THE JOHN -: CRAIG LIBRARY COLLEGE OF AGRICULTURE Cornell University Library | QH 366.C78 a organic mann Cornell University Library The original of this book is in the Cornell University Library. There are no known copyright restrictions in the United States on the use of the text. http://www.archive.org/details/cu31924003061979 THE PRIMARY FACTORS OF ORGANIC EVOLUTION THE PRIMARY FACTORS Qa“ ORGANIC EVOLUTION BY E. D. COPE, PH. D. MEMBER OF THE U. S., NATIONAL ACADEMY OF SCIENCES; PROFESSOR OF ZOOLOGY AND COMPARATIVE ANATOMY IN THE UNIVERSITY OF PENNSYLVANIA CHICAGO THE OPEN COURT PUBLISHING COMPANY (LONDON: 17 Jounson’s Court, FLEET St., E. C.) 1904 (® JH4 07 Tue Open Court PUBLISHING Co, CHICAGO, ILL., 1896. The Lakeside Press KK. K. DONNELLEY & SONS CO., CHICAGO. PREFACE. HE present book is an attempt to select from the mass of facts accumulated by biologists, those which, in the author's opinion, throw a clear light on the problem of organic evolution, and especially that of the animal kingdom. As the actual lines of descent can be finally demonstrated chiefly from paleontologic re- search, I have drawn a large part of my evidence from this source. Of course, the restriction imposed by limited space has compelled the omission of a great many facts which have an important bear- ing on the problem. I have preferred the paleontologic evidence for another reason. Darwin and the writers of his immediate school have drawn most of their evidence from facts which are embraced in the science of cecology. Weismann and writers of his type draw most of their evidence from the facts of embryology. The mass of facts recently brought to light in the field of paleon- tology, especially in the United States, remained to be presented, and the evidence they contain interwoven with that derived from the sources mentioned. Many of the zodlogists of this country, in common with many of those of other nations, have found reason for believing that the factors of evolution which were first clearly formulated by La- marck, are really such. This view is taken in the following pages, and the book may be regarded as containing a plea on their behalf. In other words, the argument is constructive and not destructive. The attempt is made to show what we know, rather than what we do not know. This is proper at this time, since, in my opinion, a certain amount of evidence has accumulated to demonstrate the doctrine here defended, and which I have defended as a working hypothesis for twenty-five years. In the following pages I have cited many authors who have contributed to the result, but it has been impossible to cite all who vi PRIMARY FACTORS OF ORGANIC EVOLUTION. have written on one part or another of the subject. If some very meritorious. essays have not been cited, it has been generally be- cause I have confined myself to those in which facts or doctrines were first presented, and have not had so much occasion to refer to ‘ those of later date. Mr. Romanes, in his posthumous book, Volume II. of his Dar- win, and After Darwin on Post-Darwinian Questions, expresses the following opinion of the position which has been taken by the Neo-Lamarckians of this country. He says that they do not dis- tinguish between the ‘‘statement of facts in terms of a proposition, and an explanation of them in terms of causality.” Had Mr. Ro- manes been acquainted with the literature of the subject published in America and elsewhere during the last three years, he would have had reason to change this view of the case. I think he would have found in it demonstration ‘‘in terms of causality."’ At the outset it must be stated that a knowledge of the history of organic evolution rests primarily on the science of morphology, and secondarily on the kinematics of the growth of organic struc- tures. The phenomenon to be explained is the genealogical suc- cession, or phylogeny of organisms ; and the access to this subject is through the sciences of paleontology and embryology. The phe- nomena of the functioning of the organism, or physiology, are only incidentally referred to, as not the real object of inquiry. Since organic species are much more numerous than the tissues of which they are composed, organogenesis must claim attention more largely than histogenesis. It is true that histogenesis is fundamental, but it is a science as yet in its early infancy, and little space can be given to it. The exact ow of organic evolution will never be finally solved, however, until our knowledge of histogenesis is com- plete. ; The research depicted in the following pages has proceeded on the assumption that every variation in the characteristics of organic beings, however slight, has a direct efficient cause. This assump- tion is sustained by all rational and philosophical considerations. Any theory of evolution which omits the explanation of the causes of variations is faulty at the basis. Hence the theory of selection cannot answer the question which we seek to solve, although it embraces an important factor in the production of the general re- sult of evolution. In the search for the factors of evolution, we must have first a knowledge of the course of evolution. This can only be obtained PREFACE, vii in a final and positive form by investigation of the succession of life. The record of this succession is contained in the sedimentary deposits of the earth's crust, and is necessarily imperfect. Advance in knowledge in this direction has, however, been very great of re- cent years, so that some parts of the genealogical tree are tolerably or quite complete. We hope reasonably for continued progress in this direction, and if the future is to be judged of by the past, the number of gaps in our knowledge will be greatly lessened. In the absence of the paleontologic record, we necessarily rely on the em- bryologic, which contains a recapitulation of it. The imperfections of the embryonic record are, however, great, and this record differs from the paleontologic in that no future discovery in embryology can correct its irregularities. On the contrary every paleontologic discovery is an addition to positive genealogy. If the present work has any merit, it is derived from the fact that the basis of the argu- ment is the paleontologic record. E. D. Cope. PuHILapELpuHIA, November 1, 1895. ~ CONTENTS. PAGE PREFACE 5 RU F v TABLE OF Gesu 5 Bes ‘ 2 ix LisT oF ILLUSTRATIONS ‘ : xiii INTRODUCTION Fi $ I PART I. THE NATURE OF VARIATION. PRELIMINARY ea F es 19 CHAPTER I. VARIATION. pees ‘ ‘ q 21 . Variations of Specria’ Characters. . F #3 25 a. Variations in Cicindela . - 25 6, Variations in Osceola doliata ‘ 29 ¢. Color-Variations in Cnemidophorus . 41 d, Variations in North American Birds and Mam- mals in Relation to Locality. . ‘ i 45 2. Variation of Structural Characters . . r 58 3. Successional Relation . .. . z 62 CuaAPTER II. PHYLOGENY. 1. General Phylogeny PhS Se BCS 74 2. Phylogeny of the Vertebrata . . . : 83 «, Phylogeny of the Classes ee ee ee ee 83 6. The Line of the Pisces . 3 ; 99 c. The Line of the Batrachia ‘ 108 d. The Line of the Reptilia. ; Hows oe ETS e. The Line of the Aves: . 123 f. The Line of the Mammalia. ; + « $26 g. Review of the Phylogeny of the Mammalia 138 x PRIMARY FACTORS OF ORGANIC EVOLUTION. PAGE hk, Phylogeny of the Horse . fa a 2 146 z. The Phylogeny of Man & eS 150 3. The Law of the Unspecialized . . . 172 CHAPTER III. PARALLELISM. Preliminary . : 175 1. Parallelism in the Bisehionaa 3 176 2. Parallelism in the Cephalopoda 182 3. Parallelism in the Vertebrata. . 192 4. Inexact Parallelism or Czenogeny. 3 200 5. Objections to the Doctrine of Parallelism . 205 CHAPTERIV. CATAGENESIS. . . . ; 211 PART II. THE CAUSES OF VARIATION. PRELIMINARY. . fe 2 Os 225 CHAPTER V. PHYSIOGENESIS. Preliminary . z ‘ ‘ 227 a, Relation of Size of Mollusca to Environment. . 229 6. The Conversion of Artemia into Branchinecta 229 c. Production of Colors in Lepidopterous Pupz . 230 @. Effect of Light on the Colors of Flatfishes . 238 e. Effect of Feeding on Color in Birds . 238 f. Blindness in Cave Animals. . . . . 241 CHAPTER VI. KINETOGENESIS. ae tags 3. yi Sahay ola 246 . Kinetogenesis of Miusevlas Sirnetane ge ok : 249 z. Kinetogenesis in Mollusca. . . 255 a, Origin of the Plaits in the Cidlatmmatts of the Gas- teropoda 255 . Mechanical Origin of Ghantens | in the Lamelli- branchiata . 261 c. Mechanical Origin of fie. impressed Fite in Cephalopoda. . . ‘ 268 3. Kinetogenesis in Vermes and Artheopoda ga Sy . 268 4. Kinetogenesis in Vertebrata . . . a a te 275. i. Kinetogenesis of Osseous Tissue . a TS «, Abnormal Articulations . . . + 275 CONTENTS. xi ! PAGE 6. Normal Articulations. . . . = 6 3 « 283 . The Physiology of Bone Mantding toe ee 285 ii. “Moulding of the Articulations. . . . . . . . 287 a, The Limb Articulations . Ms oe, . . 287 6. The Forms of Vertebral Centra . 6 a & « 362 iii. Increase of Size Through Use. . . . + 304 a, The Proportions of the Limbs and their Seamenle 305 6. The Number of the Digits . . . . . . 309 ce. The Horns . . . « = 314 iv. Mechanical Origin ot Dental Types, : . . 318 Preliminary . . ae ocae, et cai A Ca SLe a. The Origin of Canines Teeth Bh far oat be +. 327 6. The Development of the Incisors. . . . . . 328 c. The Development of Molars fot. Hot - 331 d, Origin of the Carnivorous Dentition el ge 932 e. Origin of the Dental Type of the Glires. . . . 345 v. Disusein Mammalia. . . . .... , 352 a, Natatory Limbs. . . Hi gs, “gp Ak 352 6, Abortion of Phalanges in Uawalats ie gear + 353 c, Atrophy of Ulnaand Fibula . .. . » . 355 d, Atrophy of Incisor Teeth . . . . . . . . 356 vi. Homoplassy in Mammalia. . . . . . . . . 357 vii. Origin of the Divisions of Vertebrata . . . . . 362 5. Objections to Kinetogenesis . . . . . . . « «375 CuapTER VII, NatTuRAL SELECTION. . . . 1. «© « © 385 PART III. THE INHERITANCE OF VARIATION. PRELIMINARY. . «©. . s ee ee ee 2 & ee BOT CuHapTerR VIII. HeEreEpitTy. 1. The Question Stated . 7 ne eee . . 398 2. Evidence from Embryology . . Sica + + 401 a, Vertebrata . . . . . eos . » . 40I 6. Arthropoda. . . ee ee ke GOR 3. Evidence from Paleontology . . 2 & = #405 «a, The Impressed Zone of the Nauiiloids é » 405 4. Evidence from Breeding . . . i, Mili Te wees nd Se «, Of Characters Due to Nutrition pe . 6 423 4. Of Characters Due to Exercise of Function . . 426 c. Of Characters Due to Disease. . . . . . 430 xii PRIMARY FACTORS OF ORGANIC EVOLUTION. PAGE d. Of Characters Due to Mutilation and Injuries . 431 . Of Characters Due to Regional Influences . 435 5. The Conditions of Inheritance ; . 438 6. Objections to the Doctrine of Tatlevitaned of Mcieleea Characters; « « « 4 © «& © ® * @ «» & « «© 458 CuHaPTER IX. Tue ENERGY oF EvoLuTIon. Prelimifiary,, s. ae. we Ok ee a a ZS 1. Atiageniesis’ 6.2 4 i) Seow Sa A HR cae 475 2. Bathmogenesis. we ap Pdin Ts) odds) 28g) g - 484 CHAPTER X. THE FUNCTION OF CONSCIOUSNESS. 1. Consciousness and Automatism fu z A 495 2. The Effects of Consciousness . , 509 CHAPTER XI. THE Opinions oF NEO-LAMARCKIANS . 2 518 Fig. LIST OF ILLUSTRATIONS. - Horn on Cicindela . . Osceola doliata triangula . . Osceola doliata clerica . . Osceola doliata collaris. . Osceola doliata temporalis. . Osceola doliata doliata . . Osceola doliata syspila . . Osceola doliata parallela . . Osceola doliata annulata . . Osceola doliata coccinea . Cnemidophorus tessellatus . . Cnemidophorus gularis . . Lacerta muralis ‘ . Shoulder-girdle of Pigilomidiica cota: : . Do. of Rana temporaria, tadpole with budding itmbe: . Do., adult . © eos . Bufonide . Scaphiopidz and Pelobatidze . Hylide. nah . Cystignathidze . Ranide. . Feet of Uma sbpparts Cape, atl Plonnpass garrulus Smith . Eusthenopterum ‘foowds Whiteaves . Paired fins of Cladoselache Dean . Extremities of skeletons of caudal fins of fates . Cricotus crassidiscus Cope, head and belly . Cricotus crassidiscus Cope, vertebral column and pelvis 112 PAGE 27 32 32 34 34 36 38 38 40 40 42 43 44 64 64 64 66 66 67 67 67 72 go 92 96 . Iii xiv PRIMARY FACTORS OF ORGANIC EVOLUTION, PAGE Fig. 29. Crania of Stegocephalia and Cotylosauria . . . . 117 30. Diagrams of crania of Reptilia. . . . .. . 118 31. Do. . ee 119 32. Archaeopteryx Behograpibien Waipu. ae stie, 22 125 33. Ungulata, anterior feet 129 34. Ungulata, posterior feet . . . . . .. . 130 35. Ungulata, anterior and posterior feet . é 131 36. Phenacodus vortmanii Cope. . . . 134 37. Fore and hind limbs of Phenacodus prima and Homo sapiens . 