperm around the embryo, but a few have blue endosperm. The con- verse is likewise the case. It must be noted that the effect is only on the so-called " endosperm," or nutritive layer around the embryo ; the envelope of the seed, for instance, is never affected. What happens seems to be this. The pollen-tube arising from the pollen-grain contains two generative nuclei, which arise by the division of one. Of these two nuclei, one fertilises the egg- cell, the other fuses with what are called the polar-nuclei (a fact discovered by Nawaschin and Guignard). Thus there is a sort of double fertilisation within the embryo-sac ; the one results in the embryo, the other gives origin to the endosperm. Thus we see that xenia (in the well-authenticated case of maize) is no mysterious influencing of maternal tissue by the pollen-tube, and that it does not require Darwin's hypothesis of a migration of " gemmules " from the fertilised ovum into the surrounding tissue. It is a phenomenon sui generis, due to the very peculiar " double fertilisation." As Weismann points out, it corroborates the view that the nuclei are the vehicles of the hereditary qualities. Many of the alleged cases of xenia are cited in Prof. Delage's great work (1903, p. 252), the most picturesque being that of an apple-tree of Saint Valery. " This tree was sterile through the abortion of its stamens. Every year the young girls gathered branches from other apple-trees in flower, and shook them over the flowers of the non-staminate tree to fertilise them. Tillet De Clermont-Tonnerre (1825) relates that the resulting fruits recalled in their size, colour, and taste, those of the trees which had furnished the pollen." It is to be feared, however, that many of the alleged cases of xenia will not stand examination. Thus the records in regard MATERNAL IMPRESSIONS 161 to peas do not seem to be relevant, since the two halves of the pea-seed are of course the cotyledons and part of the embryo. Some of the phenomena seem simply ordinary cases of Mendelian inheritance (see Chapter X.). Some of the cases where it is said that the whole fruit is affected — e.g. in grapes and oranges — well deserve further investigation. § ii. Maternal Impressions It is a time-honoured belief that the mental states — especially vivid sense-impressions and strong emotions — of a pregnant mother may so affect the unborn offspring that structural changes result which have some correspondence with the maternal ex- perience. The belief was hardly doubted till Blondel began to criticise it early in the eighteenth century. Every one allows that the mother's health in the widest sense may react on the offspring, within what limits we hardly know ; but it is a very different matter to believe in definite and specific structural effects. There can be no doubt that the firmly rooted theory is in the main quite unscientific, except in the sense that it expresses the instinct to discover some cause for peculiar phenomena. A child has hypertrichosis : did not the mother look too long at a picture of John the Baptist in a hairy robe ? A white mother has a dark child: what can she say but that she was frightened by a Moor ? The abundant literature on the subject has been carefully studied by Dr. J. W. Ballantyne, and it need hardly be said that his general verdict is wholly against the tenability of the theory, except in a very refined form. The mental experiences of the mother have been held to ex- plain peculiarities of colour, abnormal hairiness, birth-marks, malformations, and even conception itself. The post hoc ergo propter hoc argument has never been more wildly used, and the result has been a retardation of the study of ante-natal pathology. Jacob's trick of using peeled wands to influence the colour of ii i62 TELEGONY his stock is still practised in modified form. A famous breeder of cattle has assured me that to obtain a particular colour of calf from a cow which persistently refused to produce what he wanted, he followed the patriarch's prescription with success. He had her covered blindfold ; after the sire had gone he brought to her a heifer of the desired colour, and that was the first object she saw when the bandages were removed ; she was left with the heifer as a companion to occupy her mind, and the result in due time was a calf of the desired colour. Nor was this an isolated case. What can one say — the credibility of the witness being secure — except the unsatisfactory word " coincidence " ? One requires to know in what direction, as regards colour, the sire was prepotent. One requires to know how many failures are forgotten in pro- portion to the successes remembered ? It is admitted that shock and distress and the like may have prejudicial effects on the unborn offspring. It is stated that after the Irish famine and after the siege of Paris there were many children born with stigmata of various sorts, and these were sometimes referred back to particular experiences instead of to the general state of malnutrition and nervous exhaustion. But to associate a particular structural defect with a particular mental impression seems an untenable position. The modus operandi is difficult to conceive of. Sometimes, indeed, the maternal-impression theory is demonstrably untenable, when the impression occurs late in pregnancy, for most of the great events in development occur very early. We have also to re- member the multitude of cases in which, in spite of very startling maternal experiences, the offspring is quite normal. In com* parison with this multitude of cases where nothing happens, the number of really puzzling cases is very small, and may be dismissed as coincidences. At the same time it is always unwise to speak of impossi- bilities in regard to matters which are inadequately known and imperfectly understood. That we cannot imagine the nature of a A PUZZLING CASE ;6$ physiological nexus does not prove its non-existence. Thus, as in regard to the transmission of acquired characters and telegony, we may be scientifically sceptical and give a verdict "non- proven," without dogmatically saying " impossible." We can understand how contact with a puzzling case gives the observer pause. A medical practitioner of keen scientific intelli- gence told me of a patient who, during pregnancy, had seen her husband suffer a serious accident. His arm was cut open by a falling block. As the impression seemed to weigh on the woman's mind in its relation to the unborn child, the doctor was asked to reassure her — which he did, with confidence and no doubt with skill. He was rather startled, however, when the time came, to find that the child he ushered into the world had a mark on the arm suggestive of the father's wound, and on the same arm. We must remember that for a prolonged period the unborn child is part and parcel of the mother — almost an integral part of herself — and we are beginning to know enough of the influence of mind upon body to make us cautious in dogmatising as to the possibilities of what Ballantyne * finely calls " the mysterious wireless telegraphy of ante-natal life." * While expressing his disbelief in the potency of maternal impressions to cause conditions in the foetus resembling the impression, Dr. J. W. Ballantyne cautiously adds (" Discussion on Heredity in Disease," Scottish Med. and Sing. Journ. vi. 1900, p. 310) that " to whatever extent we believe the mind capable of influencing the state of a part of the body, to that same extent, or to a degree rather less, the mother's mind might influence her parasitic growth — i.e. the foetus in utero. But this amount of belief would of course vary very much in accordance with the elasticity of our belief regarding the influence of the mind over the body." CHAPTER VII THE TRANSMISSION OF ACQUIRED CHARACTERS " A right answer to the question whether acquired characters are or are not inherited underlies right beliefs, not only in Biology and Psychology, but also in Education, Ethics, and Politics." — HERBERT SPENCER. " II n'est pas demontre que les modifications acquises sous 1'influence des conditions de vie soient generalement hereditaires, mais il parait bien certain qu'elles le sont quelquefois. Cela depend sans doute de leur nature." — YVES DELAGE. [This is the opinion of one of the acutest of living biologists, but we find ourselves forced to a negative position.] § i. Importance of the Question. § 2. Historical Note. § 3. Definition of the Problem. — s § 4. Many Misunderstandings as to the Question at Issue. 5. Various Degrees in which Parental Modifications might affect the Offspring. § 6. Widespread Opinion in favour of Affirmative Answer. • § 7. General Argument against the Transmissibility of Modifications. § 8. General Argument for the Transmissibility of Modi- fications. § 9. Particular Evidences in support of the Affirmative Answer. § 10. As regards Mutilations and the Like. § ii. Brown-Sequard's Experiments on Guinea-pigs. § 12. Negative Evidence in favour of the Affirmative Answer. § 13. The Logical Position of the Argument. § 14. Indirect Importance of Modifications. § 15. Practical Considerations. 164 THE PROBLEM NOT MERELY ACADEMIC 165 § i. Importance of the Question No one is at present entitled to rank the transmission of " acquired characters " — i.e. somatic modifications — among the facts of inheritance, and the logical place for a discussion of this subject should be beside other disputed questions, like the occurrence or non-occurrence of telegony. But we have given special prominence to a discussion of this problem because of its great importance both practically and theoretically, and because of the abundant debate which has been aroused over it. Not a Merely Academic Problem. — The question as to the transmissibility of characters acquired during life by the body of the parent as the result of changes in environmental or func- tional influences is much more than a technical problem for biologists. Our decision in regard to it affects not only our whole theory of organic evolution, but even our every-day conduct. The question should be of interest to the parent, the physician, the teacher, the moralist, and the social reformer — in short, to us all. If the particular results of changes or peculiarities in individual nurture, education, and experience do not directly and specifi- cally affect the inherited nature of the offspring, there must be a revision of some current psychological and pedagogical opinions ; but it must be borne in mind that man's rich external heritage of tradition and convention, custom and institution, law and literature, art and science, makes his case quite peculiar, for the results of man's external heritage are often such as might have come about if acquired characters were heritable. If the particular results of changes or peculiarities in individual " nurture " do not directly and specifically affect the inherited nature of the offspring, there must be a revision of that theory of organic evolution which is usually called Lamarckian, in which it is a central postulate that whatever is acquired may also be transmitted, l66 TRANSMISSION OF ACQUIRED CHARACTERS Spencer's Estimate of the Importance of this Question. After contrasting the two hypotheses of the transmissibility and the non-transmissibility of acquired characters, Herbert Spencer said : " Considering the width and depth of the effects which the acceptance or non-acceptance of one or the other of these hypotheses must have on our views of life, the question, Which of them is true ? demands beyond all other questions whatever the attention of scientific men. A grave responsi- bility rests on biologists in respect of the general question, since wrong answers lead, among other effects, to wrong belief about social affairs and to disastrous social actions." This authoritative statement removes all need of apology for the prominence which we have given to the question. An Interminable Question. — The attention of scientific men which Herbert Spencer demanded for this problem has not been grudgingly given. The subject has been keenly debated for many years ; there are, as our bibliography will show, scores of papers and not a few books devoted to its discussion. Indeed, one of the most tolerant of biologists, Prof. W. K. Brooks, has spoken of it as " the interminable question." Those who give the affirmative answer have not succeeded in proving their case ; as for the other side, how can they prove a negative ? Therefore, while we have no hesitation as to the verdict of " non- proven" to which the evidence at 'present available points, we do not expect a satisfactory issue until many years of experimental work have supervened. Why, then, if a satisfactory termination be not at present possible, and if no unanimity even among experts can be looked for, should we enter upon the discussion once more ? Prof. Brooks states our warrant in a quotation from Berkeley's Siris : " It is Plato's remark in his Thecztetm, that while we sit still we are never the wiser, but going into the river and moving up and down is the way to discover its depths and shallows. If we exer- cise and bestir ourselves we may even here discover something." HISTORICAL NOTE 167 Experiment is doubtless most urgent, but misunderstandings in regard to the problem are still so prevalent that we take courage in attempting a re-discussion, from which we have tried to eliminate obscurity and prejudice. § 2. Historical Note Doubt as to the transmission of acquired characters is certainly not novel, though Galton and Weismann deserve credit for denning the scepticism. Brock has pointed out that the editor, whoever he was, of Aristotle's Historia Animalium seems to have differed from his master on this subject. Aristotle had referred to the transmission of the exact shape of a cautery mark, but the editor insinuated a doubt as to credibility of instances of this sort. Kant. — In modern times Kant was one of the first to express a firm disbelief in the transmission of individual peculiarities ; Blumenbach inclined to the same opinion ; but neither seems to have defined precisely what he intended to exclude from the bundle of inheritance. Prichard. — James Cowles Prichard (b. 1786), a well-known anthropologist, anticipated as early as 1826 some of the character- istically modern views on evolution. His importance has been pointed out by Prof. Edward B. Poulton. In the second edition of his Researches into the Physical History of Mankind (1826), Prichard stated the case in favour of the general evolu- tionist interpretation of animate nature, recognised the operation of natural and artificial selection, and not only drew a clear distinction between acquired and inborn peculiarities, but argued that the former were not transmitted. He was not rigidly consistent, however, and his convictions seem to have weakened in after years ; yet his anticipation of one of Weismann 's positions by more than half a century is very interesting. In more recent times we find sporadic expressions of scepticism i68 TRANSMISSION OF ACQUIRED CHARACTERS as to the transmission of acquired characters — e.g. by the mor- phologist His and the physiologist Pfliiger; but, as we have said, the focussing of the question was due to Galton and Weismann. Galton. — Thus Galton in 1875 stated his opinion that the current theory of the inheritance of characters acquired during the lifetime of the parents " includes much questionable evidence, usually difficult of verification. We might almost reserve our belief tha_t/the structural cells can react on the sexual elements at all, and we may be confident that at the most they do so in a very faint degree — in other words, that acquired modifications are barely, if at all, inherited in the correct sense of that word." Galton's position at that time may be summed up as follows : (1) In regard to climatic variations, Galton doubted the reality of any reaction of the " body " upon the germs, but believed that the germs are themselves directly affected. (2) The same is true in regard to many diseases that have been acquired by long-continued irregular habits. /(3) The cases of the apparent inheritance of mutilations are out- numbered by the overpowering negative evidence of their non-inheritance. (4) It is hard to find evidence of the power of the personal structure ( -*" to react upon sexual elements that is not open to serious objection. That which appears the most trustworthy lies almost wholly in the direction of nerve changes, as shown by the inherited habits of tameness, pointing in dogs, and the results of Dr. Brown-Sequard's experiments on guinea-pigs. Weismann. — But Weismann gave the scepticism an even sharper point. He denied all transmission of acquired modifi- cations, partly because he found the evidence so flimsy and anecdotal, partly because he could not conceive of any mechanism whereby the transmission of a particular acquired modification could be effected, and partly because his whole theory of heredity and variation raised strong probabilities against the view that VIEWS OF G ALTON AND WEISMANN 169 acquired characters were transmitted. On Weismann's view the sole fountain of specific change is in the germ-plasm of the sex-cells. It is true that the environment makes dints on the organism, but only upon its body ; the reproductive cells, through which alone the change could be transmitted, are either unaffected or are not affected in such a definite way as to bring about the transmission of the parental modification. It is true that the results of changed function (use and disuse) are often very marked, and very important for the individual; but they are not transmitted as such or in any representative degree, and therefore are of no direct account in the evolution of the species. Thus the ground is taken from under the feet of Buffonians and Lamarckians, and the whole burden of organic progress is laid upon germinal variation and the processes of selection. The following sentences indicate Weismann's original posi- tion: (1) "Acquired characters are those which result from external influence upon the organism, in contrast to such as spring from the constitution of the germ." (2) " Characters can only be inherited in so far as their rudiments (' Anlagen ') are already given in the germ-plasm." (3) " Modifications which are wrought upon the formed body, in consequence of external influences, must remain limited to the organism in which they arose." (4) " So must it be with mutilations, and with the results of use or disuse of parts of the body." (5) " No such modifications of the soma (affected by environment or by use and disuse) can be transmitted to the germ-cells, from which the next generation springs. They are, there- fore, of no account in the transformation of the species." (6) "The only principle that remains for the explanation of the transformation of the species is direct germinal variation." On germinal variations natural selection operates in the usual way. The helpful subsidiary theory of germinal selection was afterwards suggested, and various saving clauses were added, which do not, however, affect the clearness and strength of Weismann's original position. 170 TRANSMISSION OF ACQUIRED CHARACTERS Lamarck's Laws.