137 38. Skull of Anapionephais Hoomnentus Cope. Lower jaw of Anaptomorphus emutus Cope be oe I51 39. Tomitherium rostratum Cope, mandible . 156 40. Tomitherium rostratum Cope, fore arm ‘ 157 41. Tomitherium rostratum Cope, ilium and femur 158 42. Megaladapis madagascariensis Forsyth ae 160 43. Skullofthe manof Spy ... . 161 44. Outlines of calvaria of the Meandecthal man, of the Spy men. 162 45. Vertical sections of symphysis mandibull of govilla’ of orang; of chimpanzee; of Spy men . 164 46, Sections of symphysis mandibuli of modern Liégois and of an ancient Parisian. ae 166 47. Molar teeth of man. . . . : 167 48. Parallelism in Brachiopoda . . , 181 49. Circulatory systems. . 194 50. Succession of horns of Cervzs shapes L. : 197 51. Shoulder-girdles of Anura ba oho Ae 198 52. Lernea branchialis dos div, ie 212 53. Entoconcha mirabilis Mill. ; 214 54. Mycetozoa. . . ‘ ; 220 55. Typhlogobius eben miensis Stead. » 245 56. Fusus parilis Con., displaying non-plicate édtowielte. 257 57. Mitra lineo/ata Heilprin, showing the plications of the columella - 259 58. Siphocyprea prbliaie. Heilaein, showing plica- tions of lips ‘ ‘ : : 260 59. Ostrea edulis, embryo . . . . 262 60. Myaarenaria . . play! der! gh. B 263 61. Modiola plicatula . . . . . ae 264 LIST OF ILLUSTRATIONS. xv PAGE . Ostrea virginiana . . : 264 : Diagrammatic representation of the segiaents of the leech. . . , . 270 . Diagrammatic se peesentstion se the rings age a pam: tive crustacean. . y a « 2975 . Diagrams of hand of Grangois ane of Astana . 274 . Elbows of man and horse rae : 280 . Elbow of horse . . . 281 . Periptychus rhabdodon Cops; slowing Do ee 5 288 . Hind foot of Poébrotherium labiatum Cope. . 290 . Hind foot of three-toed horse F 290 . United first bones of two middle toes of deer: auldiope 291 . Wrist-joints at distal extremity of forearm . . . 292 . Elbow-joint of Crocuta maculataL. . . . - 294 . Elbow-joint of chimpanzee . . be Gr wee. SBOE . Elbow-joint of Cervus elaphus . . . . . > 296 . Cervus canadensis in motion a @ Ss te 4 9 207 . Cervus elaphus . . . . 298 . Diagram of carpus of a Taxcopoi, a a diplerthrons ungulate ‘i ’ . 299 . Raccoon pacing. . . .... . 299 . Rhinocerus unicornis carpus. . 5 . 300 . Equus cabatlus fore foot ‘ 300 . Gazelladorcas . . . 301 . Pes of Marychochavasn montanus and Bos taurus . 307 . Anterior feet of primitive Ungulata : . 308 . Right posterior foot of Prothippus and Hesbeotheniiis 310 . Manus of Artiodactyla. . 4 312 . Burrs on antlers of Cosoryx necatus Leidy 316 . Diagram of excursion of lower jaw in mastication 320 . Cervus, molars . ; : 321 . Cusps of superior peal aad ieee So a 322 . Two true molars of both jaws of a ruminant 323 . Sections of superior molar teeth . . 323 . Chirox plicatus Cope, palate and molar teeth 324 . Meniscoéssus conguistus Cope, last two superior mo- lars. je eS : : be » 325 . Lemur collaris, dentiGion i ‘ . 326 . Esthonyx burmeisterti Cope, dentition ; . . 328 xvi PRIMARY FACTORS OF ORGANIC EVOLUTION. PAGE Fig. 97. Psittacotherium multifragum Cope, mandibular ra- mus ; ‘ : . 329 98. Diagrammatic representations of horizontal sections of tricuspidate molars of both jaws in mutual rela- tion é » 333 99. Delsathevium fandordeis Cope, Ragmedinty tea 336 100. Centetes ecaudatus, skull and molars 1 « 337 ror. Inferior molar crowns representing transition from the simple to the quadritubercular vis 338 102. Stypholophus whitie Cope; apposition of inferior and superior molars . : wi ti . 339 103. Cynodictis geismarianus Cope; skull a - 341 104. Aelurodon sevus Leidy; coadaptation of crowns of superior and inferior molars in mastitation 342 105. Smilodon neogeus Lund; skull. ' 344 106. Sections of crowns of molars of Ungulata .° 345 107. Castorotdes ohioensis Foster; skull . . . 347 108. Castorotdes ohioensis Foster ; skull from below 350 109. Lschyromys typus Leidy ; cranium and mandible . 351 110. Balena mysticetus ; fore limb ou 353 r11. Feet of Amblypoda . ; ‘ . 354 112. Feet of Proterotheriide ; : » 358 113. Dorsal vertebrz of casnpaadudgiaus ficties 370 114. Vertebral column of Eryops megacephalus Cope 371 114a Sleeve of coat : Bae ‘ 371 115. Metatoceras cavatiformis Hyatt . . 406 116, Do. ; , 407 117. Temnochilus crassus er 407 118. Metacoceras dubium Hyatt . » 408 119. Hyatt on Cephalopoda 410 120. Diagram explanatory of Diplogenesis 441 INTRODUCTION. ‘HE doctrine of evolution may be defined as the teaching which holds that creation has been and is accomplished by the agency of the energies which are intrinsic in the evolving matter, and without the interference of agencies which are external to it. It holds this to be true of the combinations and forms of inorganic nature, and of those of organic nature as well. Whether the intrinsic energies which accom- plish evolution be forms of radiant or other energy only, acting inversely as the square of the distance, and without consciousness, or whether they be ener- gies whose direction is affected by the presence of con sciousness, the energy is a property of the physical basis of tridimensional matter, and is not outside of it, according to the doctrine we are about to consider. As a view of nature from an especial standpoint, evolution takes its place asa distinct science. The science of evolution is the science of creation, and is as such to be distinguished broadly from the sciences which consider the other operations of nature, or the functioning of nature, which are not processes of crea- tion, but processes of destruction. This contrast is especially obvious in organic evolution, where the two processes go on side by side, and are often closely in- 2 PRIMARY FACTORS OF ORGANIC EVOLUTION. termingled, as for instance in muscular action, where both destruction of proteids and growth of muscular tissue result from the same acts, or use. Physiology, or the science of functions, concerns itself chiefly with destruction, and hence physiologists are especially prone to be insensible to the phenomena and laws of progressive evolution. The building of the embryo, remains a sealed book to the physiologist unless he take into account the allied biological science of evo- lution, as resting on the facts of botany, zodlogy, and paleontology. In his reflections on the relations of mind to matter he is likely to see only the destructive functioning of tissue, and not the history of the build- ing of the same during the ages of geological time. J. B. P. A. Lamarck! thus contrasts the theories of direct creation, and creation by evolution. The former asserts: ‘‘ That nature or its author in creating animals has foreseen all possible kinds of circumstances in which they may have to live, and has given to each species a permanent organization as well as a prede- termined form, invariable in its parts; that it forces each species to live in the place and the climate where one finds them, and to preserve there the habits which it has.” He then states his own, or the evolutionary, opinion to be: ‘*That nature in producing succes- sively all species of animals, commencing with the most imperfect or simple, and terminating its work with the most perfect, has gradually complicated their organization ; and these animals spreading themselves gradually into all habitable regions of the globe,— each species has been subjected to the influence of the circumstances in which it is ; and these have produced the habits which we observe, and the modifications of 1 Philosophie Zoologique, Paris, 1809, Vol. I., Chap. VII. INTRODUCTION. 3 its parts.” On an earlier page of the same chapter, Lamarck thus formulates the laws of organic evolu- tion, to which his name has been attached. First law. <‘‘In every animal which has not passed the time of its development, the frequent and sustained employment of an organ gradually strengthens it, de- velops and enlarges it, and gives it power proportional to the duration of its use; while the constant disuse of a like organ weakens it, insensibly deteriorates it, progressively reduces its functions, and finally causes it to disappear.” Second law. ‘‘All that nature acquires or loses in individuals by the influence of circumstances to which the race has been exposed for a long time, and in con- sequence of the influence of the predominate employ- ment of such an organ, or of the influence of disuse of such part, she preserves by generation, in new indi- viduals which spring from it, providing the acquired changes be common to both sexes, or to those which have produced new individuals.” We have here a theory of the origin of characters; , viz., of the increased development or loss of parts as a result of use or disuse. We have also the theory that the peculiarities thus acquired are transmitted to the succeeding generation by inheritance. The next formal statement of the efficient cause of organic evolution was presented by Messrs. Charles Darwin and Alfred R. Wallace in 1859.1 The cause assigned is natural selection, and Mr. Darwin thus states what is meant by this expression in his work The Origin of Species.? ‘*If under changing conditions of life organic beings present individual differences 1 Proceedings of the Li: Soctety of London. 2Ed. 1872, p. 102. 4. PRIMARY FACTORS OF ORGANIC EVOLUTION. in almost any part of their structure, and this cannot be disputed ; if there be, owing to their geometrical rate of increase, a severe struggle for life at some age, season, or year, and this certainly cannot be disputed ; then considering the infinite complexity of the rela- tions of all organic beings to each other and to their conditions of life, causing an infinite diversity of struc- ture, constitution, and habits, to be advantageous to them, it would be a most extraordinary fact if no varia- tions had ever occurred useful to each being’s own wel- fare, in the same manner as so many variations have occurred useful to man. But if variations useful to any organic being ever do occur, assuredly individuals thus characterized will have the best chance of being preserved in the struggle for life; and from the strong principle of inheritance, these will tend to produce off- spring similarly characterized. This principle of pre- servation, or the survival of the fittest, I have called natural selection. It leads to the improvement of each creature in relation to its organic and inorganic condi- tions of life; and consequently in most cases, to what must be regarded as an advance in organization. Nevertheless, low and simple forms will long endure if well fitted for their simple conditions of life.” It is readily perceived that this statement makes no attempt to account for the origin of variations, but that it simply formulates, as observed by Mr. Darwin, the doctrine of survival of such variations as are most useful to their possessors. This fact is more distinctly pointed out in the same work (p. 63) where the author remarks: ‘‘Several writers have misapprehended or objected to the term natural selection. Some have even imagined that natural selection induces variabil- ity, whereas it implies only the preservation of such INTRODUCTION. 5 variations as arise and are beneficial to the being under its conditions of life. No one objects to agriculturists speaking of the potent effects of man’s selection, and in this case the individual differences given by nature, which man for some object selects, must of necessity first occur.” Itis evident then that Mr. Darwin did not attempt to account for the origin of variations, but that the service rendered by him and by Mr. Wallace to the doctrine of evolution consists in the demonstra- tion of the reality of natural selection. Darwin also assumes in the statement first quoted above, the in- heritance of acquired characters. In 1865 the Principles of Biology of Herbert Spen- cer appeared. In this work the attempt is made to set forth the laws of organic evolution, in a way which represents an advance beyond the positions of his predecessors. He adopts and harmonizes both the Lamarckian and Darwinian doctrines, and is at times more specific in his application of Lamarck’s doctrine of the stimulus of the environment, and of use, than was Lamarck himself. Very often, however, Spencer contents himself with generalities ; or takes refuge in the ‘‘instability of the homogeneous,” as an efficient cause. This phrase, however, like his other one, ‘‘the unknowable,” is but a makeshift of temporary ignor- ance, and is neglected by Spencer himself, when he can see his way through it. He approaches the cause of the varied forms of leaves of plants in this language :! «And it will also be remembered that these equalities and inequalities of development correspond with the equalities and inequalities in the incidence of forces.” Language of similar significant but rather indefinite import is frequently used throughout this volume. 