— It may be fairly said that the fons et origo of the affirmative position was Lamarck. Though he did not originate, he formulated and illustrated the theory of the in- heritance of acquired characters. He maintained the trans- missibility of modifications due to increased and decreased and changed use, and also of modifications due to environmental change, whether directly induced, or indirectly induced by altered function. The giraffe has attained its long neck by stretching it for many generations ; swimming birds have got webbed feet because they stretched their toes in the water ; wading birds have got long legs because they stretched them ; the mole has very small eyes because it has ceased to use them ; the whalebone whale has no functional teeth because it has acquired the habit of swallowing its food without mastication ; and so on. Lamarck's two laws of nature, which he said no observer could fail to confirm, were : * (1) In every animal that has not passed beyond the term of its development, the frequent and sustained use of any organ strengthens it, develops it, increases its size, and gives it strength proportionate to the length of time of its employ- ment. On the other hand, the continued lack of use of the same organ sensibly weakens it ; it deteriorates, and its faculties diminish progressively, until at last it disappears. (2) Nature preserves everything that she has caused the individual to acquire or to lose by the influence of the circumstances to which the race has been for a long time exposed, and consequently by the influence of the predominant use of certain organs (or in consequence of their continued disuse). She does this by the generation of new individuals, which are produced with the newly acquired organs. This occurs* provided that the acquired changes were common to the two sexes, or to the individuals that produced the new forms. Prof. E. Ray Lankester has pointed out (1894) that Lamarck's * I have taken the translation from T. H. Morgan's Evolution and Adaptation (1903), p. 226. CRITICISM OF LAMARCKS LAWS if i first and second laws are QontradLctory the one of the other. In correspondence with the normal conditions of the environment, organisms show " responsive " quantities in their parts ; but. change a young organism to an environment quantitatively different, and it shows new responsive quantities in the parts of its structure concerned, new or acquired characters. " So far, so good. What Lamarck next asks us to accept, as his ' second law/ seems not only to lack the support of experi- mental proof, but to be inconsistent with what has just preceded it. The new character, which is ex hypothesi, as was the old character (length, breadth, weight of a part) which it has re- placed— a response to environment, a particular moulding or manipulation by incident forces of the potential congenital quality of the race — is, according to Lamarck, all of a sudden raised to extraordinary powers." It is declared to be trans- missible, that is, it alters the potential character of the species, so as to persist when other quantitative external conditions are substituted for those which originally determined it. But this has never been experimentally proved, and there is strong reason for holding it to be improbable. " Since the old character (length, breadth, weight) had not \become fixed and congenital after many thousands of successive generations of individuals had developed it in response to environment, but gave place to a new character when new conditions operated on an individual (Lamarck's first law), why should we suppose that the new character is likely to become fixed after a much shorter time of responsive existence, or to escape the operation of the first law ? Clearly there is no reason (so far as Lamarck's statement goes) for any such suppo- sition, and the two so-called laws of Lamarck are at variance with one another. " In its most condensed form my argument has been stated thus by Prof. Poulton (Nature, vol. li., 1894, p. 127) ; Lamarck's ' first law assumes that a past history of indefinite duration 172 TRANSMISSION OF ACQUIRED CHARACTERS is powerless to create a bias by which the present can be con- trolled ; while the second assumes that the brief history of the present can readily raise a bias to control the future.' ' (See E. Ray Lankester's Kingdom of Man. 1907, pp. 128-130.) Lamarck ism remains alive. — The Lamarckian position is still stoutly maintained — usually in more or less modified form — by many prominent naturalists, especially in France and America. It is often held along with a more or less half-hearted Darwinism, just as Darwin combined some Lamarckism with his own selec- tionist doctrine — even in spite of his protest, " Heaven forfend me from Lamarck nonsense of a tendency to progression, adapta- tions from the slow willing of animals, etc." Though Alfred Russel Wallace has said, " The hypothesis of Lamarck has been repeatedly and easily refuted by all writers on the subject " ; though Huxley said, " The Lamarckian hypothesis has long since been justly condemned" ; though Ray Lankester has said that perhaps the greatest step of progress in modern aetiology will be the complete removal of all taint of Lamarckism,— there remains a vigorous school of Lamarckians and a still more vigorous school of Neo-Lamarckians, who, whatever be the truth in regard to the transmission of acquired characters, have got a firm grip of the often-overlooked commonplace that the organism is an active, self-assertive, self-adaptive living creature — to some extent master of its fate. § 3. Definition of the Problem A Protest. — Much time and energy have been wasted on the discussion as to the transmissibility or non-transmissibility of " acquired characters " or somatic modifications, through lack of precise definition of the terms. Usually, though not always, the fault has been with the supporters of the affirmative position, who have failed to observe the rules of the game by ignoring the definitions of those who find themselves forced to a negative DEFINITION OF THE PROBLEM 173 Conclusion. By all means let there be a critical discussion as to the best definition of " an acquired character," " a modifica- tion," " a somatic change induced on the body by environmental or functional influences " ; by all means let there be a criticism of terms and categories — the minting of a perfectly unambiguous word for somatic modifications would be most welcome : but if the sheaves of facts and alleged facts are to be thrashed out with the end of getting at the wheat of truth, we must adhere to certain definitions — notably, of course, to those given by Weismann, who brought the problem in its modern aspect into focus. Even a sense of humour should hinder a young medical practitioner from thinking that he makes for progress by ad- vancing an argument which has no cogency unless the biological dictionary be first re-edited. It should be evident that a dis- cussion over which some of the wisest heads in Europe and America have pondered cannot be, as some have had the effrontery to declare it, a mere play of words. Is it too much to ask of those who are keen to break a lance with the Biologist of Freiburg that they should first at least read The Germ- Plasm ? What is an Acquired Character? — In our previous dis- cussion of " heredity and variation " we have briefly expounded the distinction between germinal, blastogenic, constitutional, endogenous " variations," and bodily, somatogenic, acquired, exogenous " modifications." An acquired character, or a somatic modification, may be defined as a structural change in the body of a multicellular organism, involving a deviation from the normal, directly induced during the individual lifetime by a change in environment or in function (use and disuse), and such that it transcends the limits of organic elasticity, and there- fore persists after the factors inducing it have ceased to operate. Illustrations. — Dwarfing of Japanese trees, deformation of trees by the wind, blanching of plants grown in darkness, changes directly induced by transplantation, persistent sun-burning, change 174 TRANSMISSION OF ACQUIRED CHARACTERS of colour after particular diet, callosities induced on the skin by pressure, e.g. those at first produced on the finger-tips of one who is learning to play the violin, dwarfing of animals in confined space, increased muscular development by exercise, atrophy of muscles through disuse, chronic fatigue of nerve-cells, alterations in the walls of the food-canal through particular diet, changes in the skeleton as the result of specialised activities, increased growth of hair, etc., after importation to a warm climate, accumulation of fat as the result of modified nutrition, and so on through a long list. To understand the question clearly we must spend a little time and thought over it. Let us briefly consider the various relations between an organism and its surroundings. T. Relation of Dependence between Organism and Environ- ment.— It is a familiar fact that a living creature is de- pendent upon its surroundings. A great part of life consists in action and reaction between the organism and its environ- ment. It is a profound commonplace that between the animate system — so incomprehensibly unified — and its inanimate milieu, there is a continual coming and going of matter and energy. On this life depends. The may-fly during its short aerial life must breathe even if it does not feed ; the philosopher requires his dinner, just as his dog does. This may be called the relation of constant and normal environmental dependence — necessary to the development and to the continuance of the organism. 2. Transient Adjustments. — But surroundings are changeful, and the living creature changes with them. A great part of life consists of effective responses to external changes ; consciously or sub-consciously the organism adjusts itself to changes in its environment, or works in the direction of adjustment. There is bright sunshine and our pulse beats more quickly ; the external temperature rises and we perspire. Thousands of these changes are familiar, saving life from monotony. Yet in regard to many there remains no abiding result that can be detected. There are structural changes attendant on normal nerve-fatigue, but in rest and food we gain almost complete recuperation. No LONG PERSISTING ADJUSTMENTS 175 doubt there is always some lasting impression, for even the bar of iron is never quite the same after it has been once struck ; but the results of the slight organic changes we have been alluding to are usually lost as the sand-ripples are lost when the tide turns. They are the merely transient results of re- sponses to frequently recurring environmental changes to which the organism is well accustomed. 3. Adjustments which persist for a Considerable Time.— Insensibly, however — for it is all a matter of degree — we pass from transient results to others which last for a considerable time. We are browned by the sun on our summer holiday, and the result may last far into the autumn. The change, though still very superficial, has taken a firmer hold. The world is full of illustrations — the increase in the child's weight after a month at the farm, the increase in the size of the muscles after a course of Sandow exercises, the warping of the plant-stem which has been illumined from one side only, the blanching of the banked-up celery. But these results do not last long after the inducing conditions have ceased to operate. Sooner or later there is a return to the normal. Like a bow unstrung, the organism rebounds approximately to its previous state. The stimulus ceases or the absent stimulus is restored, and the organism, as if at the command "As you were," returns to the status quo. 4. Modifications. — Insensibly, however — for it is still only a matter of degree — we pass from these temporary changes to others which are demonstrably permanent. For there are cases where the new stimulus provokes a structural change, which persists after the stimulus has ceased. As we have put it, metaphorically, the limit of organic elasticity has been trans- cended. These are what in technical language we call " acquired characters " or " modifications." The Englishman who works half his lifetime under a tropical sun may become so tanned that the result does not disappear 176 TRANSMISSION OF ACQUIRED CHARACTERS during all the years in which he enjoys his pension at home. He has changed his skin, but he cannot by any means change it back again. Through prolonged disuse from early years a muscle may pass into a state of atrophy, and may so remain throughout life. Pressure on the little toe may so deform it, that even in the " easiest " shoes it can never right itself. A tree may be blown out of shape by the wind, and the crooked bough may never be straightened. Over-exertion may strain the heart permanently. A sudden shock may be followed by a whitening of the hair from which there is no natural recovery. 5. Modifications and Variations. — When we analyse the observed differences between fellow members of a species, we find that some of them can be definitely associated with peculiarities of function and environment. They can be more or less ac- counted for physiologically in terms of some change in surrounding influences or of some change in function thereby induced. They may not be* hinted at in the young forms, but they begin to appear when the peculiar conditions begin to operate, and they are usually exhibited in some degree by all organisms of the same kind which are subjected to the same change of conditions. Furthermore, they can be experimentally brought about. These are " modifications." By those who measure observed differences they are usually slumped along with true variations, but this appears to us to lead to confusion. True variations are those peculiarities / which remain when all the modifications are subtracted from the total of observed differences. It goes without saying that the distinction cannot always be drawn in practice. Often, however, it is quite apparent, and in any case the theoretical distinction is clear. Variations, in the strict sense, cannot be causally related to peculiarities in habit or surroundings ; they are often hinted at in the earliest stages — even before birth ; and they are very unequal even MODIFICATIONS AND VARIATIONS 177 among organisms whose conditions of life seem absolutely identi- cal. We refer them to changes in the germinal material before or during fertilisation. We call them endogenous, constitutional, blastogenic ; and there is no doubt that they are transmissible, though they are not always transmitted. Is there really an Antithesis ?— Some subtle minds have found satisfaction in maintaining that the distinction between an acquired modification and an inborn variation is a distinction without a difference. In his interesting Problems of Biology Mr. George Sandeman points out that every acquired quality is congenital (i.e. there are in the organisation the rudimental possibilities of it), and that every congenital quality is also acquired (i.e. it requires to be nurtured by appropriate con- ditions if it is to develop). In this epigram there is undoubtedly truth, but is it relevant ? No doubt the possibility of the modification must be in the organism, just as the possibility of an explosion is in the barrel of gunpowder. The environment is not creative ; yet, as a matter of fact, it seems possible to distinguish between the actual modification which we see and measure and the possibility of it which we presuppose. Similarly, it is very true that the potentialities so marvellously embodied in the fertilised egg- cell require appropriate environing conditions if they are to be realised, for, as His observed long ago, "it is a piece of unscientific mysticism to suppose that heredity will build up an organism without mechanical means." The common jelly-fish (Aurelia aurita) often has a pentamerous instead of a tetramerous symmetry. This is a variation of germinal, endogenous origin. Of course it requires an environ- ment to develop in, but we cannot causally relate the structural peculiarity to any peculiarity in the environment. It seems to be logically quite distinguishable from a modification. Discussing words is often indescribably tiresome, but it is better than misunderstanding them. ' ' Inheritance of acquired characters ' ' 12 iy8 TRANSMISSION OF ACQUIRED CHARACTERS may be a most unfortunate phrase, but it has come to have a perfectly definite technical meaning and usage, which any normal person can understand in a few minutes. Prof. W. K. Brooks, in his Foundations of Zoology, says that he never uses the phrase " inherit- ance of acquired characters " except under protest, and this may be commendable restraint ; but it seems to us inconsistent with his usual wisdom to go on to say, " If any assert that the dog inherits anything which his ancestors did not acquire, their words seem meaningless ; for, as we use words, everything which has not existed from the beginning must have been acquired — although one may admit this without admitting that the nature of a dog is, wholly or to any practical degree, the inherited effect of the environment of his ancestors." But as the word " acquired " is now a technical term, meaning wrought out on the body as the result of changes in environmental or functional stimuli, we fail to see that, as we use the words, there is anything meaningless in the first assertion, or any warrant for the second. Summary. — What forms the material basis of all inheritance, in all ordinary cases of sexual reproduction among multicellular organisms, is the fertilised ovum. The question under dis- cussion is, physiologically stated, whether we can conceive that structural changes in the body of a parent, induced by changes in functional or environmental influence, can so specifically affect the reproductive cells that these will, if they develop, reproduce in any degree the modification acquired by the parent or parents. The question under discussion, logically stated, is whether there are any secure phenomena of inheritance which forcibly suggest the reality of the transmission of acquired characters ; or whether, if such phenomena there be, a simpler interpretation may not be found. If, summing up in Galton's phrase, we call environmental and functional influences " nurture/' our question is seen to be the exceedingly important one, May the results of peculiarities in parental " nurture" be as such trans- mitted, or is it the germinal " nature" alone that constitutes the inheritance ? SOME MISUNDERSTANDINGS CONSIDERED 179 § 4. Many Misunderstandings as to the Question at Issue The precise question is this : Can a structural change in the body, induced by some change in use or disuse, or by a change in surrounding influence, affect the germ-cells in such a specific or representative way that the offspring will through its inheritance exhibit, even in a slight degree, the modification which the parent acquired ? Before we pass to discuss the evidence pro and con it will be useful to notice some frequently recurring misunderstandings, the persistence of which would make further argument futile. Misunderstanding I — How can there be progressive evolution if acquired characters are not transmitted ? — Those who have not thought clearly on the subject often shake their heads sagely and remark that they " do not see how evolution could have been possible at all unless what is acquired by one generation is handed on to the next." To this we have simply to answer (i) that our first business is to find out the facts of the case, careless whether it makes our interpretation of the history of life more or less difficult, and (2) that in the supply of germinal variations, whose transmissibility is unquestioned, there is ample raw material for evolution. We know a little about the abundant crop of variations at present supplied ; there is no reason to believe that it was less abundant in the past. Misunderstanding II — Interpretations are not facts. — There are many adaptive characters in plants and animals which may be superficially interpreted as due to the direct result of use and disuse or of environmental influence. The Lamarckians have so interpreted them, and the Lamarckian way of looking at adaptations has become habitual to many uncritical minds. They see on modern flowers the footprints of insects which have visited them for untold ages ; they speak of the dwindling of the whale's hind-limbs through disuse, of 180 TRANSMISSION OF AC QV I RED CHARACTERS the hardening of the ancestral horses' hoofs as they left the marshes and ran on harder ground ; they picture the giraffe by persistent effort lengthening out its neck a few millimetres every century, as the acacia raised its leaves higher and higher off the ground; and they say that animate nature is so full of evidences of the inheritance of acquired characters that no further argument is needed. But all this is a begging of the question. It is easy to find structural features which may be interpreted as entailed acquired characters, */ acquired characters can be entailed. Obviously, however, we must deal with what we can prove to be modifi- cations, or with what we can plausibly regard as modifications because we find their analogues in actual process of being effected to-day. It is easy to say that the blackness of the negro's skin was produced by the tropical sun, and that it is now part of his natural inheritance. It is easy to say this, but absolutely futile. Let us first catch our modifications. The Golden Rod (Solidago virgaurea) growing on the Alps is precocious in its flowering when compared with representatives of the same species growing in the lowlands. Hoffmann found that Alpine forms transplanted to Giessen remained precocious, therefore the acquired precocity had become heritable. But there is no evidence that the precocity was acquired ; it may have been the outcome of the selection of germinal variations. The African Wart-hog (Phacochcerus) has the peculiar habit of kneeling down on its fore-limbs as it routs with its huge tusks in the ground and pushes itself forward with its hind-limbs. It has strong horny callosities protecting the surfaces on which it kneels, and these are seen even in the embryos. This seems to some naturalists to be a satisfactory proof of the inheritance of an acquired character. It is to others simply an instance of an adaptive peculiarity of germinal origin wrought out by natural selection. SOME MISUNDERSTANDINGS CONSIDERED 181 Misunderstanding III — Begging the question by starting with what is not proved to be a modification. — There is no rele- vancy in citing cases where an abnormal bodily peculiarity re-appears generation after generation, unless it be shown that the peculiarity is a modification, and not an inborn variation whose transmissibility is admitted by all. Short-sightedness may recur in a family-series generation after generation, but there is no evidence to prove that the original short-sightedness was a modification. In all probability, short-sightedness is in its origin a germinal variation, like so many other bodily idiosyncrasies. In regard to some diseases, such as rheumatism, it is often said dogmatically by those who know little about the matter that the original affection in the ancestor was brought about by some definite external influence — such as a cold drive or a damp bed ; but it seems practically certain that in all such cases we have to do with an inborn predisposition, to the expression of which the cold drive or the damp bed was merely the liberating stimulus, comparable to the pulling of the trigger in a loaded gun. The liberating stimulus is, of course, of great importance, both in the case of the gun's discharge and the organism's disease, but it only goes a little way towards a satisfactory interpretation in either case. Not that we can explain the origin of rheumatism or shortsightedness or any such thing — there is no explanation in calling them germinal variations that cropped up; but we are almost certain that they never are modifications or acquired characters. Herbert Spencer twits those who are sceptical as to the trans- mission of acquired modifications with assigning the most flimsy reasons for rejecting a conclusion they are averse to ; but when Spencer cites the prevalence of short-sightedness among the " notoriously studious " Germans, the inheritance of musical 'talent, and the inheritance of a liability to consumption, as evidence of the inheritance of modifications, we are reminded of the pot calling the kettle black. 182 TRANSMISSION OF ACQUIRED CHARACTERS Over and over again in the prolific literature of this discussion the sjdlogism is advanced, either in regard to gout or something analogous — Gout is a modification of the body, an acquired character ; Gout is transmissible; Modifications are sometimes transmissible. It may be formally a good argument, but there is every reason to deny the major premiss. There is no proof that the gouty habit had an exogenous origin — that it was, to begin with, for instance, the direct result of high living ; though it is generally admitted that excesses in eating or drinking may give a stimulus to its expression. " The conclusion I have arrived at," says Prof. D. J. Hamilton (1900, p. 297), " is that the gouty habit of body has arisen as a variation, and as such is hereditarily trans- missible, and that excess of diet and alcohol merely renders the habit of body apparent." It may also be pointed out that gout and rheumatism and the like are rather processes of meta- bolism than structural modifications, though the latter may ensue. After pointing out the irrelevancy of citing cases of the here- ditary recurrence of polydactylism, haemophilia, colour-blindness in man, or the absence of horns in cattle or of tails in cats, as instances of the transmission of acquired characters, Prof. Ernst Ziegler says (1886, p. 13) : " Only that can be regarded as ' ac- quired ' which is produced in the course of the individual life, during or after the period of development, exclusively under the influence of external conditions ; the term is in no wise applicable to peculiarities which, as one says, arise of themselves from a predisposition already present in the germ." Let us state the case once more. There is no doubt that the expression of a germinal variation during the lifetime of an individual may be sometimes definitely associated with a. particular external stimulus. It may thus be mistaken for a modification, and mis- takenly spoken of as " acquired." But the relation between the MISUNDERSTANDINGS: IRRELEVANT CASES 183 provoking stimulus and the expression of the innate tendency or predisposition is more or less arbitrary — various kinds of stimuli will have the same result ; whereas the relation between an environ- mental influence and the induced modification is more or less constant — similar influences having similar results — and is more strictly causal. An external stimulus may provoke the expression of a germinal variation, as when a mouse provokes hysteria; but this is different physiologically from what occurs when the sun produces sun-burning. A certain abnormal psychosis, which may not have been hinted at during early years, suddenly emerges under provocation. It is carelessly spoken of (even in the law courts) as due to that provocation — a fright, a wound, a debauch, a railway accident, a night's exposure, and so on, and it is carelessly thought of as " acquired " ; it is recovered from, but it re-appears in the off- spring : therefore an acquired character may be transmitted. But there is the strongest probability that what was called an acquired psychosis was primarily germinal, and might have emerged under quite different stimulation — for instance, under the normal events of puberty and parturition. Another version of this misunderstanding is seen in references to the improvement of a breed in the course of generations, as the result, it is supposed, of functional modifications. Practice makes perfect in the individual, therefore also in the race. But we have seen no cases cited where the results were not hopelessly complicated by the occurrence of selection and elimination, which, by acting on constitutional variations, may quite well account for what is hastily referred to modification-inheritance. Herbert Spencer was keenly aware of the misunderstanding which we have been discussing. " Such specialities of structure as are due to specialities of function are usually entangled with specialities which are, or may be, due to selection, natural or artificial. In most cases it is impossible to say that a structural peculiarity which seems to have arisen in offspring from a functional peculiarity in a parent is wholly independent of some congenital peculiarity of structure in the parent, whence this functional peculiarity arose. We are restricted to cases with which natural or artificial selection can have had nothing to do, and such cases are difficult to find." Yet it is strange that he should point to such facts as the following : the bones of the wing in the domestic duck weigh less and the bones 1 84 TRANSMISSION OF ACQUIRED CHARACTERS of the leg more in proportion to the whole skeleton than do the same bones in the wild duck ; in cows and goats which are habitually milked the udders are large ; moles and many cave-animals have rudimentary eyes. Cases like these may be in part regarded as instances of individually re-acquired modifications, but they are for the most part readily interpreted as due to the selection of germinal variations. Misunderstanding IY — Mistaking the reappearance of a modification for transmission of a modification. — It is of little service to cite cases where a particular modification reappears generation after generation unless it be shown that the change recurs as part of the inheritance, and not simply because the external conditions which evoked it in the first generation still persisted to evoke it in those that followed. Reappearance is not synonymous with inheritance. Illustration. — When Prof. Nageli brought Alpine plants (Hieracium, etc.) to the Botanical Garden at Munich, many became in the first year so much changed that they were hardly recognisable as the same species, and their descendants in the garden were likewise quite different from their Alpine ancestors. The small Alpine hawkweeds became large and thickly branching, and blossomed freely. In some cases many generations were observed — even for thirteen years ; there was no doubt as to the reappearance of the acquired characters ; but it was not thereby proved that the reappearance was due to the inheritance. On the contrary, that the reappearance was due to the persistence of the novel conditions, to the changes which these directly impressed on each successive crop, was shown by the fact that when the plants were removed to poor, gravelly soil, the acquired characters disappeared, and the plants were re-transformed into their original Alpine character. " The re-transformation was always complete, even when the species had been cultivated in rich garden soil for several generations." Misunderstanding Y — Mistaking re-infection for transmission. > — A particular form of the fourth misunderstanding has to do with facts so special that it may be conveniently treated of separately. It has to do with microbic diseases. It is ad- MISUNDERSTANDINGS: MORE IRRELEVANCIES 185 mitted that a parent infected with tubercle-bacillus or with the microbe of syphilis may have offspring also infected. But such cases are irrelevant in the discussion. Infection, whether before or after birth, has nothing to do with inheritance. As Dr. Ogilvie says (1901, p. 1072), " Wherever the transmission of infectious disease from parent to offspring has been adduced to support the doctrine of the inheritance of acquired^ characters, it has been done in utter misconception of its meaning and scope.'* Medical men have sometimes condescended to make a subtle distinction between " hereditary " and " congenital " syphilis — the latter manifested at birth, the former some time afterwards ! It seems strange that they have failed to recognise that there is no reason to use the word " hereditary " at all in this con- nection. What occurs is an injection, and it is theoretically immaterial at what stage the infection occurs.* A microbe cannot be part of an inheritance.f Misunderstanding YI — Transmission in unicellulars is not to the point. — It is not to the point to cite cases where uni- cellular organisms, such as bacteria or monads, have been profoundly and heritably modified by artificial culture, so that, for instance, the descendants of a virulent microbe have been made to lose their evil potency. It is irrelevant because in regard to unicellular organisms we cannot draw the distinction * It may be the germ-cells that are infected — especially when the direct source of infection is the father ; or it may be the embryo that is infected through the placenta : but the difference in the time of the infection is of no theoretical interest, nor can it be inferred from any difference in the outward symptoms, as these appear in the offspring. f The egg of the green freshwater polyp (Hydra viridis) always contains little greenish corpuscles which are not present in the youngest stages af oogenesis. It is almost certain that these are minute unicellular Algae 'Zoochlorella). But no one can regard these useful symbions as actually part of the inheritance. The eggs of the silk-moth are often infected by a minute but fatal Protozoon which is present in the body of the moth. It seems uncertain at what precise point these pebrine organisms become associated with the egg, but however early it may be, the infection has nothing to do with inheritance. (See Ziegler, 1905, p. 5.) 1 86 TRANSMISSION OF ACQUIRED CHARACTERS between body and germinal matter, apart from which the concept of modifications is of no value. In artificial culture the whole character of the unicellular organism — its particular metabolism — is altered ; it multiplies by dividing into two or more parts, which naturally retain the altered constitution. But this is worlds away from the supposed case of an alteration in the structure of the little toe so affecting the germ-cells that the offspring inherit a corresponding deformation.* Prof. Adami (1901, p. 1319) says: "By subjecting a growth of pigment-producing bacteria to the action of a temperature just below that which will cause their death, we can bring about a loss of pigment production, so that the rapidly-succeeding genera- tions are perfectly colourless ; but gradually, in the course of time, the cultures made from the original (heated) tube regain the power of pigment production. This may be in two or three days, or, again, only after several transplantations at the end of two or three weeks ; and when we remember that a bacillus divides and so forms a new generation in, on the average, something considerably less than an hour, it is seen that the acquired character may be impressed upon a race for some hundreds of generations. The more intense the alteration to which the bacillus is subjected, the longer and the more frequently the race is subjected to the altered temperature conditions, the longer it is before there is a sign of return to the normal." These are interesting and reliable facts, but their citation as evidence of the inheritance of " acquired characters " is misleading, since no bacilli show any hint of the distinction between somatic and germinal material on which the definition of " acquired char- acters " depends, nor do they multiply except by division and * It is surprising that even Prof. Oscar Hertwig (1898) supports his argument in favour of the transmissibility of somatic modifications by citing cases of inheritance in unicellular organisms. We are told that the irritability of certain Algae to light may be modified by exposure to strong light and to high temperature, and that " nobody would be sur- prised " if the progeny also showed " some similar property." But this is hardly evidence of the transmission of a modification ! We are also told that under artificial conditions some bacteria may lose their toxic properties, and may transmit this somewhat negative character of lost virulence. This is admitted by all, but it is an ignoratio elenchi. TRANSMISSION IN UNICELLULARS 187 spore-formation. What occurred in the cases referred to was probably a temporary dislocation or disturbance of the character- istic organisation of the cells, with the result that pigment pro- duction was suppressed. When the inhibiting conditions were removed the original organisation recovered itself in the course of generations. But there is a great difference between such cases and, let us say, the transmission of sun-burning, or of specially strong muscles, or of a callosity on the skin, or a dwarfed form, which are instances of bodily modifications, technically called acquired characters. In the case of the bacilli the disturbed organi- sation was halved or multiplied in each reproductive process, and the effect originally induced was inherited from generation to generation, eventually disappearing as the restoration of normal conditions allowed the original organisation to re-assert itself in its integrity ; in the case of the supposed inheritance of a callosity we have to assume either that the influence which induced this, or the influence of it after it had been induced, also affected the germinal material in the reproductive organs in such a way that the contained germ-cells, when liberated, developed into an organism with more or less of the callosity. It must be evident, without further dis- cussion, that the cases are not at all on a par, and that inheritance in unicellulars has not been considered with sufficient carefulness even by experts. Prof. L. Errera (1899) reported an experiment with a simple but multicellular mould (Aspergillus niger), which adapted itself to a medium more concentrated than the normal. The second generation of the mould was more adapted than the first, and the adaptation to the concentrated medium was not wholly lost after rearing in the normal medium again. This looks like evidence of the inheritance of the acquired adaptive quality which was brought about as a direct modification. But the case does not really help us, since the distinction between soma and germ-plasm is not more than incipient in the mould in ques- tion. And even if the distinction were more marked, it would only show that the germ-plasm is capable of being affected along with the body, by a deeply saturating influence, which nobody as ever denied. 1 88 TRANSMISSION OF ACQUIRED CHARACTERS Misunderstanding YII — Changes in the germ-cells along with changes in the body are not relevant. — Another misunderstanding is due to a failure to appreciate the distinction between a change of the reproductive cells along with the body, and a change in the reproductive cells conditioned by and representative of a particular change in bodily structure. The supporters of the hypothesis that modifications may be transmitted point to the tragic cases where some poisoning of the parent's system, by alcohol, opium, or some toxin, is followed by some deteriora- tion in the offspring. There is no doubt as to the fact; the question is as to the correct interpretation. (1) In some cases it may be that the whole system of the parent is poisoned — reproductive cells as well as body ; the effect may be as direct on the germ-cells as on the nerve-cells. These, therefore, are not cases on which to test the transmissi- bility of an acquired character — i.e. of a particular somatic modification. If a local poisoning had a structural effect on some particular organ, and if that structural effect was repro- duced in any degree in the offspring, the case would be relevant ; but when the whole organism is soaked in a poison the case is irrelevant. If it could be said that the sunshine, which brings about sun-burning in the skin, soaks through the organism even to its reproductive cells and specifically affects them, in a manner analogous to the saturating poison, we should have a physiological basis for expecting the inheritance of sun-burning. But we cannot make this assumption. We have no warrant for believing that the modification of a part re-echoes in a definite specific way through the organism until even the penetralia of the germ-cells reverberate. (2) A parent organism is poisoned, and there are structural results of that poisoning. The offspring are born poisoned, and show similar structural peculiarities. This may be due to the fact that the germ-cells were poisoned along with the parental body ; but it may also be due, in the case of a mother, CHANGES IN GERM*CELLS 189 to a poisoning of the embryo before birth, in a manner comparable to pre-natal infection. (3) In some cases — e.g. of alcoholism in successive generations — - there may be poisoning of the germ-cells along with the body, there may be poisoning of the embryo before birth, and of the infant after ; but it may also be that what is really inherited is a specific degeneracy of nature, an innate deficiency of control, perhaps, which led the parent to alcoholism, and which may find the same or some other expression in the child. Cases are known in which the children of a dipsomaniac father and a quite normal mother have exhibited a tendency to alcoholism, insanity, and the like. In this case the possibility of poisoning the unborn child is eliminated, but there remain three possibilities of interpretation, — that there was specific poisoning of the paternal germ-cells; that what was inherited was the constitutional weakness which expressed itself as alcoholism in the father ; and that there were detrimental influences in the early nutrition, environment, education — " nurture," in short — of the offspring. But while we have admitted a good deal, we have not admitted the transmissibility of a particular structural modification brought about in the