1 The Principles of Biology, by Herbert Spencer, Amer. Ed., 1873, II., p. 143. 6 PRIMARY FACTORS OF ORGANIC EVOLUTION, But in some cases Spencer is more specific. With reference to the inequality in the basal lobes of the erect leaves of Tilia and other plants, he says :1 «A considerable deviation from bilateral symmetry may be seen in a leaf which habitually so carries itself that the half on the one side of the midrib is more shaded than the other half. The drooping branches of the lime show us leaves so arranged and so modified. On examining their attitudes and their relations one to another, it will be found that each leaf is so inclined that the half of it next the shoot grows over the shoot and gets plenty of light ; while the other half so hangs down that it comes a good deal into the shade of the preceding leaf. The result is that having learned which fall into these positions, the species profits by a large deveiopment of the exposed halves ; and by survival of the fittest acting along with the direct effect of extra exposure, this modification becomes established.”” In his discussion of the origin of the characters of ani- mals, Spencer is also sometimes specific. Respecting the development of muscular insertions he remarks :? ‘‘Anatomists easily discriminate between the bones of a strong man and those of a weak man by the greater development of those ridges and crests to which the muscles are attached ; and naturalists on comparing the remains of domesticated animals with those of wild animals of the same species, find kindred differ- ences. The first of these facts shows unmistakably the immediate effect of function on structure, and, by obvious alliance with it, the second may be held to do the same, both implying that the deposit of dense sub- stance capable of great resistance habitually takes 104, cit., p. 143. 2 Loc, cit., p. 200. INTRODUCTION. 7 place at points where the tension is excessive.” Quite as specific is his ascription of the forms of epithelial cells to definite causes, as follows :! « fust the equal- ities and inequalities of dimensions among aggregated cells, are here caused by the equalities and inequalities among their mutual pressures in different directions ; so though less manifestly, the equalities and inequali- ties of dimensions among other aggregated cells, are caused by the equalities and inequalities of the osmatic, chemical, thermal, and other forces besides the me- chanical, to which their different positions subject them.” In spite of this not infrequent definiteness, Mr. Spencer occasionally falls into the error of ascribing the origin of structures to natural selection, as in the case of the forms of flowers,? and the armor-plates of paleozoic fishes.? Spencer assumes the inheritance of acquired characters throughout. ‘In 1866 Haeckel’s Schépfungsgeschichte appeared. In this work the author presents a mass of evidence which sustains the doctrine of evolution, and he com- bines the views of Lamarck and Darwin into a general system. He says:* ‘“We should, on account of the grand proofs just enumerated, have to adopt Lamarck’s theory of descent for the explanation of biological phe- nomena, even if we did not possess Darwin’s theory of selection. The one isso completely and directly proved by the other, and established by mechanical causes, that there remains nothing to be desired. The laws of inheritance and adaptation are universally acknowl- 10. cit., p. 260. 2 Of, cit., p. 153. 3 Op. cit., p. 288. 4The History of Creation, Amer. Edition, II., p. 355. 8 PRIMARY FACTORS OF ORGANIC EVOLUTION. edged physiological facts, the former traceable to prop- agation, the latter to the nutrition of organisms.” Apart from the statement that adaptation is traceable “to the nutrition of organisms,” we find nothing in Haeckel’s earlier writings which attempts the ex- planation of the origin of variations, beyond the gen- eral position assumed by Lamarck. The distinctive merit of Haeckel is his formulation of phylogeny. Much of this was speculative at the time he wrote, but so far as the Vertebrata are concerned, it has been largely confirmed by subsequent discovery. Up to this period, the form in which the doctrine of evolution had been presented, was general in its application; that is, without exact reference to the structural definitions of natural taxonomic groups. No attempt was made to show the modes of the origin of any particular class, order, or genus, and only in the most general way in the case of a few species, by Mr. Darwin. Phylogeny was untried, except by Haeckel; and this distinguished author did not attempt to ac- count specifically for the origins of the divisions whose filiations he set forth. In the year in which Haeckel’s work above cited appeared, Professor Hyatt of Boston and myself took the first step towards the formulization of a rational theory of the origin of variation, which should accord with specific examples of taxonomy. Quite indepen- dently, we selected the simple series presented by the characters of genera in their natural relations, Hyatt in the cephalopodous Mollusca, and I in the Batrachia Salientia. It is probable that Hyatt’s! article was pub- lished shortly before mine.” He says of the genera of Cephalopoda: “‘In other words, there is an increasing 1 Memoirs Boston Society Natural History, 1866, p. 193. INTRODUCTION, 9 concentration of the adult characteristics in the young of higher species and a consequent displacement of other embryonic features which had themselves also previously belonged to the adult period of still lower forms.””% My own language is:! «*That the presence, rudimental condition, or absence of a given generic character can be accounted for on the hypothesis of a greater rapidity of development in the individuals of the species of the extreme type, such stimulus being more and more vigorous in the individuals of the types as we advance towards the same, or by a reversed im- pulse? of development, where the extreme is charac- terized by absence or ‘mutilation’ of characters.” The phenomena of the aggregation of characters in pro- gressive evolution, and the loss of characters in retro- gressive evolution, were termed by me acceleration and retardation in an essay published in 1869.3 In these papers by Professor Hyatt and myself is found the first attempt to show by concrete examples of natural taxonomy, that the variations that result in evolution are not multifarious or promiscuous, but definite and direct, contrary to the method which seeks no origin for variations other than natural selection. In other words, these publications constitute the first essays in systematic evolution that appeared. By the discovery of the paleontologic succession of modifications of the articulations of the vertebrate, and especially mamma- lian skeleton, I first furnished an actual demonstration of the reality of the Lamarckian factor of use, or mo- 1 Transactions American Philosophical Society, 1866, p. 398; reprinted in The Origin of the Fittest, p. 92. 2 The expression ‘‘reversed"’ is unfortunate, d/mnished being the proper word to convey the meaning intended. 8 The Origin of Genera, Philadelphia, 1869. 10 PRIMARY FACTORS OF ORGANIC EVOLUTION. tion, as friction, impact, and strain, as an efficient cause of evolution.! This demonstration led me to the necessary inference that when the agency directive of motion is consciousness, this also has been an impor- tant factor of evolution, in demonstration of the sup- position of Erasmus Darwin.? Hyatt has demonstrated first on paleontologic evidence, the inheritance of a mechanically acquired character. Important contribu- tions to corresponding histories of the Mollusca have been made by Hyatt,’ Dall,* Jackson,’ and Beecher.® Many other contributions, into which the paleontologic evidence does not enter, have also been made by vari- ous authors in Europe and America. The authors quoted up to this point had all as sumed that the progress of evolution depends on the inheritance by the offspring of new characters acquired by the parent, and had believed that such is the fact in ordinary experience. In 1883, Weismann, in an essay on heredity, announced the opinion that charac- ters acquired by the body could not be transmitted to the reproductive cells, and could not therefore be in- herited. This doctrine rests on the relation of the germ-cells to those of the rest of the body, which is expressed in the following language of his predecessor Jaeger: ‘‘Through a great series of generations the 1“The Origin of the Hard Parts of Mammalia,” American Journal of Morphology, 1889, p. 137. 2 Origin of the Fittest, 1887, p. 357. 3 Phylogeny of an Acquired Characteristic,'’’ Proc. Amer, Philosophical Society, 1893, p. 349; ‘‘ The Genesis of the Arietide," Memocrs Mus. Compar. Zoblogy, Cambridge, Mass., 1889, XVI., No. 3. 4Dall, W. H., ‘The Hinge of Pelecypods and Its Development, Amer. Jour, Sct, Arts, 1889, XXXVIIL., p. 445. 5 Jackson, R. T., ‘‘ Phylogeny of the Pelecypoda, the Aviculidz, and Their Allies,’’ Memoirs Boston Society Natural History, 1890, IV., p. 277. 8American Journal Sei, Arts, 1893. INTRODUCTION. II germinal protoplasm retains its specific properties, dividing in every reproduction into an ontogenetic por- tion and a phylogenetic portion, which is reserved to form the reproductive material of the mature offspring. This reservation of the phylogenetic material I de- scribed as the continuity of the germ-protoplasm. ... Encapsuled in the ontogenetic material the phyloge- netic protoplasm is sheltered from external influences, and retains its specific and embryonic characters.” In other words, the reproductive cells are removed from the influence of those stimuli which affect and effect growth in the cells of the other parts of the body, so that no character acquired by the rest of the body can be inherited. The bearing of this theory on evolution is thus stated by Weismann:! ‘The origin of heredi- tary individual variations cannot indeed be found in the higher organisms, the metazoa and metaphyta, but is to be sought for in the lowest, the unicellular.” ‘‘The formation of new species, which among the lower pro- tozoa could be achieved without amphigony (sexual union), could only be attained by means of this process in the metazoa and metaphyta. It was only in this way that hereditary individual differences could arise and persist.”” In other words, variation in organic beings above the unicellular forms, has been and is, introduced only by sexual reproduction. The conclusions of Weismann were derived prin- cipally from embryologic research, and his disciples have been chiefly recruited from embryologists. These conclusions have been supported by extensive and ex- haustive investigations, which have added greatly to our knowledge of the subject. In order to account for 1£ssays, p. 296. For a complete account of Weismann’s views, see The Germ-Plasm, 1893. iz PRIMARY FACTORS OF ORGANIC EVOLUTION, the appearance of characters in the embryonic succes- sion, through influences confined to the germ-plasma, Weismann invented a theory which requires the pres- ence of distinct molecular aggregates within it, which represent the potentialities or causes. To these he has given the names of ids, idants, determinants, etc. As Weismann’s contribution to evolution has been con- fined to the department of heredity, I will consider it more particularly in the third part of this book, which is devoted to that subject. Weismann has, however, subsequently modified his views to a considerable extent. He has always admitted the doctrine of Lamarck to be applicable to the evolution of the types of unicellular organisms. His experiments on the effect of temperature on the production of changes of color in butterflies, showed that such changes were not only effected, but were sometimes inherited. This he endeavors to explain as’ follows.! ‘‘Many climatic variations may be due wholly or in part, to the simultaneous variation of correspond- ing determinants in some parts of the soma and in the germ-plasm of the reproductive cells.” This is an ad- mission of the doctrine which in 1890 I called Diplo- genesis,” and which is adopted in the present work. It appears to have been first propounded by Galton in 1875. In the chapter on Heredity I hope to offer some reasons for believing that the suggestion of Galton embraces the true doctrine of heredity. From what has preceded, two distinct lines of thought explanatory of the fact of organic evolution may be discerned. In one of these the variations of organisms which constitute progressive and regressive 1 The Germ Plasm, Contemporary Science Series, 1893, p. 406. 2American Naturalist, Dec., 1889; published in 1890. INTRODUCTION. 13 evolution appear fortuitously, and those which are beneficial survive by natural selection, while those which are not so, disappear. Characters both benefi- cial and useless or harmless, which are acquired by the adult organism, are not transmitted to the young, so that no education in habit or structure acquired by the adult, has any influence in altering the course of evo- lution. This is the doctrine of Preformation. From this point of view the cause of the variations of organ- isms has yet to be discovered. The other point of view sees in variation the direct result of stimuli from within or without the organism ; and holds that evolution consists of the inheritance of such variations and the survival of the fit through nat- ural selection. This is the doctrine of Epigenesis. To this I would add that in so far as sensations or states of consciousness are present, they constitute a factor in the process, since they enable an organism to modify or change its stimuli. The position of each of these schools on each of the questions to which reference has been made, may be placed in opposition as follows : 1. Variations appear in defi- nite directions. 2. Variations are caused by the interaction of the organic being and its environment. 3. Acquired variations may be inherited. 4. Variations survive directly as they are adapted to changing environments. (Natural selec- tion.) 1. Variations are promiscu- ous or multifarious. 2. Variations are ‘‘congeni- tal” or are caused by mingling of male and female germ-plas- mas. 3. Acquired variations cannot be inherited. 4. Variations survive directly as they are adapted to changing environments. (Natural selec- tion.) 14. PRIMARY FACTORS OF ORGANIC EVOLUTION. 5. Movements of the organ- ism are caused or directed by sensation and other conscious states. 6. Habitual movements are derived from conscious experi- ence. 7. The rational mind is de- veloped by experience, through memory and classification. 5. Movements of organism are not caused by sensation or conscious states, but are a sur- vival through natural selection from multifarious movements. 6. Habitual movements are produced by natural selection. 7. The rational mind is de- veloped by natural selection from multifarious mental activ- ities, It is not the object of the present book to present all the available evidence on both sides of each of the ques- tions above enumerated. I propose merely to submit certain facts, in support of the doctrines contained in the left-hand column of the above table. My aim will be to show in the first place, that variations of charac- ter are the effect of physical causes ; and second, that such variations are inherited. The facts adduced in support of these propositions will be necessarily prin- cipally drawn from my own studies in the anatomy, ontology, and paleontology of the Vertebrata. It will be my aim, moreover, to co-ordinate the facts of evo- lution with those of systematic biology, so that the re- sult may be as clearly presented as possible. The fail- ure to do this by the founders of evolutionary doctrine has given their work a lack of precision, which has been felt by systematic biologists. The detailed ap- plication of the principles of Lamarck and Darwin has been the work of their successors, and has necessarily thrown much new light on the principles themselves. In pursuing the object above stated, I shall be obliged INTRODUCTION. 15 to consider briefly in the following pages, the validity of the general propositions on which the doctrine of evolution rests. Less space will be given, however, to those which are less relevant, than to those which are more relevant to the doctrine of neo-Lamarckism. PART I. THE NATURE OF VARIATION. PRELIMINARY. HE structural relations of organisms may be ex- pressed in the following canons :} 1. Homology.—This means that organic beings are composed of corresponding parts ; that the variations of an original and fixed number of elements constitute their only differences. A part large in one animal may be small in another, or vice versa ; or complex in one and simple in another. The analysis of animals with skeletons, or Vertebrata, has yielded several hundred original elements, out of which the twenty-eight thou- sand included species are constructed. The study of homologies is thus an extended one, and is far from complete at the present day. 2. Successional Relation.—This expresses the fact that species naturally arrange themselves into series in consequence of an order of excess and deficiency in some feature or features. Thus species with three toes naturally intervene between those with one and four toes. So with the number of chambers of the heart, of segments of the body, the skeleton, etc. There are greater series and lesser or included series, and mis- takes are easily made by taking the one for the other. 1 Origin of the Fittest, p.6, The laws here stated are as expressive of the relations of plants as of animals. 20 PRIMARY FACTORS OF ORGANIC EVOLUTION. 3. Parallelism.—This states that all organisms in their embryonic and later growth pass through stages which recapitulate the successive permanent condi- tions of their ancestry. Hence those which traverse fewer stages resemble or are parad/el with the young of those which traverse more numerous stages. This is the broad statement, and is qualified by the details. 4. TLeleology.—This is the law of fitness of structures for their special uses, and it expresses broadly the general adaptations of an animal to its home and habits. The first and fourth of the laws above enumerated are taken for granted as generally accepted, and are -not especially considered in the following pages. The second law, that of successional relation, is discussed and illustrated under the two heads of Variation and Phylogeny; the first expressing contemporary relations, and the second, successive relations in time. The third -law, or that of parallelism, is considered in a chapter devoted to that subject. CHAPTER I—ON VARIATION. PRELIMINARY. LL species are not equally variable. Some species vary little or not at all, even under domestication. Thus the varieties of the turkey (Aledeagris gallopavo) and the guinea-fowl (Wumida meleagris) are few, and are confined to albinistic or melanistic conditions. The barnyard fowl (Gallus sp.), on the other hand, varies enormously, as does also the pigeon (Columba “ivia). Among domesticated Mammalia the variations of cats (Felis domestica) are few as compared with those of dogs (Canis sp.). Variability is not peculiar to do- mesticated animals. A large proportion of animals and plants are, in a state of nature, variable, and some of these are much more so than others. The common garter-snake (Zu¢enta sirtalis) varies exceedingly, while the variations of the allied ribband snake (Zutenia Saurita) are minute or none. But little variation has been observed in the polar bear (Ursus maritimus), while the common bear (VU. arctos) presents many vari- eties. Similar conditions are found among fishes. Thus the larger species of pike, the muskallonge (Lucius nobi- ior), the pike (LZ. estor), and the pickerel (L. vermicu- Jatus) are constant in their characters, while the small pickerel (Z. vermiculatus) presents numerous varieties. 22 PRIMARY FACTORS OF ORGANIC EVOLUTION, Many of the varieties of the animals referred to in- habit the same territory, although some are restricted to particular regions. Of geographical varieties or races much is known. Asarule, all widely distributed species present them. Examples are the brown bear of the Northern Hemisphere (Ursus arctos); the cobra di capello snake of the warmer parts of Asia (Waja tripu- dians), and that of Africa (aya haje). In North Amer- ica the king-snake (Ophibolus getulus) and the milk- snake (Osceola doliata) are represented by distinct races in different regions. On the other hand, the copperhead (Ancistrodon contortrix) and the Eastern rattlesnake (Crotalus horridus), which have a wide range, scarcely vary at all. The chub (Aydopsis bigut- tatus) is an example of a fish distributed everywhere east of the Rocky Mountains, which presents scarcely any variation. Variations are not promiscuous or multifarious, but are of certain definite kinds or in certain directions. Thus amid all their varieties, dogs never present black cross-bands on the back like those of the dog-opossum (Thylacinus cynocephalus) of Tasmania, nor do they present ocellated spots like those of the leopard, nor longitudinal stripes like those of certain squirrels. The same is true of the many varieties of cattle (Bos tau- rus), and of numerous other mammalia. Domestic fowls never vary to blue or green, colors which are common to many other birds; nor are canaries known to produce blue or red natural sports. All variations are in the first place necessarily restricted by the ex- isting characters of the ancestor; but beyond this it is evident that other conditions determine the nature of the variation. It is not supposable, for instance, that the pale tints of animals which live in dry regions ON VARIATION. 23 originated by an accident or without a determining cause. The increased amount of dark pigment ob- served in animals which dwell in especially humid re- gions must have a corresponding cause, and it is nat- urally to be supposed, of a kind the opposite of that which. produces the pale colors. I shall adduce some illustrations which show that color variations in species, as well as structural varia- tions in higher groups, have appeared in certain defi- nite series, and observe a successional relation to each other, which may or may not coincide with geograph- ical conditions. The same relation is observed in the order of appearance of variations on the body. Eimer and Weismann have shown that the grad- ual modification of color markings has originated in lizards and in caterpillars at the posterior end of the body and has gradually extended forwards. This has been discovered both by comparisons of the variations of the adults, and by studies of the order of their appear- ance in ontogenetic growth. Eimer shows that longi- tudinal bands have been produced in some animals by the confluence of spots placed in transverse series, which themselves are the remains of interrupted trans- verse bands. Thus he believes that the spots of the leopard group of the large cats were derived from the breaking up of transverse bands of the character of those now possessed by the tiger. The uniform colo- ration of the lion is the result of the obliteration of the spots. Traces of these spots may be distinctly seen in lions’ cubs. In plants variation is said to be equally definite by Henslow. He says: ‘‘In 1847 Professor J. Buckman sowed the seed of the wild parsnip in the garden of the Agricultural College at Cirencester. The seeds 24 PRIMARY FACTORS OF ORGANIC EVOLUTION. began to vary, but in the same way, though in differ- ent degrees. By selecting seed from the best rooted plants the acquired ‘somatic’ characters of an en- larged root, glabrous leaves, etc., became fixed and hereditary; and the ‘Student,’ as he called it, having been ‘improved’ by Messrs. Sutton and Sons, is still regarded as ‘the best in the trade.’ This is definile variation, according to Darwin’s definition, for those weeded out did not differ from the selected, morpho- logically, except in degree, the variations towards im- provement not being quite fast enough to entitle them to survive.” Finally I wish especially to point out that variation in animals, and probably in plants, (with which I am not so familiar,) gives no ground for believing that ‘«sports’’ have any considerable influence on the course of evolution. This is apparent whether we view the serial lines of variations of specific, generic, or higher characters ; or whether we trace the phylogeny of the animal and vegetable types by means of the paleonto- logical record. The method of evolution has appar- ently been one of successional increment or decrement of parts along definite lines. More or less abruptness in some of the steps of this succession there may have been ; since a definite amount of energy expended in a given direction at a given point of history might pro- duce a much greater effect than the same amount ex- pended at some other period or point of evolution. This might be due to the release of stored energy, which could only be accomplished by a coincidence of circumstances. A simple illustration of the phenome- non of abrupt metamorphosis is to be found in the passage of matter from the gaseous to the liquid, and ON VARIATION. 25 from the liquid to the solid state. I have stated the case in the following language:} ‘As one or more periods in the life of every spe- cies is characterized by a greater rapidity of develop- ment ” (ontogenetic) ‘‘than the remainder ; so in pro- portion to the approximation of such a period to the epoch of maturity or reproduction is the offspring liable to variation. During the periods corresponding to those between the rapid metamorphoses, the char- acters of the genus would be preserved unaltered, though the period of change would be ever approach- ing. Hence the transformation of genera may have been rapid and abrupt, and the intervening periods of persistency very long. Thus, while change is really progressing, the external features remain unchanged at other than those points, which may be called ex- pression points. Now the expression point of a new gen- eric type is reached when its appearance in the adult falls so far prior to the period of reproduction as to transmit it to the offspring and their descendants.” 1. VARIATIONS OF SPECIFIC CHARACTERS. a. Variations in the Coleopterous Genus Cicindela. Dr. George H. Horn has traced the variations in the color patterns of the elytra of the North American species of this abundant and well-known genus. He shows that they form series, in the following language :? ‘‘Any one in glancing over this series will perceive that there is a great similarity of marking between many species. This similarity, which may be con- sidered as the type of marking, and is illustrated by l Origin of the Fittest, p. 79. 2 Extomological News, Philadelphia, Feb., 1892, p. 25. 26 PRIMARY FACTORS OF ORGANIC EVOLUTION, No. 1 of the accompanying plate (Fig. 1) is the under- lying pattern from which all the forms observed in our Cicindela have been derived. “Before going further it is well to present the fol- lowing propositions that the argument ard the illus- trations may be understood. ‘<1, The ¢ype of marking is the same in all our spe- cies. «¢2, Assuming a well-marked species as a central type the markings vary, a, by a progressive spreading of the white, 4, by a gradual thinning or absorption of the white, c, by a fragmentation of the markings, d, by linear supplementary extension. «¢3, Many species are practically invariable. These fall in two series, a, those of the normal type, as vulgaris, hirticollis, and fenutsignata, 4, those in which some modification of the type has become permanent, probably through isolation, as marginipennis, togata, and lemniscata. ‘¢4. Those species which vary, do so in one direc- tion only. That is, supposing a species begins typ- ically with markings similar to vudgaris, the variation may be either in the direction of thickening and in- crease of white, as in hyperborea, generosa, and others, or in the direction of thinning or fragmentation of the white with perhaps an entire loss of markings as in hemorrhagica, splendida, or obsoleta. «The first two propositions must be considered as applying to the species of the genus collectively, the last two to the species separately. ‘“«The accompanying plate has been prepared to illustrate these propositions. It must, however, be 27 ON VARIATION. a ~ o ss) & BS 3) id oO sj ro) ‘=| i) io} | 4 a wt & 28 PRIMARY FACTORS OF ORGANIC EVOLUTION. understood that, in tracing the derivations from the typical, it is not possible to use one species, as these modifications go on gradually through a number of species, one sometimes beginning where another ends. ‘In the plate, No. 1 represents vu/garis, which is a fairly typical species, following through generosa (2-3), pamphila (4), hyperborea var. (5), togata (6), gra- tiosa (7), canosa (8), we finally arrive at a perfect white elytron as seen in some varieties of dorsalis. “‘Following in the other direction through ¢enuz- Signata (9), marginipennis (10), hentztd (11), sexguttata (12), hemorrhagica (13), and splendida var. (14), it will be observed that through a gradual thinning or ab- sorption of the markings, or by their fragmentation and obliteration, we arrive at the opposite result of elytra without any white markings whatever, as in many forms of odsoleta, scutellaris, punctulata, and he- morrhagica. ‘« Those species which vary from the type in hav- ing the markings broken into spots, as in 12-gutta/a or hentzit, may lose the spots by a gradual decrease of size, so that they all seem to disappear nearly at the same time; or the spots may disappear successively, those on the disc being the first to go, while the mar- ginal spots remain. «‘From our series it would be difficult to say which spot is the most persistent, but it is probably the lu- nule, as there are more with entirely dark elytra with slight traces of this spot than with any other, as shown in abdominalis and punctulata. ‘¢Forms like Zemniscata (16) seem very far removed from the type, but many forms of zmperfecta (15) show how the markings gradually leave the margin and tend ON VARIATION. 29 by fusion to form a vitta at first somewhat oblique, but finally becoming nearly median. ‘« The last two figures on the plate represent the markings of gaddci (17) and saudcy? (18), in which the ends of the bands or lunules are greatly prolonged. The latter form, which represents dorsalis as well, is but rarely seen so perfectly marked, the tendency being toward a greater extension of the white. The other species is scarcely variable, although equally a coast form. ‘Those species which retain a permanent diver- gence from the normal standard, such as /ogata (6) or lemniscata (16), are doubtless descendants from a nor- mal type which has varied, and in which a variety has become isolated and perpetuated itself.” The accompanying plate is copied from the original drawing by Dr. Horn, and which accompanies the paper now cited. &. Variations tn the Osceola doltata. The Milk-Snake, Osceola doliata Linn., ranges in North America over the Eastern, Central, and Austro- riparian districts, and is absent from the Sonoran and Pacific districts. It is found also in the humid regions of Mexico and Central America, as far as the Isthmus of Darien. Beyond this point it does not oc- cur, but a very similar snake (Opheomorphus mimus) is found in New Grenada. I have called attention to the color variations of this species in a brief paragraph in the introduction to my check list of Batrachia and Reptilia in North America, 1875,) and have given the characters of the 1Bulletin of the U. S. National Museum, No. I, p. 4. 30 PRIMARY FACTORS OF ORGANIC EVOLUTION. color types, or subspecies, in an analytical key, ina ‘‘Review of the Characters and Variations of the Snakes of North America,” 1892.! I have also given a series of figures representing the North American color forms, for which I am indebted to the United States National Museum, which are here reproduced. Before going further into the patterns of the Osce- ola doliata, 1 give a synoptic key of them. I. No yellow band posteriorly from orbit (a yellow half-collar). a. Dorsal spots or saddles (red) open at the side, the borders of adjacent spots forming pairs of black rings. Interspaces between red saddles open below ; scales not black-tipped ; front more or less black ; first black ring on nape only : O. d. coccinea, Interspaces between red saddles closed by black spot be- low ; scales black tipped ; front black; first black ring complete : O. a. polyzona. Interspaces not closed; rings, including first, complete on belly; first yellow band crossing occipital plates; front black ; scales not black-tipped : O. d. conjuncta, aa. Dorsal spots closed at the sides below, forming saddles. 6, Saddles closed by a single black tract on the middle of the belly; no spots between the saddles. Dorsal spots undivided medially; front black ; first black ring complete : O. d. annulata, Dorsal spots divided longitudinally by a median black connection ; front black : O. d. gentilis. 46. Inferior borders of saddles separate and not confluent with each other. Saddles completed on gastrosteges; no alternating spots ; no black collar : O. d. parallela. Saddles completed on gastrosteges ; spots opposite inter- vals forming a single series on the middle line of the belly : O. d. syspila, Saddles completed above the gastrosteges; alternating spots which do not meet on the middle line of the belly: O. d. doliata. 1 Proceedings of the U.S, National Museum, XIV, p. 589-608. ON VARIATION, 31 II. A yellow band posteriorly from orbit, bounded below by a black or brown one. a, Saddle spots closed laterally on gastrosteges ; alternate spots entirely on gastrosteges. A half collar behind parietal plates, no superciliary stripe : - O. d. temporalis. aa. Saddle spots closed above gastrosteges ; alternate spots on scales. : A half collar nearly or quite touching occipital plates, no bands ; alternate spots partly on gastrosteges: O. d, collaris, Neck with longitudinal bands ; alternate spots partly on gastrosteges . O. a. clerica. Neck with bands ; alternate spots entirely on scales: O. d. triangula. In Fig. 2 are represented vertical, lateral, and in- ferior views of parts of the body of the subspecies ¢77- angula, taken from a specimen in my collection from Westchester County, New York, which I owe to the kindness of my friend, Mr. T. H. Mead. The characters of this form are seen in (1) the presence of a light band extending from the posterior angle of the eye downward and backward, which is bounded by a black border above and below; (2) a black cross-band on the posterior border of the pre- frontal plates; (3) chevron shaped mark with the apex on the posterior part of the frontal plate, whose limbs extend posteriorly as a band on each side of the neck, where they are fused together, and continue as a sin- gle, broad band for a short distance; (4) a series of lateral spots which do not extend beyond the scales on to the gastrosteges, and which alternate with the dor- sal spots; (5) a series of spots on the ends of the gas- trosteges which alternate with the last mentioned; (6) a series of spots on the centers of the gastrosteges which alternate with the spots mentioned under (5). 32 PRIMARY FACTORS OF ORGANIC EVOLUTION. an SS. %, ee eatnt +) ee, See oe C é Osceola doliata triangula. Ke | iy A ini SC) Fig. 2. $4 ee SR Osceola doliata clerica. Fig. 3. ON VARIATION. 33 The ground color in this form is gray, and the spots are a rich brown with black borders. The belly has a white ground color. In Fig. 3 we have the subspecies clerica, where the following modifications appear. The fusion of the limbs of the chevron is more complete, and the dorsal spots are more expanded transversely. They extend to within two or three scales of the gastrosteges, while in the form ¢rzangulus they are five scales distant. The alternate spots touch the gastrosteges. This figure is taken from a specimen in the Museum of the Phila- delphia Academy from southern Illinois. In Fig. 4 we have an individual from Elmira, IIli- nois, which illustrates the characters of the form co/- laris. Here the chevrons are distinct from the first dorsal spot, whose anterior black border forms a half collar on the neck. This specimen is instructive, as it displays the last connection between the chevron and the first spot, in a black line on each side. This is wanting in the typical ce/aris. The collar of ground color is complete in its an- terior border, as well as the posterior in the form ¢em- poralis (Fig. 5), owing to the disappearance of the chevron. ‘The transverse band on the prefrontals has also disappeared. The anterior extremity of the post- orbital stripe is cut off, and consists of a spot of ground color. The dorsal saddle spots are wider, reaching the gastrosteges, while the intermediate spots are exclu- sively gastrostegal. The spots whichealternate with them, have fused on the middle line. Fig. 5 is from a specimen from the State of Delaware. In subspecies do/iata the postocular stripe has dis- appeared, and the chevron:is replaced by a black patch on the parietal and temporal plates. In other respects 34 PRIMARY FACTORS OF ORGANIC EVOLUTION. i r) xX Se tear Osceola doliata collarts. Fig. 4. Osceola doliata temporalis, Fig. 5. ON VARIATION. 