parental body as a result of the toxin. An illustration of what we mean by the distinction " along with, but not through the body," is afforded by an experiment of Paul Bert's. He tried to acclimatise some Daphniae (small fresh-water crustaceans) to salt water by gradually adding salt to the aquarium. At the end of forty -five days, when the water contained 1*5% of salt, all the adults had died ; but the eggs in their brood-chambers survived, and the new generation arising from these flourished well in the salt medium (cit. Packard, 1894, p. 345). Packard sees in this case an argument for the heritability of a modification, but it seems to us merely an instance of the direct modification of the germ-cells or of the embryos. Cuenot, whom Packard cites, gives the correct interpretation : " This experiment shows with admirable clearness that the germ-plasm has, owing to the modifi- igo TRANSMISSION OF ACQUIRED CHARACTERS cation, become accustomed to the salt, causing it to produce a generation so different from the preceding." Misunderstanding YIII — Failure to distinguish between the possible inheritance of a particular modification and the possible inheritance of indirect results of that modification, or of changes correlated with it. — At first sight this seems hair-splitting, but it is a crucial point. Through his vigorous exercise the blacksmith develops a muscular arm worthy of admiration ; the shoemaker acquires skeletal and muscular peculiarities less admirable. There are many permanent and profound modifications associated with particular occupations. Are we to believe, it is asked, that the occupation of the parents has no influence on the offspring ? Are we to believe, it is asked, that the children of soldier, sailor, tinker, tailor, are in no way affected by the parental functions ? It would be interesting to have precise data in regard to this, but it is generally admitted that when parents have healthful occupations their offspring are likely to be more vigorous. The matter is complicated by the difficulty of estimating how much is due to good nurture before and after birth. It is not unlikely, too, that some profound parental modifications may influence the general constitution, may even affect the germ- cells, and may thus have results in the offspring. But unless the offspring show peculiarities in the same direction as the original modifications, we have no data bearing precisely on the question at issue. A belief in the inheritance of modifications was perhaps expressed in the old proverb, " The fathers have eaten sour grapes, and the children's teeth are set on edge " — a proverb which Ezekiel with such solemnity said was not any more to be used in Israel. Now if " setting on edge " was a structural modification, and if the children's teeth were " set on edge " as their fathers' had been before them, there would be a pre- sumption in favour of the transmission of this acquired character, though it would be still necessary to inquire carefully whether INDIRECT RESULTS OF MODIFICATIONS 191 the children had not been in the vineyard too. But if, as Ro- manes said, the children were born with wry necks, we should have to deal with the inheritance of an indirect result of the parents' vagaries of appetite, and not with any direct repre- sentation in inheritance of the particular modification produced in the paternal dentition. Misunderstanding IX — Appealing to data from not more than two generations. — It has often been pointed out that animals transported to a new country or environment may exhibit some modification apparently the result of the novel influence, and that their offspring in the same environment may exhibit the same modification in a greater degree. Thus sheep may show a change in the character and length of their fleece, and their progeny may show the same change more markedly. ' But it is perfectly clear that if the evidence does not go beyond this, nothing is proved that affects the question at issue. It was to be expected that the offspring should sho'w the modifica- tion in a more marked degree than their parents did, since the offspring were subjected to the modifying influences from birth, whereas their parents were influenced only from the date of their importation. What would be welcome is evidence that the third generation is more markedly modified than the second ; then there would be data worth considering. Only then would it be necessary to consider Weismann's somewhat subtle discussion as to the influence of climate. § 5- Various Degrees in which Parental Modification might affect the Offspring It may seem, at first sight, unscientific to discuss various hypothetical degrees in which parental modifications might affect the offspring, when we do not know that modifications can be in any degree transmitted. But unless we are greatly mis- 192 TRANSMISSION OF ACQUIRED CHARACTERS taken, our theorem, if carefully attended to, will serve to make the issue clearer. In regard to germinal variations, whose transmissibility is undoubted, it is well known that there may be different degrees of transmission, or, more cautiously stated, that the offspring's hereditarily determined reproduction of the parental variation may have diverse expressions. It seems just, therefore, to imagine that there might be different degrees in the transmission of modifications. (1) The first degree of transmissibility would be illustrated if the offspring showed in any measure the same modification as the parent had acquired. If the sun-burnt parent had a congenitally swarthy child, that might be an indication of modification-transmission of the first degree of directness. It might be an illustration of what has been so carefully searched after — the transmission of a particular acquired character. We cannot too strongly emphasise that this and nothing else is what Weismann has denied ; this and nothing else is the crux of " the interminable argument." And for the sake of argument, the possibility (i) must be kept quite distinct from the possibilities (2) and (3). (2) If the offspring exhibited a new character, not the same as the parent's acquired modification, but affecting similar tissue, though in a different fashion, we might be justified in speaking of this as modification-transmission of the second degree of directness. It might be an illustration, not of the inheritance of a particular acquired character, but of something correlated therewith, if the much sunburnt parent of a thoroughly blond stock had a child with very dark hair on a very white skin. But the inference would not be certain. (3) If the offspring exhibited a novel character, analogous to a modification, yet neither similar to the modification acquired by the parent nor affecting the same region of the body, it might be said that we had to deal with modification-inheritance of AFFIRMATIVE ANSWER 193 the third, degree of directness. It might be an illustration of the inheritance of an indirect effect of a parental modification if the sons of fathers who had eaten sour grapes had wry necks. But we should require many instances before admitting the hereditary nexus. § 6. The Widespread Opinion in favour of Affirmative Answer It seems to be a widespread opinion that acquired characters may be transmitted, but often the opinion wavers when it is explained what this precisely means — namely, that a modification in the body, brought about by a change in function or environ- ment, may so specifically affect the reproductive elements that when these develop there is in the offspring something corre- sponding to the parental modification. Opinion of " Practical Men." — In fairness we must admit that the verdict of the practical man, whether physician or breeder, gardener or farmer, is still in many cases an unhesitatingly affirmative answer. One of the keenest of physicians has said that a few months in practice would dispel all doubt as to the inheritance of acquired characters ; but there are equally keen physicians who have taken a different view. It may also be that the first had not freed himself from Misunder- standings V and VII. Prof. Brewer, an American authority on breeding, who gives an emphatic affirmative answer, notes : " The art of breeding has become in a measure an applied science ; the enormous economic interests involved stimulate observation and study, and what is the practical result ? This ten years of active promulgation of the new theory has not resulted in the conversion of a single known breeder to the extent of inducing him to conform his methods and practice to the theory. My conclusion is that they are essentially right in their deductions founded on their experience and observations — namely, that ao 194 TRANSMISSION OF ACQUIRED CHARACTERS quired characters may be, and sometimes are, transmitted, and that the speculations of the Weismann school of naturalists are unfounded." But perhaps this widespread opinion does not mean so much as it seems ; for it is very difficult to get busy practical men to take the trouble to appreciate an exact distinction such as is involved in the phrase, " the inheritance of an acquired character." Against the opinion quoted we may balance that of an ex- perienced botanical physiologist, Prof. MacDougal. "Despite general assertions to the contrary, no evidence has yet been obtained to prove that the influence of tillage, ' cultivation/ or the mere pressure of environment factors has produced any permanent changes in hereditary characters of unified strains of plants." Great Yariety of Opinion. — There is little to be gained by a citation of opinions, for there are equally authoritative names on both sides. But there are some points of interest. Thus we have already noticed that the scepticism as to the inheritance of acquired characters is not a modern fad. It is also note- worthty that, while the majority of zoologists disbelieve in modi- fication-inheritance, the reverse seems to be the case with botanists. Is this because modifications are even more marked and more recurrent in plants than in animals, or because the distinction between soma and germ-plasm is much less definite in plants than in animals ? But there is this use at least in noting the discrepancy of opinions, that we are warned from dogmatism. It cannot be an easy question when we find Spencer on one side and Weismann on the other, Haeckel on one side and Ray Lankester on the other, Turner on one side and His on the other, and so on. Herbert Spencer was so convinced that he went the length of writing : " Close contemplation of the facts impresses me more strongly than ever with the two alternatives— either there REASONS FOR AFFIRMATIVE ANSWER 195 has been inheritance of acquired characters, or there has been no evolution." * Haeckel is so convinced for the affirmative that he stakes his particular form of religion upon it, asserting that " belief in the inheritance of acquired characters is a necessary axiom of the monistic creed " ; and what may sound to some even more serious is his declaration that, rather than agree with Weismann in denying the inheritance of acquired characters, " it would be better to accept a mysterious creation of all the species as described in the Mosaic account." Sir William Turner has said that " to reject the influence which the use and disuse of parts may exercise, both on the individual and on his offspring, is like looking at an object with only a single eye " — which is not perhaps a very emphatic condemnation, since most microscopic research is monocular. Moreover, the doyen of British anatomists does not state the case with his usual precision. Why is the Affirmative Position so widely held? — Even in regard to our own muscular and nervous systems, we are familiar with illustrations of the fact that practice increases capacity, and that desuetude is apt to be followed by loss of power. A force de forger on dement forgeron. Organs improve with the using and deteriorate in disuse. We are also well aware that changes in the environment or conditions of life, and notably in our food, cause changes in our body. It seems a " natural " assumption to suppose that these gains and losses and changes may be in some degree transmissible. Apart from the " naturalness " of this assumption, there are probably four reasons why the affirmative position is so widely held: (i) There are many facts which suggest modification-inheritance * The italics are ours. See Herbert Spencer. " The Inadequacy of Natural Selection," Contemporary Review, February and March, 1893. Appendix B, Principles of Biology, 2nd ed. vol. i. 1898, p. 621- ig6 TRANSMISSION OF ACQUIRED CHARACTERS until they are examined critically. The late Duke of Argyll, in one of his scientific excursuses, said the world was strewn with illustrations of the inheritance of acquired characters, and Dr. W. Haacke, a very wide-awake evolutionist, has compared the evidences for the affirmative to the sand on the sea-shore for multitude, yet neither furnishes us, so far as we are aware, with a single case that will bear analysis. The affirmative may be an obvious interpretation of the results of evolution, but the ob- vious interpretation is seldom the right one. The sun does not go round the earth. (2) The affirmative is an interpretation which seems to make the theory of organic evolution simpler ; it suggests a more direct and rapid method than the natural selection of germinal varia- tions. If to a growing and varying nature or germinal inheritance there were continually being added the results of peculiarities in nurture, the rate of evolution would be quickened, both upwards and downwards. But our first business is to find out whether the hypothesis actually consists with experience. Dr. Walter Kidd has argued carefully and ingeniously that all departures of hair-direction from a simple and primitive type may be interpreted as due to mechanical causes, namely, stimuli repeated immensely often. The difficulty here and always is with the presuppositions of the interpretation. (3) We are so accustomed in human affairs to the entailment of acquired gains from generation to generation, to standing on the shoulders of our ancestors' achievements, that many find it difficult to refrain from projecting this on 'organic nature. They forget that the greater part of our entailing process comes about through our social heritage, which is altogether apart from our natural inheritance. \ (4) A fourth reason is that many fictitious or anecdotal cases of the inheritance of acquired characters continue circulating. The inheritance of a letter branded upon the arm, which Aristotle notes, is still in the popular currency, though it is perhaps an GENERAL ARGUMENT AGAINST 197 extreme type of what His calls a handful of anecdotes. It is reported that Sioux Indians tattoo discs on the cheekbone prominences of their squaws, and it is said that similar marks may be seen on some new-born children (Nature, in., 1870, p. 168). And besides fictitious cases there are some puzzling phenomena, which the supporters of the negative position are wont to dismiss as " coincidences "—which, it must be confessed, is never a very satisfactory way of dealing with difficult cases. § 7. General Argument against the Transmissibility of Modifications Most of the evidence brought forward in support of the belief in the inheritance of acquired characters is terribly anecdotal ; but apart from this Weismann was led to a position of entire scepticism by his realisation of the continuity of the germ-plasm. The Apartness of the Germ-cells. — If the germ-plasm or the material basis of inheritance be something relatively apart from the body, and from its everyday metabolism, something often segregated at a very early state in development, there is a pre- sumpjion against its being readily affected in a specific manner by detailed exogenous changes wrought on the structure of the body. It seems accurate to say that the reproductive cells which have the potentiality of becoming offspring never arise from differentiated body-cells. Whether they are recognisable as such, late or early, the germ-cells are simply those cells which retain in all its integrity the complex, definite, and stable organi- sation of the fertilised ovum from which the whole organism develops. They have their power of reproducing creatures more or less like the parents just because they are continuous, through an unspecialised cell-lineage, with the fertilised ovum from which the parental body arose. All the somatic cells are, of course, likewise the progeny of the fertilised ovum, but in their lineage there is differentiation and specialisation. We imagine that in them the numerous items or potentialities in the fertilised 198 TRANSMISSION OF ACQUIRED CHARACTERS ovum are distributed and allowed to express themselves. In the germ-cell lineage they are kept concentrated and latent.* In any case the germ-cells in the reproductive organs are not actively functioning elements of the body ; they are in a quite peculiar way apart from the general soma ; and Weismann has reasonably emphasised the difficulty of picturing any means whereby the modification of a particular corner of the body can react upon the germ-cells in a manner so specific that these can, when they develop, reproduce the particular parental modification or any approach to it. This argument, and the answers to it, must be carefully considered. i. The Germ -cells may be affected by the Body. — In the first place, it has been answered that the body does undoubtedly, in some cases, exert some influence on the gonads, so that the difficulty is reduced to this : Can a modification of part of the body exert a specific or representative influence on the germinal material ? But what is the precise nature of the alleged influence of the body on the gonads ? It is pointed out that nervous changes can excite the reproductive organs, that food-stuffs may increase their activity, that alcohol and other stimulants may influence them, and so on. But there is a great difference between any such excitation of the gonads and the propagation of a particular modification, let us say, from the skin to the germ-cells. And there is a great difference between a poisoning of the germ-cells along with the body, and the influencing of them in a manner so specific that they can, when they develop, reproduce the particular parental modification. (See Misunderstanding VII.) * In certain conditions, as yet unknown, certain body-cells may revert to a primitive mode of behaviour — like some kinds of criminals in society. Thus the cells which develop into cancerous growths behave in some ways like germ-cells, especially in their mode of division. (See the researches of Farmer, Walker, and Moore.) But such cases need not lead us to Hertwig's extreme conclusion that every cell is potentially a germ -cell. DARWIN'S AND SPENCERS THEORIES 199 2. Hypotheses as to Possible Mechanism of Transmission. — In the second place, attempts have been made to construct hypotheses by aid of which we might conceive how a modifi- cation of, say, the skin, can exert a specific or representative influence on the germinal material. Thus, Darwin suggested his provisional hypothesis of pan- genesis, according to which the parts of the body give off gemmules which pass as samples to the germ-cells. But his suggestion remains a pure hypothesis — and an unnecessary one unless new facts come to light — and is nowadays maintained by no one except in extremely modified form — e.g. in the Pangen-theory of De Vries. Spencer deserves credit for at least facing the difficulty of conceiving a modus operandi whereby a particular modification in, say, the brain or the thumb, can specifically affect the ger- minal material in such a way that the modification or a tendency towards it becomes involved in the inheritance. Briefly stated, his theory is as follows : Spencer's Theory of the Mechanism of Transmission. — Spencer made the legitimate postulate that, intermediate between the biological unit or cell and the chemical molecule, there were " con- stitutional units," the vehicles of specific characters, ancestral and parental traits, and the individual peculiarities of the organism itself. He supposed that they were very stable in their " fundamental traits," but plastic as regards their " superficial traits." He supposed that they had " such natures that while a minute modification, representing some small change of local structure, is inoperative on the proclivities of the units throughout the rest of the system, it becomes operative in the units which fall into the locality where that change occurs." He supposed " an unceasing circulation of protoplasm throughout an organism," such that, " in the course of days, weeks, months, years, each portion of protoplasm visits every part of the body " — a wild assumption. Finally, " we must conceive that the complex forces of which 200 TRANSMISSION OF ACQUIRED CHARACTERS each constitutional unit is the centre, and by which it acts on other units while it is acted on by them, tend continually to re-mould each unit into congruity with the structures around, superposing on it modifications answering to the modifications which have arisen in these structures. Whence is to be drawn the corollary that in the course of time all the circulating units — physiological, or constitutional, if we prefer so to call them — visit all parts of the organism ; are severally bearers of traits expressing local modifica- tions ; and that those units which are eventually gathered into sperm-cells and germ-cells (i.e. egg-cells), also bear these superposed traits." Thus the constitutional units are supposed to circulate and to visit one another throughout the body. When they come to a modified structure and visit its modified constitutional units, they are supposed to be themselves impressed ; thus impressed, they are supposed to be gathered into the germ-cells, which thus come to bear the " superposed traits " resulting from modifications. If we were sure that modifications were ever transmissible, we might be glad of this hypothetic interpretation of the business. But it is a difficult hypothesis to think out, and it would hardly be tolerable even if there were facts which it was needed to interpret. In particular, the conception of "an unceasing circulation of protoplasm," so that " each portion of protoplasm visits every part of the body," seems not only unwarranted, but contradicted by well-established facts. 3. A Mechanism may exist though it remains Unknown.— In the third place, we must recall Prof. Lloyd Morgan's warning that although we cannot imagine how a modification might, as such, saturate from body to germ-cells, this does not exclude the possibility that it may actually do so. Oscar Hertwig also maintains that our ignorance of any mechanism which could secure the transmission of an acquired character is not a good argument against the possibility of its occurrence. There are, he says, many facts in biology which are quite secure, though no causal nexus can be worked out at present (All- gemeine Biologie, 1906, p. 621). It must be noted, however, A CONCRETE CASE: SPENCERS HANDS aoi that, so far as we can understand, a very complex and special mechanism would be necessary if a modification in, say, the eye is specifically to affect the germinal material. Dr. George Ogilvie (1901) writes : " In a subject so involved in obscurity the present incomprehensibility of certain relations can hardly serve as an argument against their existence. The development of the apparently uniform germ-plasm into the infinite differentiation of a complex cell-state is, although no longer a matter of doubt, perhaps not less inconceivable." But this illustration is not altogether appropriate, since our inability to conceive the precise " how " of development rests on our inability to restate in simpler terms any of the fundamental facts of life, such as growth, assimilation, or reproduction, whereas the supposed relation between soma and germ-cells is inconceiv- able in rather a different sense. A better illustration, it seems to us, would be found in the difficulty of exactly stating how particular changes in the gonads are correlated with particular changes in the body — e.g. in the changes associated with puberty, conception, ovarian and testi- cular disease. Yet here we can at least imagine what the general nature of the physiological nexus may be — in terms, for instance, of internal secretions or " hormones." A Concrete Case : Spencer's Hands.— It may illumine the abstract argument to take a concrete case. Why had Herbert Spencer small hands ? He says that it was because his grand- father and father were schoolmasters, who did little manual work from day to day, save in wielding the pen and sharpening the pencil. Through disuse of the sword and the spade their hands were " directly equilibrated " towards smallness. But since Mr. Spencer senior was " a combination of rhythmically acting parts in moving equilibrium," the dwindling of the hands and the moulding of the physiological units thereof reverberated through the whole aggregate ; a change towards a new state of equilibrium " was propagated throughout the parental system — 202 TRANSMISSION Of ACQUIRED CHARACTERS a change tending to bring the actions of all organs, reproductive included, into harmony with these new actions," or inactions. The modified aggregate impressed some corresponding modifica- tion on the structures and polarities of the germ-units. And this was how Herbert Spencer had small hands. At least, so he tells us. Disuse of Parts. — It seems " natural " to suppose that organs have dwindled part passu with their disuse, and because of their disuse. But the two statements are not synonymous. The dwindling may be due to germinal variations in the line of reduc- tion, which are appropriate because of some change in the ani- mal's habits and environment. It may even be that the organism meets an endogenous reduction of certain parts by itself changing its habits and habitat. Moreover, it is important to notice, as Emery, Kennel, and Ziegler have pointed out, that there has pro- bably been a " Kampf der Theile im Organismus " (a struggle of parts within the organism) not merely in individual ontogeny, but also in the racial phylogeny. Dwindling of one part occurs when some adjacent part attains increased differentiation. " Thus snakes have not lost their limbs because they did not use them, but because of their evolution in the direction of excep- tionally large trunk and tail musculature. In man the strong dentition of his Simian forebears has become weaker, not through disuse, but because the extraordinary increase of the brain has been correlated with a weaker development of other parts of the head " (H. E. Ziegler, 1905, p. 3). § 8. General Argument for the Transmissibility of Modifications The Germ-cells are not Insulated. — While it must be ad- mitted that the germ-cells have a certain apartness from the daily life of the body, and that they are unspecialised cells that have not shared in the differentiation characteristic of the body- cells, is there not some risk of exaggerating the distinction between somato-plasm and germ-plasm ? GENERAL ARGUMENT FOR 203 In many simple animals, such as sponges and hydroids, the germ-cells simply make their appearance at certain times of year among the commonplace somatic cells. In many plants the distinction between body and germ-cells can hardly be drawn until the period of reproduction sets in. Thus Spencer refused to accept the contrast between body-cells and germ-cells as expressing a fact, and referred to the numerous cases in which small pieces of a plant or a polyp may grow into entire organisms. To this objection Weismann answers, — (i) that the* distinction between somatic cells and germ-cells has been gradually em- phasised in the course of evolution, and that in the simpler multicellular organisms it is still incipient ; (2) that it is quite conceivable that, even in some complex organisms, the body-cells, though differentiated, may retain some residual unused germ- plasm ; and (3) that there may be a quite definite and distinct germ-plasm, though there is no demonstrably distinct lineage of germ-cells. Again, however, we must remember that the blood, or lymph,, or other body-fluids form a common medium for all the parts of the animal, gonads included ; the results of changes in nutrition may saturate throughout the body and affect the germ-cells inter alia. The nervous system makes the whole organism one in a very real sense ; in plants there are often intercellular bridges of protoplasm binding cell to cell, and this is true in not a few cases among animals. Moreover, there are subtle, dimly understood correlations between the reproductive organs and the rest of the system. If changes in the reproductive organs can effect changes in remote parts, such as the larynx and the mammary glands, why may not there be reciprocal influences ? In short, the organism is a unity, and to divide it up, in any hard- and-fast way, into soma and germ-cells may land us in the same fallacy as parcelling the mind into separate faculties. It must be admitted, therefore, that it is quite erroneous to thjnk of the germ-cells as if they led a charmed life, uninfluenced 204 TRANSMISSION OF ACQUIRED CHARACTERS by any of the accidents and incidents in the daily life of the body which is their bearer. But no one believes this, Weismann least of all, for he finds the chief source of germinal variations in the stimuli exerted on the germ-plasm by the oscillating nutritive changes in the body. Weismann's Concessions — There are some who find in this " a concealed abandonment of the central position of Weismann," and who say : "If the germ-plasm is affected by changes in nutrition in the body, and if acquired characters effect changes in nutrition, then acquired characters or their consequences will be inherited." But it is quite illegitimate (§ 5) to slump acquired characters and their consequences as if the distinction were immaterial. The illustrious author of The Germ-Plasm has made it quite clear that there is a very great difference between admitting that the germ-plasm has no charmed life, insulated from bodily influences, and admitting the transmissibility of a particular acquired character, even in the faintest degree. The point, let us repeat, is this : Does a structural change in a part of the body, induced by use or disuse, or by change in surround- ings, influence the germ-plasm in such a specific or representative way that the offspring will thereby exhibit the same modification that the parent acquired, or even a tendency towards it ? The Real Difficulty. — Even when we recognise, as fully as we can, the unity of the organism, that each part shares in the life of the whole, it is very difficult to think of any modus operandi whereby a local modification can specifically affect the germ- plasm. The argument that we can as little understand the modus operandi whereby an influence passes from the gonads to distant parts of the body is not really sound. For we know that in some cases the reproductive organs, besides being areas for the multiplication of germ-cells, are organs of internal secre- tion, producing specific substances which are carried away by the blood-stream, and serve as the stimuli awakening the dormant potentialities of distant parts. REAL DIFFICULTY 205 Nor does the fact that morbid processes in a particular part may result in a diffusion of toxins, which saturate even the germ- cells, help us much in our attempt to picture how a modification could become transmissible. For there is not the slightest reason for supposing that the ordinary modifications in which naturalists are interested, which experimental evolutionists can bring about, are associated with the formation of specific toxins which might diffuse through the whole system. Spencer's Statement of the a priori Argument. — As Herbert Spencer was perhaps the keenest and most convinced upholder of the affirma- tive position, it seems just to give his statement of the a priori argument. We have made a comment on each of the steps. (1) "That changes of structure caused by changes of action must be transmitted, however obscurely, appears to be a deduction from first principles — or if not a specific deduction, still, a general implication." "For if an organism, A, has, by any peculiar habit or condition of life, been modified into the Form A1, it follows that all the functions of A1, reproductive function included, must be in some degree different from the functions of A." "An organism being a combination of rhythmically acting parts in moving equilibrium, the action and structure of any one part cannot be altered without causing altera- tions of action and structure in all the rest." Comment. — (a) It is not denied that some deeply saturating modifications of the body, affecting the nutritive stream, may affect the reproductive organs. This is not the point at issue, (b) How far a modification is likely to affect the reproductive organs must be determined by observation and experiment. The appreciability of the change will depend on the amount and nature of the modification, and on the intimacy of the correlation subsisting in the organism. Dislodging a rock may alter the centre of gravity of the earth, but it does not do so appreciably. (2 ) " And if the organism A, when changed to A1, must be changed in all its functions, then the offspring of A1 cannot be the same as they would have been had it retained the form A." Comment. — This is logical, but is it true ? The change from A to A1 may be important, it may appreciably alter the 206 TRANSMISSION OF ACQUIRED CHARACTERS metabolism, but it does not follow that it can appreciably alter the architecture of the germ-plasm. Spencer's as- sumption that the change in the constitutional units of the body must affect the constitutional units in the germ- cells remains an assumption. (3) "That the change in the offspring must, other things equal, be in the same direction as the change in the parent, appears implied by the fact that the change propagated throughout the parental system is a change towards a new state of equilibrium — a change tending to bring the actions of all organs, reproductive included, into harmony with these new actions." Comment. — It seems to us to pass the wit of man to conceive how or why an improved equilibrium in, let us say, the use of the hand should involve any corresponding or represen- tative change of equilibrium in the germinal material. The drawback to abstract biology based on first principles is that it enables its devotees to develop arguments which seem plausible until they are reduced to the concrete. § 9. Particular Evidences in support of the Affirmative Answer The question is whether modification-inheritance does or does not occur, and we must no longer postpone our consideration of the concrete evidence used to support the affirmative position. Our reason for not placing this section in the foreground of the chapter is mainly that a multitude of misunderstandings have had to be cleared away before the so-called direct evidence could be profitably considered. When one naturalist, Dr. W. Haacke, declares that instances of modification-inheritance are as plentiful as sand on the shore, and another, Prof. E. Ray Lankester, declares that the Lamarckian position has its only remaining defence, and that no secure one, in Brown-Sequard's experiments, we have obvious justification for our preliminary discussion. The instances adduced as evidence of modification-inheritance might be classified according to the errors involved, but we have arranged them rather in reference to the general nature of the modifications discussed, whether environmental or functional, IMPROVEMENT IN TROTTING HORSES 207 whether tending to increase or decrease, and so on. The alleged inheritance of the direct effects of mutilations, injuries, and the like is discussed separately in §§ 10 and u. Improvement in Trotting Horses. — Over a hundred years ago (1796) the utmost speed of the English trotter was stated at a mile in 2 min. 37 sec. Since 1818, accurate records have been kept, which show a gradual increase decade after decade in the speed and in the percentage of swift trotters. The standard has risen and the breed has improved. The mile can now be run in 2 min. 10 sec., or less. It is claimed by Cope and others that we have here direct evidence of the trans- mission of the structural results of exercise. Brewer (cit. Cope, 1896, pp. 426-30) relates that about 1818 the record speed of the trotting horse was 3 min. to the mile ; in 1824 it was reduced to 2 min. 34 sec. ; in 1848, to 2 min. 30 sec. ; in 1868, to 2 min. 20 sec. ; in 1878, to 2 min. 16 sec. ; in 1888, to 2 min. nj sec. ; and finally to 2 min. 10 sec. " The gain in speed has been cumulative. ... It has gone on along with systematic exercise of special function in suc- cessive generations ; . . . there is nothing that would lead to even suspect that the changes due to exercise of function had not been a factor in the evolution ; ! ! ! there is every appearance and indication that the changes acquired by in- dividuals through the exercise of function have been to some degree transmitted, and have been cumulative, and that this has been one factor in the evolution of speed." It is impossible to prove the negative above suggested — namely, that function has not been a factor ; but the affirmative position is robbed of all cogency by the admitted occurrence of rigorous artificial selection. The improvement supposed to be entailed may not have been a modification at all ; but, supposing it was, the interpretation of the result simply by the hypothesis of use-inheritance gives a false simplicity to the case. It overlooks the selective breeding which increases the constitutional swiftness, 208 TRANSMISSION OF ACQUIRED CHARACTERS and the process of elimination which persistently weeds out the less swift from the stud. And even apart from artificial selection and elimination there may be a progressively cumulative suc- cession of variations making for greater and greater swiftness. We may even picture how this might come about, if we adopt Weismann's conception of germinal selection. Case of Squatting Punjabis. — It has been stated that the Punjabis of India show certain peculiarities of musculature and skeleton which are associated with the frequency with which these people assume on all possible occasions the squatting posture. It is asserted that the peculiarities of structure are due to the peculiarities of function, but this requires definite proof (Misunderstanding III). They may be adaptations origina- ting in germinal variations. It is necessary to know whether the peculiarities are in any degree represented on new-born Punjabi babies, but even then it would be simpler to regard them as variations than as transmitted modifications. There can be no conclusiveness in regard to peculiarities whose first appearance is hidden in obscurity. If squatting increased from generation to generation, and if the structural peculiarities increased pari passu, the case would be interesting ; but even then we should have to inquire whether we were not dealing with a progressive variation. Peculiarities of Occupations. — In his interesting paper on the anatomy of the shoemaker, Dr. Arbuthnot Lane describes the peculiarities induced by this occupation, which tends to form a distinct anatomical type. The same is true of the tailor. " The bent form, the crossed legs, thumb-and-forefinger action, and peculiar jerk of the head while drawing the thread, are the main features of the sartorial habit," and they are associated with permanent changes in muscles, insertion surfaces, and articulations. These are indubitable modifications : what of their transmission ? No one, Dr. Lane says, would expect any perceptible changes in the first generation, but he thinks that he has observed inherited effects in the third. LARGE AND SMALL HANDS 209 We can only say that this line of inquiry deserves to be followed up, especially since our minute acquaintance with the human body and the accumulation of facts in regard to its varia- tions make a discrimination between modification and variation more secure than is possible in many other cases. It should be remembered, however, that if the shoemaker's sons and grandsons and subsequent descendants all " stuck to the last," there might tend to be an accumulation of general constitutional peculiarities — e.g. of meditativeness and of the physical effects of persistent sedentary work, which might dispose the organism to re-acquire particular modifications in a more marked degree. Large and Small Hands. — Darwin (Descent of Man, p. 18) refers to the alleged fact that the infants of labourers have larger hands than those of the children of the gentry ; but this, and