35 this form is more like the form co//aris. The dorsal saddle spots are separated by a row or two of scales from the gastrosteges, and their alternating spots are partly on the scales. The ground color in this form, as in the zemporalis, approaches red. This is the form of the tier of states between latitude 4o° and the Gulf States. The subspecies sysfiJa is represented in Fig. 8. The head pattern is like that of doliata with the black patch more or less reduced—in the specimen figured being represented by a cross stripe. The dorsal saddle spots are more expanded than in any form yet encoun- tered, their lateral borders being completed below the scales and entirely on the gastrosteges. The alternate spots now meet and fuse on the middle line of the ab- domen, and the second series of alternating spots has disappeared. This is distinctively a southern form, extending west to central Oklahoma. The dorsal saddles are so far extended in the next subspecies, parallela, as to form two parallel stripes with a narrow strip of ground color between, on the middle line of the abdomen. The alternating spots have disappeared. In the specimen figured, which is from Florida, and is in the United States National Museum, the supraocular spots seen in /emforalis, are indicated. The ground color is red. Black begins to appear on the head. From the form syspz/a two types of color modifica- tion may be traced. One of these brings the borders of the saddle spots together on the median line, form- ing a median black stripe; this is the subspecies annw- Jata, which belongs to western Texas and the adjacent parts of Mexico. The top of the head is black (Fig. 10). In the other, the lateral borders of the saddle 36 PRIMARY FACTORS OF ORGANIC EVOLUTION. Figs. 6-7. Osceola doliata doliata, ON VARIATION. 37 spots have disappeared altogether, so that the body is more or less completely encircled by pairs of black rings, the alternating spots having disappeared. This might be supposed to have resulted from a continua- tion of the process by which the alternating spots have disappeared, and the edges of the saddles been brought closer and closer together. The continued transverse extension of the spot color would finally obliterate the lateral borders completely, as actually occurs in this last form, the coccinea of authors, which is the com- mon type of the Gulf Coast. But the black has not covered the head and muzzle of this form as in the an- nulata. ‘These regions are on the contrary red, as is the spot color generally, while the ground color is pale yellow. A tendency to a development. of black pigment in the saddle spots is seen in two other forms. The sub- species gentilis resembles annulata, but has a black longitudinal dorsal band which divides each saddle spot in two equal halves. This is a rare form, only known from the Indian Territory. The common Mex- ican form (folyzona) has the paired rings of coccinea, the black head of annulata, but each scale of the red intervals is tipped with black. The relations of these forms may be expressed ina tabular form, given on page 39. The main series corresponds with a distribution in latitude, commencing with the ¢riangula of New Eng- land and New York, and passing gradually to the coc- cinea of the Gulf Coast regions, and Jolyzona of Mex- ico and Central America. The forms of the right-hand column are (except the paradllela) from the central warmer parts of the continent. This series of color-forms of the Osceola doliata 38 PRIMARY FACTORS OF ORGANIC EVOLUTION. a i» 0, 10, 8 Osceola doliata syspila. Fig. 8. PA este Sa Osceola doliata parallela, Fig. 9. ON VARIATION. 39 demonstrates the following points. First: the color- variation is determinate and not indeterminate. It consists in, a, the successive enlargement of the dorsal spots toward, to, and across, the belly; 4, the diminu- tion and extinction of the longitudinal stripes on the head ; ¢, do. of the spots of the inferior surface of the body; @, in the increase of red in the color of the dorsal spots, coincidentally with the changes men- tioned. Second: these color-changes follow parallels gentilis polyzona annulata conjuncta parallela coccinea syspila / doliata “~~_temporalis collaris clerica triangula of latitude, the red color and accompanying changes developing from north to south. Third: so far as re- gards eastern North America, there is a diminution of size in passing from north to south; the O. @. coccinea being the smallest of the subspecies. In Mexico, the size is recovered, as the O. @. polyzona equals in di- mensions the O. d. triangula. The young of the northern O. d. ¢riangula pre- sents the colors of the dorsal spots nearly as brilliant as those of the southern O. d. coccinea, and they fade 40 PRIMARY FACTORS OF ORGANIC EVOLUTION. Be e: my 9, CoN ™~ Fig. 10. Osceola doliata annulata. Osceola doliata coccinea, Fig. 11. ON VARIATION. 41 with age to the adult character. The pattern in the young at the period of hatching is the same as that of the adult. ¢. Color-Variations in Cnemidophorus. Another illustration of the nature of color-variation is to be found in certain species of the lacertilian gen- era Cnemidophorus in America, and Lacerta in Eu- rope. In both genera the color-markings differ in the same individual at different ages, and the age at which the acult coloration ds assumed, differs in different lo- calities. Some of the species, e. g., Cnemidophorus sexlineatus, never abandon the coloration of the young of other species and subspecies. The same condition is characteristic of the C. deppet of Mexico, the C. Zem- niscatus of Brazil, and other.species. - The process of color-modification in the C. Zessellatus and C. gudaris of North America is, as J have pointed out,! as follows: The young are longitudinally striped with from two to four stripes on each side of the middle line. With in- creasing age, light spots appear between the stripes in the dark interspaces. Ina later stage these spots increase in transverse diameter, breaking up the dark bands into spots. In some of the forms these dark spots extend themselves transversely and unite with each other, forming black cross-stripes of greater or less length. Thus we have before us the process by which a longitudinally striped coloration i is transformed into a transversely striped one. The large number of specimens of the C. ¢essellatus and C gularis in the National Museum collection show that the breaking up of the striped coloration appears 1 Proceeds.. Amer. Philos, Soc., 1885, p. 283. Transac, Amer, Philos. Soc 1892, p. 27. 42, PRIMARY FACTORS OF ORGANIC EVOLUTION. =i) esemmanlly Pepcid Om ‘C131 *snqvpjassaz snaoygopruauy ‘y-p ON VARIATION. 43 first at the posterior part of the dorsal region (i. e., the sacral and lumbar). The confluence of the spots ap- 2 & = z ° be s 8 - oR BASES be ay Aan yy TU hy aleeatetes r af A Rs pb ro pears there first; and finally (C. gularis semifasciatus), where the color markings disappear, leaving a uniform hue, this also appears first at the posterior part of the 44 Fr 31g *$2]DANUL DILIIVT PRIMARY FACTORS OF ORGANIC EVOLUTION. MBARBRRKTS eee ae <3 SoS rrs — ON VARIATION. 45 body. In the C. ¢essellatus rubidus the dark spots dis- appear first on the anterior regions. According to Eimer,! among many color-variations of the Lacerta muralis there exists a series of types closely similar to those observed by me to characterize the two species of Cnemidophorus mentioned. I give figures of these series in all three species. It will be observed that in the second and third forms (B and C) of the Z. muralis, the pale portions of the dark stripes do not assume the very light hue of the ground color as they do in the corresponding phase of the Cnemido- phorus tessellatus (C and D, Fig. 12), but this interme- diate condition is exactly paralleled by the subspecies mariarum of the Cnemidophorus gularis. The corre- spondences are represented in the table on page 46. There are some color forms in the Lacerta muralis which are not repeated in the North American Cnemi- dophori, particularly those which result in a strong contrast between the dorsal colors as a whole and the darker lateral colors, as a band. The color variety, No. 7, of the Cnemidophori is not reported by Eimer as occurring in the Lacerta murals. The variations from one to four form a direct series, and so do those represented by Nos. 1, 2, 3, and 5. Such variations cannot be regarded as promiscuous, especially when the same process of change is to be observed in three different species, one of which in- habits a continent remote from the other two. @. Variations in North American Birds.and Mammals in Relation to Locality. The distinguished zodlogist, Dr. J. A. Allen of New York, has made a thorough study of this subject with Lrchiv fir Naturgeschichte, 1881, p. 239. 46 PRIMARY FACTORS OF ORGANIC EVOLUTION. Cnem. tessellatus, Cnem. gularis, Other Cnemidoph't. Lacerta muralis, Longitudinally striped. . Dark interspaces pale spotted ..... Dark interspaces divided by light ColOM seine ees Dark spots confluent transversely, forming crossbars.............. Light spots not confluent; light stripes broken up; pattern reticu- Tated a sicaire oones Dark stripes interrupted by darker color. ..... Dark spots separate and on a brown PToUNd sas chee pede hada nian C. t. perplexus. C. t. tessellatus a. C. t. tessellatus B. C. t. tessellatus y. (orm thus. melanoste- C. t. rubidus, C g. gularis a. C. g. scalaris a. C. g. scataris B. C. g. &. costatus, C. t, mariarum., C. g. semifasciatus, C. octolineatus. C. sexlineatus. C. labialis, C. septemvitiatus. C. grahamit. C. variolosus. L. m, campestris. L. m, albiventris. Lm, striatomacu- Zata, L. m.veticulata. Lim, tigris, L. m, punctulato- SJasctata. L. m. maculostri- aa, ON VARIATION. 