many similar cases of which it is a type, may be sufficiently accounted for by interpreting the observed differences as constitutional char- acteristics of different stocks probably accentuated by various forms of selection. Spencer notes, " That large hands are in- herited by those whose ancestors led laborious lives, and that those descended from ancestors unused to manual labour com- monly have small hands, are established opinions." But if we accept the " opinions " as correct, it is easy to interpret the size of the hands as a stock character correlated with different degrees of muscularity and vigour, and established by selection. The hands of Japanese are in many details anatomically different from the hands of Europeans, but there is no warrant for regarding these detailed differences as other than constitutional racial differences of germinal origin accentuated modificationally in the individual lifetime. Dwindling of Little Toe. — The alleged dwindling of the little toe has been repeatedly cited as a case in point — proving the inheritance of a modification produced by tight boots. But precise data are wanting ; a dwindling has also been observed in savages who do not wear boots ; it is possible that there may 210 TRANSMISSION OF ACQUIRED CHARACTERS be in man, as there was in the ancestors of the modern horse, a constitutional variation in the direction of reducing digits ; and there are other possible explanations of the rather vague assertions. It need hardly be pointed out that unless there is a measurably progressive dwindling with similar boots in the course of generations the case has no point. A control experi- ment comparing the toes in sets of brothers respectively booted and bootless would be interesting. Results of Pressure. — Darwin (Descent of Man, p. 18) regards the thickened sole of even unborn infants as due to " the in- herited effects of pressure during a long series of generations." But here again it is impossible to exclude the interpretation that a variation in the direction of thickened solar epidermis might have selection-value from very ancient days, to the arboreal ape as well as to the bootless man. H. H. Wilder, in a paper in which he gives a detailed comparison of the palms and soles of Primates and Man (Anat. Anzeig. xiii. (1897), pp. 250-6)," distinctly refuses to commit himself to a Lamarckian theory, believing that the facts may be equally well interpreted in terms of variation and selection. Bellinger (1882) suggests that the weak development of the breasts in women of the Dachauer district is due to the old- established fashion of wearing tight corsets which are pressed flat on the breasts. It is necessary to inquire (a) whether the pecu- liarity is not a modification inflicted on each successive generation, or whether it is ever exhibited by a Dachauer woman who does not wear a corset ; and (b) whether the same peculiarity does not occur where the fashion is entirely different. Climatic Changes. — Virchow and others have laid stress on the fact that many peculiarities in races of men and of other living creatures are climatic in origin, and yet are now part of the natural inheritance. But acclimatisation is usually a slow and gradual process, involving selection of germinal variations, and it is difficult to get clear-cut cases of climatic modifications. PARTICULAR CASES 211 It must also be remembered that Weismann expressly admits that climatic influences, especially if long-continued, may influ- ence the germ-plasm along with the whole system, and may induce germinal variations that come to stay ; but this " has certainly nothing to do with the view that functional modifications of any particular, organ can cause a corresponding change in the germ-plasm." (See The Germ-Plasm, 1893, p. 408.) In adjacent areas with different climatic and other environ- mental conditions we not infrequently find closely related species or local races. It seems impossible to doubt that these are blood-relations, derived from a common ancestor. Are the}' not due to the environmental differences ? In some way, surely, the organismal differences are causally correlated with the environmental differences, and it is granted by all that pecu- liarities of climate induce changes in the nutrition, respiration, circulation, and so on. If so, the germ-plasm may be affected and variations may be provoked, some of which are adaptive. But the result of these variations may be something different from and much more profitable than the modifications directly induced. They may be expressed in relation to quite different organs. Thus it seems quite unnecessary to believe in the trans- mission of climatic modifications as such, or in any representa- tive degree. Moreover, we must never forget that the active organism must be credited with the power of seeking out en- vironments which suit its inborn nature — variations included. Plants in New Environment. — Much has been made of the changes which follow a radical change of environment. When a plant is transferred to a new soil and climate it may undergo a very marked change of habit ; its leaves may become hairy, its stem woody, its branches drooping. "These," Herbert Spencer said, " are modifications of structure consequent on modifications of function that have been produced by modifica- tions in the actions of external forces. And as these modifications reappear in succeeding generations, we have, in them, examples 212 TRANSMISSION OF ACQUIRED CHARACTERS of functionally established variations that are hereditarily trans- mitted. But this is a non-sequitur , since the modifications may reappear merely because they are re-impressed directly on each successive generation. It is Misunderstanding IV. At the same time it should be noted that radical change of en- vironment may induce germinal variations or mutations which breed true. These must be distinguished from modifications, as already explained, since we cannot interpret them physiologically as the direct somatic results of the environmental change. Another case requiring consideration is that of a Turkestan relative of our common Shepherd's Purse (Capsella bursa pastoris}. It has apparently spread from the low country to the uplands, and the specimens growing at the higher altitudes are smaller than those below, and pink instead of white. Seeds of lowland forms sown in the uplands develop into small plants with pink flowers, but the upland forms keep their characters (except the xerophytic leaves) when grown in the low country. It is possible that we have here to do with a variation coincident with a modification ; it seems, however, that the experiments require to be repeated and extended. Experiments on Brine-shrimps. — Reference is often made to the observations and experiments of Schmankewitsch (1875) on certain brine-shrimps belonging to the genus Artemia. By lessening the salinity of the water he was able to transform one type, Artemia salina, in the course of generations into another type, Artemia milhausenii. By increasing the salinity, he was able to reverse the process. Although he did not himself make any such claim, his work has often been referred to as an illus- tration of changing one species into another, and of the inheritance of acquired characters. It seems very doubtful, however, whether we have here to do with modifications at all. Schmankewitsch did not modify any one Artemia salina into Artemia milhausenii ; with a pro- gressively changing environment and in the course of generations he observed a transition of the population from the one type to the other ; it is probable that the change of salinity operated EFFECTS OF CHANGED ENVIRONMENT 213 directly on the eggs. This seems the more likely since the differ- ences between the two types (in shape of tail, details of bristles, etc.) are not such as we can interpret as the natural direct results of altered salinity. It is well known that slight alterations in the physico-chemical composition of the water have sometimes a great and mysterious influence on eggs and developing embryos. FIG. 27. — Side view of male Artemia salina (enlarged). (From Chambers's Encyclopedia.) Bateson and others have shown that there is great variability in the character of the tail and bristles of Artemia salina, of which A. milhausenii seems to be only an extreme form without tail-lobes. FIG. zja. — Tail-lobes of Artemia salina (to the left) and of Artemia mil- hausenii (to the right) ; between these four stages in the transforma- tion of the one into the other. (From Chambers's Encyclopedia ; after Schmankewitsch.) But if the changes were somatic modifications, it is still open to the critic to point out that Schmankewitsch experimented with a progressively changing environment on a series of genera- tions, and that the results were due to modifications hammered afresh on each successive generation, without there being any inheritance of these modifications. A Typical Case. — An often-quoted and typical instance was communicated to Darwin by Moritz Wagner. Some pupae of 214 TRANSMISSION OF ACQUIRED CHARACTERS a Texan species of Saturnia were brought in 1870 to Switzerland. In May, 1871, the moths emerged and were entirely true to type ; they had young, and these were fed on the leaves of Juglans regia (the Texan form feeding on Juglans nigra) ; these young developed into moths so different in colour and form from their parents that some entomologists referred them to distinct species. This was a well-marked individual modification, but the story stops just where it was beginning to be interesting. We are not told about the subsequent generations. If they, too, were fed on Juglans regia, and reared in Switzerland, they probably reproduced the new type, but this would simply mean that the modification was re-impressed on successive generations. Experiments on Lepidoptera. — Standfuss reared pupa of Vanessa urticcz at a lower than the normal temperature, and obtained a northern type (var. polaris) ; he reared them at a temperature higher than the normal, and obtained a southern variety (var. ichnusa). In the progeny he found a very small percentage (all males) which showed a change in the same direction as the parents. Fischer worked with Arctia caja, reared the pupae at 8° C., and obtained some unusually dark forms. Two of these were paired and their progeny was reared at the normal temperature. A small percentage of these— the last of the brood to emerge from the pupa-state — showed the same kind of melanistic pecu- liarity as the parents had shown. Fischer pointed out, however, that the colour-aberration in the offspring was not a repetition of the parental peculiarity, though it was in the same direction and sometimes went farther. He did not regard the case as illustrating the transmission of a specific modification, but agreed with Weismann's interpretation that the germ-cells had been prompted to vary by the lowered temperature. It should also be noted that in many butterflies there is a strong constitutional — i.e. germinal — tendency to melanistic variation, that the aberration does not occur in all the individuals subjected to the low temperature, that it occurs in very diverse degrees, and that the experimenter selected two forms to pair together. EFFECTS OF CHANGED ENVIRONMENT 215 Fresh Experiments. — Among the twentieth century experi- ments on the transmission of modifications, there are a few which suggest that a dogmatic denial of the possibility is very unwise. As a striking instance let us take Kammerer's experi- ments on salamanders. (a) The common yellow and black salamander (Salamandra macu- losa) is either viviparous, producing a large number of larvae 25-30 mm. in length with four limbs and short gills, or ovo-viviparous, lay- ing large eggs which hatch out immediately into similar larvae 2 3-2 5 mm. in length. After a few months of larval life in the water they undergo metamorphosis into land-salamanders 45-56 mm. in length. (b) The black Alpine salamander (Salamandra atra) produces at a birth two fully formed terrestrial young ones 38-40 mm. in length, the larval stage being skipped — in obvious relation to the Alpine conditions of life. (c) Kammerer kept the spotted salamander in the cold and got it, after a few pregnancies, to produce only two young ones, as in the black salamander. (d) He kept the black salamander in a warm place with plenty of water, and got it to produce 3-9 gilled larvae, thus approaching the condition in the spotted salamander. (e) Now the offspring of the salamanders thus treated (c and d) were kept for two and a half years in a vivarium, but did not become sexually mature until they were placed in the open in conditions normal to 5. maculosa. They became mature when three and a half years old. The offspring of (c) gave birth to (i) very advanced larvae, 45 mm. long with much reduced gills, metamorphosing several days after, or moderately advanced larvae, 20 mm. long, with large gills; or (2) to small larvae 26 mm. long with rudimentary gills, laid on land, and metamorphosing after four weeks into salamanders 29 mm. in length. Thus there was a partial persistence of a modified mode of reproduction in the absence of the modifying conditions. (/) The offspring of (d) bore in the water 3-5 larvae, 33-40 mm. or 21-23 mm. in length, light in colour, and possessing gills. In this case there was practically a continuance of the modifying con- ditions and there was an augmentation of the parental modification. The difficulties in regard to these very interesting experiments are : ( i ) they do not deal with a structural modification in the ordinary sense ; (2) it may be that the experimental conditions of 216 TRANSMISSION OF ACQUIRED CHARACTERS (c) and (d) acted directly on the germ cells of the original subjects of experiment ; and (3) there was some measure of artificiality in the conditions under which the second generation developed which may have disturbed the normal routine of reproduction. Breeders' Evidence. — The evidence given by breeders in sup- port of the theory of modification-inheritance, which is a tacit or an avowed belief of many, if not of most, appears to us in most cases too full of vagueness and misunderstanding to be of signifi- cance ; but it has been often adduced by expert biologists, notably by Cope (1896), who cites his cases from Brewer (1892-3), an acknowledged agricultural authority. The first argument relates to the inheritance of characters due to nutrition, and is as follows : The size of domestic animals is often of much practical import- ance, and has been attended to for many years with all the carefulness which a pecuniary stake ensures. It used to be said that " feed is more than breed," but it is now recognised that "heredity or 'breed' is the more important." There is also, of course, careful selection, " but no breeder claims that a breed is or can be kept up to extra size by selection alone." " Breeders do not believe that the characters acquired through the feeding of a single ancestor, or generation of ancestors, can oppose more than a slight resistance to that force of heredity which has been accumulated through many preceding generations, and is con- centrated from many lines of ancestry. Yet the belief is universal that the acquired characters due to food during the growing period have some force, and that this force is cumulative in suc- cessive generations. All the observed facts in the experience with herds and flocks point in this direction." The breeding of small and delicate Alderney cows was furthered by systematic underfeeding of the calves. Large-sized breeds have originated in regions of abundant food, and smaller breeds in districts of scantier forage. "This can hardly be due to accident." In short, " if these acquired characters are in no degree whatever transmitted, then certain practices of breeders, which are BREEDERS' EVIDENCE 217 founded upon the contrary belief, are delusive and expensive mistakes." We have given this argument at some length, since it deals with a subject of great practical importance, and since it is pre- sented to us with the double authority of Cope and Brewer. It is, however, on every count most disappointingly inconclusive. If the size is a function of four variables, — (a) the inheritable constitution of the stock (statistically determinable in certain of its expressions at the beginning of a period of observation) ; (b) the individual modifications produced by altered nutrition (approximately determinable by control experiments and ob- servations) ; (c) the possible occurrence of modification-inherit- ance ; and (d) the amount of discriminate selection withki a given period (also admitting of more or less precise statement), — then the only feasible way of reaching a conclusion as to the importance of any one factor — say the third in this case — is to eliminate the others one by one. As to the Alderney cows, it is admitted by all that the skilful breeder can breed small or breed large, either by relying wholly on the selection of a sufficiently variable stock, or by assisting selection by modification kept up for each generation ; but this does not touch the question at issue. And if it be a fact that large-sized races always come from regions of abundant nutrition, and vice versa, it is plainly consistent both with natural and artificial selection. As to the argument that unless modification-inheritance be a fact the practice of breeders is an expensive mistake, one is tempted to retort that the latter is at least as likely as the former ; but the sufficient answer is that breeders, even though they may think they do, never put their stake on the doubtful card. Finally, it may be noted, though this is a point rather for the biologist than for the breeder, that experiments on increased size of parts are more decisive than those which refer only to the size of the whole. 