47 ample material at his disposal. Following lines al- ready laid down by Prof. Spencer F. Baird, Dr. Allen has shown that variations in form, size, and color are directly related to latitude, and that they are not pro- miscuous or multifarious, but are definite and graded. I make the following extract from a summary of the subject published by him in The Radical Review (New Bedford, Mass.) for May, 1877: ‘¢Geographical variation, as exhibited by the mam- mals and birds of North America, may be summarized under the following heads, namely: (1) variation in general size, (2) in the size of peripheral parts, and (3) in color; the latter being subdivisible into (a) vari- ation in color with latitude, and (4) with longitude. As arule, the mammals and birds of North America increase in size from the south northward. This is true, not only of the individual representatives of each species, but generally the largest species of each genus and family are northern. There are, however, some strongly marked exceptions, in which the increase in size is in the opposite direction, or southward. There is for this an obvious explanation, as will be presently shown ; the increase being found to be almost invari- ably toward the region where the type or group to which the species belongs receives its greatest numer- ical development, and where the species are also most specialized. Hence the representatives of a given spe- cies increase in size toward its hypothetical center of distribution, which is in most cases doubtless also its original center of dispersal. Consequently there is fre- quently a double decadence in size within specific groups, and both in size, and numerically in the case of species, when the center of development of the group to which they belong is in the warm-temperate or trop- 48 PRIMARY FACTORS OF ORGANIC EVOLUTION. ical regions. This may be illustrated by reference to the distribution of the higher classes of vertebrates in North America. Among the species occurring north of Mexico there are very few that may not be supposed to have had a northern origin; and the fact that some are circumpolar in their distribution, while most of the others (especially among the mammals) have congeneric Old World allies further strengthens the theory of their northern origin. Not only do individuals of the same species increase in size toward the north, but the same is true of the species of different genera. Again, in the exceptional cases of increase in size southward, the species belong to southern types, or, more correctly, to types having their center of development within or near the intertropical regions, where occur, not only the greatest number of the specific representatives of the type, but also the largest. ‘‘For more detailed illustration we may take three families of the North American Carnivora; namely, the Canidz (wolves and foxes), the Felide (lynxes and wild cats), and the Procyonide (raccoons). The first two are to some extent cosmopolitan, while the third is strictly American. The Canide have their largest specific representatives, the world over, in the temperate or colder latitudes. In North America the family is represented by six species,! the smallest of which (speaking generally) are southern, and the larg- est northern. Four of them are among the most widely distributed of North American mammals, two (the gray wolf and the common fox) being circumpolar spe- cies ; another (the Arctic fox) is also circumpolar, but 1 The gray wolf (Canzs Zupus), the coyote (C. latrans), the Arctic fox (Vud- pes lagopus), the common fox (l, alopfex), the kit fox (V. velox), and the gray fox (V7. cinereoargentatus), ON VARIATION. 49 is confined to high latitudes. The three widest-rang- ing species (the gray wolf, the common fox, and the gray fox) are those which present the most marked variation in size. Taking the skull as the basis of comparison, it is found that the common wolf is fully one-fifth larger in the northern parts of British Amer- ica and Alaska than it is in Northern Mexico, where it finds the southern limit of its habitat. Between the largest northern skull and the largest southern skull there is a difference of about thirty-five per cent. of the mean size’ Specimens from the intermediate region show a gradual intergradation between these extremes, although many of the examples from the upper Mis- souri country are nearly as large as those from the ex- treme North. «©The common fox, though occurring as far north as the wolf, is much more restricted in its southward range, especially along the Atlantic coast, and presents a correspondingly smaller amount of variation in size. The Alaskan animal, however, averages about one- tenth larger than the average size of specimens from New England. In the gray fox, whose habitat ex- tends from Pennsylvania southward to Yucatan, the average length of the skull decreases from about five inches in Pennsylvania to considerably less than four in Central America—a difference equal to about thirty per cent. of the mean size for the species. «The Felide, unlike the Canidz, reach their great- est development, as respects both the number and the size of the species, in the intertropical regions. This family has but a single typical representative— the panther (Fe/s concolor)—north of Mexico, and this ranges only to about the northern boundary ot the United States. The other North American represen- 50 PRIMARY FACTORS OF ORGANIC EVOLUTION, tatives of the family are the lynxes, which, in some of their varieties, range from Alaska to Mexico. They form, however, the most northern, as well as the most specialized or ‘aberrant,’ type of the family. While they vary greatly in color, as well as in the length and texture of the pelage, at different localities, they afford a most remarkable exception to all laws of variation in size with locality; for a large series of skulls, repre- senting localities as widely separated as Louisiana, Northern Mexico, and California, on the one hand, and Alaska and the Mackenzie River district on the other, as well as various intermediate localities, reveals no ap- preciable difference in size throughout this wide area. The true cats, however, as the panther and the ocelots, are found to greatly increase in size southward, or to- ward the metropolis of the family. The panther ranges from the Northern States southward over most of South America. Skulls from the Adirondack region of New York have an average length of about seven and a half inches, the length increasing to eight and three- quarters in Louisiana and Texas, from beyond which points there is lack of data. The ocelot (Felis parda- Zs) finds its northern limit near the Rio Grande of Texas, and ranges thence southward far into South America. The average size of Costa Rican examples is about one-fifth greater than that of specimens from the Rio Grande. Instances of increase in size northward among the Carnivora of North America are so generally the rule that further space need not be taken in recounting ex- amples in detail. It may suffice to state that the badger (Zaxidea americana), the marten (Mustela ame- ricana), the fisher (AZ. pennanti), the wolverine (Guo luscus), and the ermine (Putorius ermineus)—all north- ON VARIATION. 51 ern types—afford examples of variations in size strictly parallel with that already noticed 4s occurring in the foxes and wolves. “¢To refer briefly to other groups, it may be stated that the Cervidz (deer family) are mainly rather north- ern in their distribution; that the largest species occur in the colder zones, and that individuals of the same species increase rapidly in size toward the north. Some of the species, in fact, afford some of the most striking instances of northward increase in size; among which are the Virginia deer and its several representa- tives in the interior of the continent and on the Pacific Slope. It is also noteworthy that the most obviously distinctive characteristic of the group—the large, an- nually deciduous antlers—reaches its greatest devel- opment at the northward. Thus all the northern spe- cies, as the moose, the elk, and the caribou, have branching antlers of immense size, while the antlers are relatively much smaller in the species inhabiting the middle region of the continent, and are reduced to arudimentary condition—a simple, slender, sharp spike, or a small and singly forked one—in the tropical spe- cies; the antlers declining in size much more rapidly than the general size of the animal. This is true in individuals of the same species as well as of the species collectively. “The Glires (the squirrels, marmots, spermophiles mice, and their affines) offer the same illustrations in respect to the law of increase in size as the species already mentioned, the size sometimes increasing to the southward, but more generally to the northward, since the greater number of the species belong decid- edly to northern types. There is no more striking in- stance known among mammals of variation in size 52 PRIMARY FACTORS OF ORGANIC EVOLUTION. with locality than that afforded by the flying squirrels, in which the northern race is more than one-half larger than the southern; yet the two extremes are found to pass so gradually the one into the other, that it is hardly possible to define even a southern and a north- ern geographical race, except on the almost wholly arbitrary ground of difference in size. The species, moreover, is one of the most widely distributed, rang- ing from the Arctic regions (the northern limit of for- ests) to Central America. “Among birds the local differences in size are al- most as strongly marked as among mammals, and in the main, follow the same general law. A decided increase in size southward, however, or toward the warmer latitudes, occurs more rarely than in mam- mals, although several well-marked instances are known. The increase is generally northward, and is often very strongly marked. The greatest difference between northern and southern races occurs, as in mammals, in the species whose breeding-stations em- brace a wide range of latitude. In species which breed from Northern New England to Florida, the southern forms are not only smaller, but are also quite different in color and in other features. This is eminently the case in the common quail (Ortyx virginianus), the meadow-lark (Sturnella magna), the purple grackle (Quiscalus purpureus), the red-winged blackbird (Age- lacus phaeniceus), the golden-winged woodpecker (Co- laptes auratus), the towhee (Pipilo erythrophthalmus), the Carolina dove (Zenadura macrura), and in nu- merous other species ; and is quite appreciable in the blue-jay (Cyanurus cristatus), the crow (Corvus amert- canus), in most of the woodpeckers, in the titmice, numerous sparrows, and several thrushes and war- ON VARIATION. 53 blers, the variation often amounting to from ten to fifteen per cent. of the average size of the species. ‘‘As a general rule, certain parts of the organisms vary more than does general size, there being a marked tendency to enlargement of peripheral parts under high temperature, or toward the tropics,—hence south- ward in North America. This is more readily seen in birds than in mammals, in consequence, mainly, of their peculiar type of structure. In mammals it is manifested occasionally in the size of the ears and feet, and in the horns of bovines, but especially and more generally in the pelage. At the northward, in individ- uals of the same species, the hairs are longer and softer, the under fur more abundant, and the ears and the soles of the feet better clothed. This is not only true of individuals of the same species, but of northern species collectively as compared with their nearest southern allies. Southern individuals retain perma- nently, in many cases, the sparsely clothed ears and the naked soles that characterize northern individuals only in summer, as is notably the case among the dif- ferent squirrels and sphermophiles. ‘¢In mammals which have the external ear largely developed—as in the wolves, foxes, some of the deer, and especially the hares,—the larger size of this organ in southern as compared with northern individuals of the same species is often strikingly apparent. It is more especially marked, however, in species inhabit- ing extensive open plains and semi-desert regions. The little wood hare, or gray rabbit (Lepus sylvaticus), affords a case in point. This species is represented, in some of its varieties, across the whole breadth of the continent, and from the northern border of the United States southward to Central America, but in 54 PRIMARY FACTORS OF ORGANIC EVOLUTION. different regions by geographical races or subspecies. In addition to certain differences of color and general size, the ears vary still more strongly. In the form inhabiting the Great Plains, commonly known as the little sage-brush hare (LZ. sylvaticus nuttall?), the ears are considerably longer than in the eastern variety, and increase in size from the north southward, reach- ing their greatest development in Western Arizona and the desert region further westward and southward, where the variety is characterized mainly by the large size of its ears, which are in this race nearly twice the size they attain in the eastern variety. In the long- eared ‘jackass’ hares of the plains, the ear likewise increases in size to the southward. In Lepus callotis, for example, which ranges from Wyoming southward far into Mexico, the ear is about one-fourth to one- third larger in the southern examples than in the north- ern. The little brown hare of the Pacific Coast (Z. trowbridge’) presents a similar increase in the size of the ear southward, as does, to a less extent, the prairie hare (LZ. campestris). Not only are all of the long- eared species of American hares confined to the open plains of the arid interior of the continent, but over this same region is the tendency to an enlargement of the ear southward stronger than elsewhere. It is also of interest in this connection that the largest-eared hares of the Old World occur over similar open, half- desert regions, as do also the largest-eared foxes. On our western plains, the deer are represented by a large- eared species. Among the domestic races of cattle, those of the warm temperate and intertropical regions have much larger and longer horns than those of north- ern countries ; as is shown by a comparison of the Texan, Mexican, and South American breeds, with ON VARIATION. 