2i8 TRANSMISSION OF ACQUIRED CHARACTERS Manly Miles gives two cases to illustrate what seems to him a general fact, the occurrence of modification-inheritance in breeding : " The fashion of raising lambs by nurses of other breeds, and drying up the dam at once to keep her in show condition, resulted in seriously diminishing the inherited capacity for milk production in the females of the family as treated." " Cows on short pastures and under careless management will form the habit of ' going dry ' early in the season, and this habit of giving milk for a short period is not only transmitted, but becomes a marked peculiarity of the females of the family that is persisted in under better conditions of food supply." But these and numerous similar cases only show, what is univer- sally admitted, that a nutritive disturbance in the mother is apt to affect the nutritive vigour of the offspring. Brewer (cited by Cope, 1896, p. 436) reports what may be called a good case. Sheep taken from a favourable region to one with alkaline or salt soil, dry climate, and corresponding forage plants, acquire a certain harshness in the wool. The change begins immedi- ately, " but is more marked in the succeeding fleeces than in the first. It is also alleged that the harshness increases with succeeding generations, and that the flocks which have inhabited such regions for several generations produce naturally a harsher wool than did their ancestors, or do the new-comers." Of course, the second generation would naturally have harsher wool than the new-comers, but if harshness really increases with succeeding generations, the case is one of the best as yet brought forward. Immunity. — Another typical line of evidence is based on the study of immunity. To this very important, but very difficult, subject we have referred in another chapter, but the particular point here may be briefly stated. It is well known that some natives are relatively immune to yellow fever ; this is now a heritable quality ; the question is whether it can be regarded as originally an acquired character. Was it in origin a modification of the bodily metabolism subsequent upon the disease ? It seems very difficult to adopt this interpretation, and most authorities incline to the other alternative of regarding immunity as a constitutional variation which has become dominant in the race by the elimination of those members who were not immune. MEDICAL ARGUMENTS 219 It may be objected, however, that there are cases where a mother rabbit or guinea-pig has been artificially rendered immune to certain diseases, and has afterwards had young born immune. This may be due to a kind of infection before birth, some anti- toxin or other having probably passed from the mother to the unborn young. (Misunderstanding No. V.) Medical Arguments. — A medical argument which has convinced many is somewhat as follows. Its cogency rests on the difficulty of drawing hard-and-fast denning lines. It is alleged that a pregnant woman with smallpox may infect her unborn offspring — a clear case of intra-uterine contagion. A tubercular mother may have an offspring without tuberculosis, but with something wrong with its heart. Here a constitutional diathesis, stimulated by a bacillus, is followed by a result in the offspring quite different from the condition in the parent. Toxins produced by bacterial disease in the parent may affect the offspring without inducing any special disease, but by weakening its constitution and power of resistance. Toxins produced, apart from bacterial disease, by a saturation of the parent with alcohol, opium, and the like, may affect the offspring both functionally and structurally, with the result that there are diseases and malformations. It has been shown experimentally that toxins (hydrocyanic acid, nicotin, alcohol, etc.) may, directly injected into the eggs of fowls, affect the development so that malformation results. It is stated that the effects of lead-poisoning on the offspring may be wholly due to the father. Therefore it seems legitimate to infer that toxins produced in the body may have a direct effect upon the germinal material. It is not shown, however, that the effect on the offspring is the same as that induced in the parent — which is the biological point under discussion — and it is a wild hypothesis that an ordinary modification liberates anything comparable to a toxin. Alcoholism. — Habitual drunkenness in a parent or in the parents produces familiar modifications, and may be followed by dire results in the offspring. But before drawing the hasty conclusion that definite structural results of alcoholic poisoning 220 TRANSMISSION OF ACQUIRED CHARACTERS on the parent's body are in the strict sense transmitted to the offspring, we do well to consider — (i) that the intemperate habits of the parent may be the expression of an inherited psychopathic disposition, and it is this which is transmitted to the offspring ; (2) that the saturation of the body with alcohol may have a direct effect on the nutrition and developmental vigour of the germ-cells ; (3) that the children of drunkards often become accustomed to alcohol as part of their food, from the days of suckling onwards. ^/Nervous Diseases. — Prof. Binswanger of Jena, a famous student of psychiatry, has expressed his inability to find evidence that a mental or nervous disease acquired during the individual life is, as such, or in partial expression, inherited by the offspring. There are, he of course allows, numerous cases in which an inheritance of mental or nervous diseases can be traced from one generation to another ; but his difficulty was to find a case where it could be securely maintained that the first occurrence of the disease was due to external influence. It may, of course, be urged, though it seems an untenable extreme, that mental and nervous diseases never have an exo- genous origin, but are always referable to germinal defect. If so, it simply forces us to say that this line of argument is closed as far as the question of the transmissibility of modifications is concerned. Modifications of Habits and Instincts. — Many animals are very plastic in their habits, and some show some plasticity even in their instincts. It seems an interesting line of experiment to try to determine whether there is any evidence of transmission of peculiar individually modified habits. For an expert discus- sion of the subject we must refer to Principal Lloyd Morgan's Habit and Instinct. There are obviously many difficulties. The experimenter must be sure that the original change of habit is really modifica- l) not an inborn idiosyncrasy. He must be careful to MODIFIED HABITS AND INSTINCTS 221 eliminate the "possibility of the offspring learning by imitation or suggestion. He must also exclude the possibility of selection. He must remember that the offspring are probably as docile, as plastic, as adaptable as their parents, or perhaps more so. Moun- taineering mules come to have an extraordinary power of adapting themselves to peculiar exercises, but mule does not inherit from mule ! A hen becomes an adept in rearing ducklings : will her own children, put to a similar task, be less fussy than she was at first ? House-martins have learned to build beneath the eaves : has there teen any hereditary transmission of this acquired habit, or is it merely "the result of intelligent adaptation through the influence of tradition " ? Have grouse inherited the habit of flying so as to avoid telegraph wires ? Is it indubitably the case that the kittens of a cat " taught to beg for food like a terrier" may spontaneously exhibit the same peculiar habit? These are some of the cases which Lloyd Morgan discusses, and his conclusion is that the evidence for the transmission of acquired habit is insufficient. § 10. As regards Mutilations and the Like When we think of the bellicose activities of our ancestors, it seems almost absurd to discuss the question of the transmissi- bility of the results of mutilations, wounds, and other injuries. Moreover, it is well known that dishorning of cattle, docking of horses' tails, curtailing of sheep, cropping of dogs' ears, and similar practices, have been continued for many generations without any known hereditary effect. The circumcision of the children of Jews and Mohammedans has gone on for many cen- turies, but there is no demonstrable structural result. Yet the question is one of possibilities, and there is a huge literature of observations and experiments. Few Useful Results. — The net result, it must be confessed, is very disappointing, and the reasons for this are not far to seek. 222 TRANSMISSION OP ACQUIRED CHARACTERS (1) Many of the experiments and observations have failed to conform to the ordinary canons of scientific method. Many of them overlook the probability of coincidence, identify post hoc with propter hoc, mix up observation and inference, or base a con- clusion on a small number of instances. It may be noted that cases suggesting the transmission of the results of mutilation and injury are most abundant in the older, less critical literature. What may be called good cases have been very scarce of recent years, though many observers have been on the watch for them. (2) Some of the kinds of experiment — e.g. the amputation of large parts or of portions of internal organs, such as the spleen — are evidently of a kind which must be rare in nature. Therefore, though such " fool's experiments," as Darwin would have called them, may have some indirect value, they tend to be of little significance to the evolutionist. (3) The experimental repetition of those mutilations and in- juries which are common in nature is of little value, since nature's experiment shows with sufficient clearness that the results are not trans'mitted. If they were there would be but little now left of man and other combative organisms. As Hartog says, " The tendency to transmit the mutilation itself would be so ruinous as to rapidly extinguish any unhappy race in which it was largely developed " (Contemp. Rev., v. 64, p. 55). As a matter of fact, even in the individual lifetime the results of mutilation are very often repaired by regeneration, which in its specialised expression is probably the adaptive outcome of prolonged selection. (4) If the results of mutilation can be in any degree trans- mitted, they must affect the germ-cells in some specific way. The improbability of this is very great in the case of many mutilations, such as lopping off. a tail. The amputation has often little demonstrable effect beyond a slight irritation of the tissues at the cut surface ; the organism's reaction bears little relation to the actual effect produced ; a considerable part of the body has been lost, but there is no constitutional disturbance — the MUTILATIONS AND THE LIKE **$ reaction is a mere scar. Why should one expect the offspring to have a shorter tail because its parent has been curtailed ? Might one not as reasonably expect a longer tail ? No one has ever observed that the descendants of much-pruned fruit-trees or decorative shrubs are any the smaller in consequence. The length of the hair in offspring is not known to be affected by the frequent cropping, clipping, or shearing of their parents. In fact, the structural results of most mutilations are not modifi- cations in the usual sense. (5) But there are cases in which the removal of a part has deeply saturating effects. Thus the removal of a thyroid gland may have an influence on many parts of the body. In such cases, therefore, the possibility of the germ-cells being in- fluenced is more conceivable. But, unless the change in the offspring — supposing that there is some change — corresponds to the direct change wrought upon the parent, we have not to deal with modification-inheritance of the first degree, which is the only question under dispute. (6) Since the structural change due to a mutilation is not on the same plane as the ordinary modifications which occur in nature, we do not expect useful results from further mutilation experiments. We may refer, however, to the suggestion made by Dr. J. W. Ballantyne,* that in this connection, as with other modification experiments, investigators err by beginning at too late a stage, after the organism is firmly set. It may be that ex- periments on early stages would yield more positive results. It may be that the germ-cells in their early generations are more reachable by, or sensitive to, somatic influences. Illustrations. — In our brief discussion of this well-worn subject, we shall for convenience distinguish three categories : (A) amputa- tions, such as docking the tail ; (B) wounds, such as the rupture of the hymen ; (c) deformations, such as the compression of the * " Discussion on Heredity in Disease," Scottish Med. and Surg. Journal, vi. (1900), p. 312, 224 TRANSMISSION OF ACQUIRED CHARACTERS Chinese lady's foot. Under each category we shall notice merely a few typical cases, which may be added to as the reader pleases by referring to the literature cited, or by consulting the great work of Delage. Amputations repeated Generation after Generation. — Circum- cision among Jews and Mohammedans, docking horses, dogs, and sheep, cutting off parts of the ears of dogs, dishorning cattle, are cases in point, and there is no evidence of transmitted result. Dar- win (1879) does indeed cite Riedel to the effect that a shortened prepuce has been induced among the Mohammedans of Celebes, but Delage notes the inconclusiveness of Riedel's observations. Haeckel (1875) and Leidesdorff (Wien. med. Wochenschr. 1877) have also stated that a rudimentary prepuce occurs more frequently in races who practise circumcision, but other statistics do not bear this out. As Ziegler says (1886, p. 27), "There is in this respect no difference between Jews and Christians ; among the latter a defective develop- ment of the prepuce is as frequent as among the former." See also Roth, Correspondenz-Blatt /. Schweizer A erzte, 1884. Weismann cut off the tails of mice for nineteen generations, Bos for fifteen, Cope for eleven, Mantegazza and Rosenthal likewise, but in no case was any inherited result observed. An American record of the production of a tail-less race almost certainly illustrates an unscientific use of the imagination. The tails of fox-terriers are often cut, and pups with short tails are sometimes observed. The following case is representative of a number of records. A fox-terrier, whose tail had been cut, had four pups, one with a full-length tail, one with a rather short tail, and two with quite short tails. But the short tails had the usual tapering vertebrae (D. E. Hutchins, Nature, Ixx., 1904, p. 6). Delage cite^s Tietz ( 1 889) to the effect that kittens with an atrophied tail are frequent in the Eiffel, where the peasants habitually curtail their cats — in mistaken kindness, for they believe that there is a worm at the root of the tail which keeps them from catching mice ! If abortive tails are unusually common in that district, the fact is of much interest, and Delage does not find sufficient explanation in the suggestion of Dingfelder (1887), that, as the peasants leave short- tailed kittens alone, an inborn variation towards short tails has been allowed to diffuse itself. It is, of course, easy to appeal to an innate tendency to shortening of the tail, but it is curious that the examples should be found so generally among domesticated animals, MUTILATIONS AND THE LIKE 225. like cats and dogs, sheep and horses, which are so often artificially docked. Amputations not repeated throughout Generations. — These form what we may call the " curtailed cat " type, the point being that a she-cat whose tail has been cut off accidentally or otherwise has been known to bear kittens, some or all of which have tails shorter than the normal. The cogency of such cases is annulled when we remem- ber,— ( i ) the existence of a Manx and Japanese breed of tail-less cats ; (2) the occasional occurrence of tail-less or short-tailed kittens as " sports " in the litters of quite normal parents ; and (3) the frequently observed variability of the tail region in many mammals. In all such cases at least two inquiries are imperative: (i) some estimate of the probability of coincidence, since the post hoc may be no propter hoc, but merely a variation which happens to resemble more or less the result of the mutilation ; and (2) an investigation into the pedigree of both parents, since there may be in either or in both an innate tendency towards a shortening of the tail. These inquiries are not usually made. A number of very interesting cases are given by Delage (1903), and it is difficult to dispose of them except by calling them " mere coincidences." One of my colleagues has told me of a case of a child with a peculiar bare patch among the hair, corresponding to a similar area on the mother's head, where the bareness was due to ringworm. The child's patch was bare save for a narrow streak of short hair, stretching about half way across. The patch was a little in front of the mother's, but was similarly situated above the left ear. What can one say but " coincidence " ? Or may one suggest that the ringworm found out a hereditarily weak spot ? Wounds repeated Generation after Generation. — We do not aim at any surgical precision in distinguishing amputations from wounds. Our point is simply that there is a difference between the effect of an amputation which may be almost negative, and the effect of a wound which disturbs the relations of parts. The classification is borne out by the fact that whereas there is not a grain of evidence, so far as we know, to lead one to believe in the inheritance of the results or any results of amputations, except when very important organs are operated upon, the same cannot, at first at least, be said in regard to the effects of wounds. The typical case here is the rupture of the hymen in the first sexual intercourse — a trivial lesion, perhaps, but one which haa 226 TRANSMISSION OF ACQUIRED CHARACTERS occurred in every generation, and one of which no inherited results are known. Nor are there known results of a kind of circumcision practised by Somalis and others on girls as well as boys. In some races ear-boring, nose-boring, and the like, have been practised by both sexes for many generations ; and no inherited result has ever been observed. Casual Wounds. — Darwin cites the case of a man whose thumbs were badly injured in boyhood, as the result of frost-bite. His oldest daughter (S) had thumbs and thumb-nails like the father's ; his third child was similar as to one thumb ; two other children were normal. Of the four children of S, the first and the third, both daughters, had deformed thumbs on both hands. The cogency of this case depends on whether there was or was not any previous family tendency to thumb-deformity. It may have been that the frost-bite was really an unimportant incident. Darwin gives another case of a man who, fifteen years before marriage, lost his left eye by suppuration. His two sons had left-sided microphthal- mia. Here we have probably to deal with an innate eye-defect in the father. Bouchut * reports the case of a man of twenty-five who injured his hands and feet by a fall from a scaffold. Of five children only one was normal. His son had one finger on each hand and two toes on each foot. A daughter (M) had two toes on each foot, one finger on the right hand, and two on the left. She married a normal man, and of her four children the oldest was normal, the others like herself. Cases like the last may seem puzzling to those unaccustomed to deal critically with the facts of inheritance. But in reality they are in most cases merely illustrations of the familiar fallacy of con- fusing post hoc and propter hoc, of mixing observation and inference (Ziegler, 1886, p. 26). Bouchut does not say that the children showed the same deformity as their father acquired ; he does not tell us about the ancestry of the father and mother, an indispensable fact if a case is to be considered seriously, since inborn mal- formations are common in some families ; finally, the frequency of inborn malformations of the fingers and toes must be borne in mind, and the possibility of coincidence allowed. Ziegler (1886, pp. 29, 30) discusses a number of cases where defects * Nouveaux Elements de Pathologie generate, Paris, 1882. Cited by Ziegler, 1886, pp. 3, 4. MUTILATIONS AND Ttf£ LIKE 2*7 in the eye occurred in the offspring of animals whose eyes had been operated on, injured, or infected. But experiments in which the eyes are infected with tubercle or the like are not relevant until all possibility of the offspring being infected is excluded ; and as for cases such as those given by Brown-Sequard (1880), where the extirpation of the eye-bulb in the parent was followed in the offspring by the loss of one eye or of both, or by corneal obscuration, it is necessary to compare the results with the statistics as to the frequency of various kinds of innate eye-defects. Deformations. — We do not know all that we should like to know in regard to the artificially deformed feet of Chinese ladies, but there is no evidence that the long-continued deformation has resulted in any hereditary change. For untold ages the herdsmen in some parts of the Nile valley have artificially deformed the horns of their cattle, making them bend forwards, twist spirally, and so forth; but no effect on offspring has ever been observed (R. Hartmann, Die Haussaugethiere der Wildl'dnder. Ann. Landwirthsch. ; Berlin, 1864, P- 28). The Rook's Bill-feathers. — Settegast and others have referred to the bristle-like feathers about the nostrils and the base of the bill in the young rook. They are said to disappear mechanically when the bird begins to bore with its beak in the ground, yet they are always present in the nestling. To cite this as an example of the non-transmission of a deformation-effect is probably quite erroneous, for there is no proof that the disappearance is causally connected with burrowing. It is probably a constitutional pecu- liarity that these feathers should be moulted and not replaced. They disappear even if the rook is not allowed to bore (see Oudemans and Haacke, cited by Delage, 1903, p. 223). On the other hand, to start from the fact that the bristles disappear even if there is no boring in the ground, and to cite this as an instance of the transmission of a deformation-effect, is equally fallacious. There is no evidence that it was a deformation-effect to start with. Some Puzzling Cases. — While the argument based on the apparent transmission of the results of mutilation appears to 'us very weak, it must be admitted that there are some cases which, if accurately recorded, are puzzling. It is desirable that any fresh cases, similar in nature to those which we propose to illustrate, should be studied carefully and without prejudice. Though they may not prove modification-inheritance, they may lead to interesting results. S2& TRANSMISSION OP ACQUIRED CHARACTERS Prof. Haeckel * records that a bull on a farm near Jena had its tail squeezed off at the root by the accidental slamming of the byre-door, and that it had thereafter a tail-less progeny. This is very interesting, but we are bound to ask — (i) how often tail- less cattle arise apart from curtailing by the byre-door ; (2) whether the bull had any tail-less offspring before it was cur- tailed ; (3) how many tail-less offspring it actually had, and so on. It may be that the answers to these questions would be quite satis- factory, but, to make the case cogent, the questions should have been forestalled. In 1874 Herr W. Besler, in Emmerich on the Rhine, wrote to Prof. L. Biichner (1882, p. 24) to report the following case. At Dobeln, in Saxony, at Eichler's Hotel there, he saw a young dog apparently bereft of ears and tail. When he remarked that the beast had been far too much cut, he was told that this was not the case, for it and its brother had been born so, out of a litter of four. The mother was normal, the father was an " Affen-Pinscher," whose ears and tail had been cut. The same condition had occurred once in a pre- vious litter. Supposing that this was more than an ostler's yarn, we should have to inquire into the ancestry of the father and mother to see whether inborn shortness of ears and tail had ever manifested itself in the family. Prof. Biichner also relates that in the autumn of 1873 a build- ing-contractor, K , in Westphalia, bought a duck whose right " wing-bone " had been broken and had mended in a crooked fashion. Next spring the duck had four ducklings, two of which showed on the right wing, and two on both wings, an extra feathered wing (4-5 in. in length), protruding immovably at an angle of 45° above the otherwise normal wing. But this duplicity, if such it was bore no precise relation to the original injury, and probably was quite unconnected with it. Biichner gives a number of other instances. Thus Williamson saw dogs in Carolina which had been tail-less for three or four genera- tions, one of the ancestors having lost the tail by accident. f But tail-lessness is also known as a germinal variation. Bronn J describes the case of a cow which lost one horn by ulcera- tion ; it had afterwards three calves which showed on the same side * Schopfungs-Geschichte, ed. 1870, p. 102. f Waitz, Anthropologie der Naturvolker, i. p. 93. % Geschichte der Natur, 1871, p. 96. MUTILATIONS AND THE LIKE 229 of the head no true horn, but a small nucleus of bone hanging to the skin. It may have been that an inborn weakness, which led to the ulceration of the mother-cow's horn, took a slightly different expression in the calves. Dr. J. W. Ballantyne quotes Kohlwey's experiments on pigeons : " He cut off the posterior (first) digit of the foot, and the mutilated bird got into the habit of turning the fourth digit backwards and using it in perching ; he got no descendant of these mutilated birds without a posterior digit, but he got a descendant of one of the pairs with its fourth digit turned backwards like the first. The mutilation was not transmitted, but the physiological adaptation tomeet itwas." Is it sufficient to regard this simply as a coincident variation ? Some of the best cases are those in which a morbid change was associated with the loss or injury of a particular structure. A cow loses its left horn by suppurative inflammation ; it has subsequently three calves in which the left horns were abortive (Thaer, 1812). But it may be that the original loss was due to a weakness of germinal origin. Prof. W. H. Brewer (1892-3) is responsible for launching a large number of rather unseaworthy instances of modification-inheritance. Inter alia, he tells the story of a pure-bred game-cock who lost an eye in a fight, and transmitted his loss. While the wound was very malignant, he was turned into a flock of game-hens of another strain, and " a very large proportion of his progeny had the corre- sponding eye defective." " The chicks were not blind when hatched, but became so before attaining their full growth. The hens after- wards produced normal chickens with another cock." A trustworthy correspondent writes : " My great-grandmother had one toe broken at a dance ; all her descendants are born with one toe bent double — my grandmother, mother, aunt, sister, and myself." But to this almost typical story what can be said except that congenital variations of the toes are common, and that the accident at the dance had nothing to do with the story ? Of great interest is the statement made by some botanists that some peculiar effects on trees due to mites, ants, etc., are trans- mitted. Thus Lundstrom says that the little shelters (acaro- domatia) produced on the leaves of lime-trees, etc., by mites, may appear when there are no mites. But, admitting that there are some puzzling cases, we cannot avoid the general conclusion that as regards mutilations, amputa- tions, wounds, and deformations, the case for the affirmative is not strengthened by further inquiry. 230 TRANSMISSION OF ACQUIRED CHARACTERS § ii. Brown-Seq uard's Experiments on Guinea-Pigs In recent discussions of modification-inheritance much pro- minence has been given to the experiments made by Brown- S6quard, Westphal, and others on the apparent transmission of artificially induced epilepsy in guinea-pigs. The reason for this prominence is that the case is not without cogency, and that a record of precise experiments (although of a somewhat ugly character) comes as a relief amid anecdotal evidence. Prof. E. Ray Lankester goes the length of saying (1890, p. 375), " The one fact which the Lamarckians can produce in their favour is the account of experiments by Brown-Sequard, in which he pro- duced epilepsy in guinea-pigs by section of the large nerves or spinal cord, and in the course of which he was led to believe that in a few rare instances the artificially produced epilepsy was transmitted." As the case has been often discussed — e.g. by Romanes (1895, vol. ii. chap, iv.) — we shall treat of it briefly. What the Experiments were. — Through a long series of years (1869-91), Dr. Brown-Sequard, a skilful and ingenious, if somewhat impetuous, physiologist, experimented on many thousands of guinea-pigs. He made a partial section of the spinal cord in the dorsal region, or cut the great sciatic nerve of the leg ; he observed that the injury was followed after some weeks by a peculiar morbid state of the nervous system, cor- responding in some of its features to epilepsy in man ; he allowed these morbid animals to breed, and found that the offspring were frequently decrepit, and that a certain number had a tendency to the so-called epilepsy. Results of the Experiments. — If it be understood that we have omitted or altered a few difficult technicalities, we may call the following statement Brown-Sequard's summary of his results. The inverted commas are ours : (i) "Epileptic" symptoms appeared in the offspring of parents who had been rendered " epileptic " by an injury to the spinal cord. BROWN-SEQUAR&S EXPERIMENTS 231 (2) " Epileptic " symptoms appeared in the offspring of parents who had been rendered " epileptic " by section of the sciatic nerve. (3) An abnormal change in the shape of the ear was observed in the offspring of parents in which a similar change followed a division of the cervical sympathetic nerve. (4) Partial closure of the eyelids was observed in the offspring of parents in which that state of the eyelids had resulted either from section of the cervical sympathetic nerve, or the removal of the superior cervical ganglion. (5) An injury to the restiform body (associated with the medulla oblongata) was followed by a protrusion of the eye (exophthalmia), and this reappeared in the offspring some- times through four generations, even affecting both sides, though the lesion in the parent had only been on one of the corpora restiformia. (6) An injury to the restiform body near the nib of the calamus was followed by haematoma and dry gangrene of the ears, and the same conditions reappeared in the offspring. (7) After a section of the sciatic nerve, or of the sciatic and crural, some of the guinea-pigs gnawed off two or three of the toes, which had become anaesthetic ; in the offspring two or three toes were absent. Sometimes, instead of complete absence of the toes, only a part of one or two or three was missing in the young, although in the parent there was a loss not only of the toes, but of the whole foot (partly eaten off, partly destroyed by inflammation, ulceration, or gangrene). (8) As effects of an injury to the sciatic nerve, there followed various morbid states of the skin and hair of the neck and the face, and similar alterations in the same parts were observed in the offspring. When the sciatic nerve had been cut in the parent, the descend- ants sometimes showed a morbid state of the nerve. There was also a similarity in the successive appearance of the phenomena, described by Brown-Sequard as characteristic of the periods of development and of abatement of the " epilepsy," especially in tfre appearance of the epileptogenic area and the disappearance of hair around that area whenever the disease showed itself. 232 TRANSMISSION OF ACQUIRED CHARACTERS Muscular atrophy of the thigh and leg followed section of the sciatic nerve, and this was also observed in the offspring. After cutting the restiform body one eye suffered deterioration ; this was seen in the offspring in one eye, or even in both. In general, the morbid conditions may affect both sides in the parents and only one in the offspring, or vice versd, or the side affected may be different. One generation may be skipped, but the duration of transmission was in some cases traced through five or even six generations. The females seemed better able to transmit morbid states than the males. As to the frequency of transmission, some inherited result was ob- served in more than two-thirds of the cases. Brown-Sequard's results were partly confirmed by his assistants, Westphal (1871) and Dupuy (1890), by Obersteiner (1875), and by Romanes (1895). Dr. Leonard Hill divided the left cervical sympa- thetic nerve in a male and a female guinea-pig, and thereby produced a droop of the left upper eyelid. Two offspring of this pair ex- hibited a well-marked droop of the upper eyelid. " This result is a corroboration of the series of Brown-Sequard's experiments on the inheritance of acquired characters." Facts to be noted, which dispose of a Number of Criticisms. — It is stated that the so-called " epileptic " state may also be induced in the dog by injury to the cerebral cortex, and may, in this case also, reappear in the offspring. If this be so it shows that we have not to deal with a tendency peculiar to guinea-pigs. It is stated that the " epileptic " condition does not occur spon- taneously— i.e. apart from injury to the nervous system — in guinea- pigs. Therefore the interpretation of the apparent inheritance as being due to a fresh variation which happened by coincidence to resemble the parental state, is inadmissible. As the tendency to " epileptic " fits (which do not last long) was seen only in the offspring of animals which had been operated on, and was manifested only after appropriate stimulus, especially after irritating an " epileptogenic " zone behind the ear on the same side as the original injury, we must pass by Gal ton's suggestion (1875) of the possibility of reappearance through imitation. Even if it be allowed that there is a certain infectiousness in " fits," this would not apply to the loss of toes, the diseased state of the ear, the pro- truding eyes, and so on. BROWN-S&QUARUS EXPERIMENTS 233 It is stated that the morbid condition of the parents was also in- duced by bruising the sciatic nerve without cutting the skin, or by striking the animals on the head with a hammer. If this be so it seems to show that the result may occur without any associated microbe influence, and possible infection of the offspring thereby (Weismann's criticism, in part). The hypothesis of microbes does not seem to be supported by any definite facts, but we note that it is not entirely excluded by Ziegler in his review of possible ex- planations (1886, p. 29). Brown-Sequard experimented with both males and females, and although he got more striking results with the latter, he did not fail with the former. This seems to lessen the force of the criticism that the offspring were affected during gestation, and therefore not, in the strict sense, hereditarily. Criticisms. — (i) The original modification was cutting, bruising, or destroying part of the nervous system ; the subsequent result was the " epileptic " state, and the various other diseased conditions mentioned. It need hardly be said that the mutilation or injury inflicted on the parent was never reproduced in the offspring, though the subsequent results sometimes were. (2) The conditions exhibited by the offspring were very diverse — general feebleness, motor paralysis of the limbs, trophic paralysis resulting in loss of toes, cornea, etc., other nervous and sensory dis- orders, and in some cases the particular " epileptic " state. In a number of cases the condition of the offspring was so different from that of the parent, that the only common feature was that in both cases there were abnormal neuroses. Romanes, while regarding his results as corroborations of those of Brown-Sequard, admitted that the epileptic condition was only rarely transmitted. (3) Even numerically there was no small diversity in the results. Thus in one set of experiments (Obersteiner, 1875), out of thirty-two young ones born of " epileptic " parents, only two showed symptoms of " epilepsy " and paralysis, three were paralytic, and eleven were only weak. Romanes did not find that any of the offspring of parents who had eaten their toes off showed, even in six generations, any defect in these parts. Even Brown-Sequard only observed this peculiar " transmission " in about i or 2 per cent, of cases. (4) Prof. Ziegler 's criticism is partly based on the allegation that guinea-pigs (as we keep them in captivity) are pathological and nervous animals, very readily thrown into an epileptic state. On 234 TRANSMISSION OF ACQUIRED CHARACTERS making a slight cut in the skin, on the occasion of a small operation on the neck, Ziegler sent an apparently healthy guinea-pig into a severe epileptic fit. But there seems considerable difference of opinion as to this nervousness of captive guinea-pigs. (5) It seems to us that the original modification was too violent to afford satisfactory data in connection with the present discussion. No matter how neatly the operations were effected, the partial sec- tion of the spinal cord, the cutting of the sciatic or of the cervical sympathetic nerve, the removal of the superior cervical ganglion, the injuring of the restiform body, imply very serious injuries, and it is hard to believe that others were not implied in some of the ex- periments— e.g. on the restiform body. But if a modification is violent it may disturb the whole organism, nutritive * and repro- ductive f functions alike, and it may naturally lead to abnormality in the offspring. Especially may it lead to general decrepitude, which, it seems to us, was the most frequent result. At the same time this hardly touches the most distinctive feature of the ex- periments, that sometimes there appeared in the offspring morbid conditions precisely similar to the results of the injury inflicted on the parents. It may be, however, that only particular parts of the body are susceptible to the influence of the original disturbance. Prof. T. H. Morgan (1903, p. 257) directs attention to the experi- ments of Charrin, Delamare, and Moussu, which have an interesting bearing on some of Brown-Sequard's results. After the operation of laparotomy on a pregnant rabbit or guinea-pig, the kidney or the liver became diseased, and the offspring showed similar affections. The experimenters suggested that some substance set free from the diseased kidney of the mother affected the kidney of the young in the uterus. " May not, therefore, Brown-Sequard's results be al explained as due to direct transmission from the organs of the paren to the similar organs of the young in the uterus ? " But thiswoul not be inheritance in the strict sense. It should be noted, however, that what has been just said does not of course apply to those cases in which Brown-Sequard experimented on the male parent. Charrin maintains on experimental grounds that " cytotoxins " may pass not only from the mother to the foetus, but from either parent to its germ-cells — ova or spermatozoa (see Revue generate des Sciences, * Dupuy, while confirming Brown-Sequard, laid emphasis on the alters tions of nutrition after the experiments. f Sommer notes a diminution of fertility after the experiments;. £ROWN-S£QUARjyS EXPERIMENTS 235 Jan. 15, 1896). Moreover, Voisin and Peron have found evidence that in epilepsy a toxin is produced which causes convulsions when injected into animals (see Archives de Neurologic, xxiv., 1892, and xxv., 1893, and Voisin's L'Epilepsie, Paris, 1897, pp. 125- 133). It is thus not a mere speculation to suppose that a toxin was produced in the guinea-pig epilepsy, and that this affected the germ-cells of both sexes. This suggestion is made by Prof. Bergson in his remarkable book L* Evolution Creatrice (1907), and he addsi to the suggestion the query, May not something of the same sort I be true in those cases where acquired peculiarities are transmitted ?* Prof. T. H. Morgan (1903, p. 255) also notes an interesting fact. " While carrying out some experiments in telegony with mice, I found in one litter of mice that when the young came out of the nest they were tail-less. The same thing happened again when the second litter was produced, but this time I made my observations sooner, and examined the young mice immediately after birth. I found that the mother had bitten off, and presumably eaten, the tails of her offspring at the time of birth. Had I been carrying on a series of experiments to see if, when the tails of the parents were cut off, the young inherited the defect, I might have been led into the error of supposing that I had found such a case in these mice. If this idiosyncrasy of the mother had reappeared in any of her descendants, the tails might have disappeared in succeeding genera- tions. This perversion of the maternal instincts is not difficult to understand, when we recall that the female mouse bites off the navel-string of each of her young as they are born, and at the same time eats the after-birth. Her instinct was carried further in this case, and the projecting tail was also removed, "Is it not possible that something of this sort took place in Brown-Sequard's experiment ? The fact that the adults had eaten off their own feet might be brought forward to indicate the possi- bility of a perverted instinct in this case also." On the other hand, this interpretation cannot apply to some other results which Brown- Sequard observed. Sommer's Experiments far from corroborating Brown-Sequard's. — In experiments the results of which were published in 1900, Max Sommer repeated some of those which Brown-Sequard and others had made, but without corroborating them. The so-called " epilepsy " was induced by cutting the sciatic nerve on one side or on both sides ; the tendency to " fits " occurred some 236 TRANSMISSION OF ACQUIRED CHARACTERS days or some weeks after the operation ; they were brought on by rub- bing particular areas of the body (the epileptogenic zones) ; whether they ever occurred spontaneously remained doubtful, since any friction on the appropriate spots — e.g. when the animal scratched itself — served to bring them on. After some months the tendency! to the attacks disappeared, and irritation of the appropriate areas] was followed by only a slight fit or by none. (This is a noteworthy] fact.) The fertility of the " epileptic " guinea-pigs was lessened. Twenty-three young ones were reared (a small number compared with those in Brown-Sequard's experiments) — six from two pairs in which the father was " epileptic," six from four pairs in which the mother was "epileptic," and seven from five pairs in which both parents were " epileptic." In no case did " epilepsy " appear in the offspring. Even paralysis of one or more of the extremities was not demonstrated, though most carefully looked for. In the parents there were several defects in the toes or ulcerations\