55 the northern stock, or those of the South of Europe with the more northern races. In the wild species of the Old World, the southern or sub-tropical are re- markable for the large size of their horns. The horns of the American prong-horn (Antilocapra americana) are also much larger at southern than at northern locali- ties.! Naturalists have also recorded the existence of larger feet in many of the smaller North American Mammalia at the southward than at the northward, among individuals of the same species, especially among the wild mice, in some of the squirrels, the opossum and raccoon, as well as in other species. ‘ zaddq -OAU0D ‘1adse{ | “aqNpoOAUt A[suIS pue peisei9 (‘artsoddg) Crep) | (epessyueq)| zz * "ANHOOIPY soraydsruayy |'aqnjoaur A[qnoq ‘sajoieqn}-F | ‘pajaoeq | “sUuryOOTIEyU] | *paaco15y ‘opeisiiisiq a) bee * INDY “NIVUG "SUSAHdOdVOAZ “SavIOW =nOnA bert ‘sat ‘LAG ‘SA0L| -Norvwuo. worwadAS ‘saiavy | anv snaavg | -VOVYISY ‘ON a 140 PRIMARY FACTORS OF ORGANIC EVOLUTION, of the Cetacea and Sirenia. The results attained by the study of the paleontology of the other orders may be summarized as follows : First. It is probable that the common ancestors of the placental and implacental lines of Mammalia are known to us in some of the types of the Jurassic per- iod. Whether they were marsupial in the sense of possessing an external pouch for the young or not, is immaterial. They were probably marsupial in brain characters, in the structure of their reproductive sys- tem, and in the absence of placenta. To this source the existing polyprotodont marsupials may be traced, through such forms as Myrmecobius. The multi- tuberculate type has a contemporary history, and one distinct from that of the Polyprotodontia, and its an- cestry has not yet been discovered. Their earliest forms (of the Jurassic and Triassic) are already highly specialized. They probably represent the Monotre- mata of their time. Second. The immediate didelphian ancestors of the monodelphous Mammalia have not yet been certainly discovered. In the oldest of the latter (of the Puerco epoch) numerous points of approach to the insectivo- rous Jurassic forms occur, especially in the prevalent trituberculy of the molars in both epochs. Third. The phylogeny of the clawed group has been traced back to a common ordinal form which has been called the Bunotheria. Of these the most genera- lized are the Creodonta, from which we may trace the Carnivora, the Insectivora, and the Tillodonta, and probably all other Unguiculata. The Ancylopoda only have undergone the alternation of the carpal and tar- sal bones, which obtains in the diplarthrous Ungulata. fourth. The phylogeny of the hoofed groups car- PHYLOGENY, 141 ries us back to the order Condylarthra, the hoofed cotemporary of the Bunotheria. The even and odd toed hoofed mammals are traceable back to the Am- blypoda, whose oldest representatives are the Panto- donta of the Puerco. The Proboscidia and Hyracoidea come directly from the Condylarthra. Moreover, the phalanges of the lemurs are not distinguishable by any important characters from the hoofs of the Hyracoidea and Condylarthra. Not only this, but the structure of the foot in these three groups is identical in regard to the mode of articulation of the first and second rows of the tarsal and carpal bones. Fifth. The characters of the feet of the Condylar- thra agree with those of unguiculate placental Mam- malia, and bind the two series together. The synthesis of the ungulate and unguiculate lines is accomplished by exceptions to the characters which define them. Thus the hoofs of Pantolambda (Amblypoda), Peripty- chus (Condylarthra), and Mesonyx (Creodonta) do not differ by any marked character. Claws occur in the Hapalide of the quadrumanous line, and the ungues of some Glires are absolutely intermediate between the hoofs and claws. Many Edentata have claws on the fore feet and hoofs on the hind feet. The Condy- larthra with tritubercular molar teeth are then trace- able to Bunotheria with tritubercular teeth, of which many are known from the Puerco beds; and the quadri- tubercular forms from corresponding quadritubercular or tritubercular Bunotheria, of which latter, some are known. . Sixth. The anthropoid line may be traced directly through the lemurs to the Condylarthra. The changes which have taken place in the skeleton are slight, and 142. PRIMARY FACTORS OF ORGANIC EVOLUTION. consist among other points in a rotation of the second row of carpal bones inwards on the first row, in the anthropoid apes and man, similar to that which has occurred among the Ungulata, but it has not become so pronounced. As a result we get the general phylogenetic scheme as shown on the following page. In this diagram, divisions of greater and lesser rank are mixed, so as to display better some of the relation- ships. Thus all the divisions whose names stand on the right side of the middle vertical line are unguicu- lates ; and those on the left side of the line, excepting Sirenia and Cetacea, are ungulates. The three names in the middle vertical line are those of the suborders of the Taxeopoda. A review of the characters of the existing Mam- malia as compared with those of their extinct ances- tors displays a great deal of improvement in many ways, and but few instances of retrogression. The succession in time of the Monotremata, the Marsupia- lia, and the Monodelphia, is a succession of advance in all the characters of the soft parts and of the skele- ton which define them (see table of classification). As to the monotremes themselves, it is more than prob- able that the order has degenerated in some respects in producing the existing types. The history of the Monotremata is not made out, but the earliest forms of which we know the skeleton, Polymastodon (Cope) of the Lower Eocene, is as specialized as the most specialized recent forms. The dentition of the Juras- sic forms, Plagiaulax, etc., is quite specialized also, but not more so than that of the kangaroos. The pre- molars are more specialized, the true molars less spe- cialized than in those animals. The history of Marsu- 143 PHYLOGENY. olssvunf Snozoviauy aNa0q aNaD0aN Bluopojordiq eyerdnsiey, eyao BeUap_ eraydoiryy epodojfouy saal[ eIOAIUIES eByUOpoalD eayyielApuos BIUOPOTLL eyeula1oU0y eNgopojoid Ajo eyerdnsieyy fwoysoyj IO eloanvasuy eydiowodoiyjuy veplooeidFy —eipros qolg vi VLIVTASINONG VLVINONQ Byenssaqnyyiny ‘Ww ea0R}97 epoddiquy Brae | quaeidiq VIVTLLAW 144 PRIMARY FACTORS OF ORGANIC EVOLUTION, pialia indicates that the primitive types were all in- sectivorous, and possessed a larger number of molars than any of the recent forms. The latter have then followed the same. course as the placentals in the re- duction of the number of teeth and specialization of those that remain. Coming to the Monodelphia, the increase in the size and complication of the brain, both of the cere- bellum and the hemispheres, is a remarkable evidence of advance. But one retrogressive line in this respect is known, viz., that of the order Amblypoda,! where the brain has become relatively smaller with the pas- sage of time. The successive changes in the structure of the feet are all in one direction, viz., in the reduc- tion of the number of the toes, the elevation of the heel, and the creation of tongue and groove joints where plain surfaces had previously existed. The diminution in the number of toes might be regarded as a degeneracy, but the loss is accompanied by a pro- portional gain in the size of the toes that remain. In every respect the progressive change in the feet is an advance. In the carpus and tarsus we have a gradual extension of the second row of bones on the first, to the inner side. In the highest and latest orders this pro- cess is most complete, and, as it results in a more perfect mechanical arrangement, the change is clearly an advance. The same progressive improvement is seen in the development of distinct facets in the cubito- carpal articulation, and of a tongue and groove (‘‘troch- lear crest”) in the elbow-joint. In the vertebre the development of the interlocking zygapophysial articu- lations is a clear advance. Progress is generally noticeable in the dental struc- 1See Naturalist, Jan,, 1885, p. 55. PHYLOGENY. 145 tures; for, the earliest dentitions are the most simple, and the later the more complex. Some of the types retain the primitive tritubercular molars, as the Cente- tide, shrews, and some lemurs, and most Carnivora (above), but the quadritubercular and its derivative forms are by far the most common type in the recent fauna. The forms that produced the complicated mod- ifications in the Proboscidia and Diplarthra appeared latest in time, and the most complex genera, Elephas, Bos, and Equus, the latest of all. The extreme sec- torial modifications of the tritubercular type, as seen in the Hyznide and the Felide, are the latest of their line also. Some cases of degeneracy are, however, apparent in the monodelphous Mammalia. The loss of pelvis and posterior limbs in the two mutilate orders is clearly a degenerate character, since there can be no doubt that they have descended from forms with those parts of the skeleton present. The reduction of flexibility seen in the limbs of the Sirenia and the loss of this charac- ter in the fore limbs of the Cetacea are features of de- generacy for the same reason. The teeth in both or- ders have undergone degenerate evolution ; in the later and existing forms of the Cetacea even to extinction. The Edentata have undergone degeneration. This is chiefly apparent in the teeth, which are deprived of enamel, and which are wanting from the premaxillary bone. A suborder of the Bunotheria, the Tillodonta of the Lower Eocene period, display a great reduc- tion of enamel on the molar teeth, so that in much- worn examples it appears to be wanting. Its place is taken by an extensive coat of cementum, as is seen in Edentata, and the roots of the teeth are often undi- vided as in that order. 146 PRIMARY FACTORS OF ORGANIC EVOLUTION. Local or sporadic cases of degenerate loss of parts are seen in various parts of the mammalian series, such as toothless Mammalia wherever they occur. Such are cases where the teeth become extremely simple, as in the honey-eating masupial Tarsipes, the carnivore Proteles, the pteropod bats, and the aye-aye. Also where teeth are lost from the series, as in the canine genus Dysodus, and in man. The loss of the hallux and pollex without corresponding gain, in various gen- era, may be regarded in the same light. In conclusion, the progressive may be compared with the retrogressive evolution of the Vertebrata, as follows: In the earlier periods and with the lower forms, retrogressive evolution predominated. In the higher classes progressive evolution has predominated. When we consider the history of the first class of ver- tebrates, the Tunicata, in this respect, and compare it with that of the last class, the Mammalia, the con- trast is very great. h. The Phylogeny of the Horse. As an example of special phylogeny I select that ‘of the horse, because it is the most completely repre- sented by specimens in our museums. I have already pointed out that the alternate type of carpus and tarsus of the Diplarthra has been derived from the linear of the Taxeopoda by a displacement inwards of the bones of their second rows. In the pos- terior foot this has changed the convex surface of the head of the astragalus into a bifacetted face. Thus was the condylarthrous astragalus modified into that of the Diplarthra. At the beginning of the line of the horses we find the condylarthrous genus of the Wasatch Eocene, Phenacodus Cope, to differ in this PHYLOGENY. 147 way from the perissodactylous genus Hyracotherium Owen. Phenacodontide are indicated as the ances- tors of all the Ungulata by their character as ‘‘buno- dont pentadactyle plantigrades,”’ characters in which they agree with the ancestors of all placental mam- mals. That they are not the ancestors of all the latter is shown by the fact that their molar type is quadritu- bercular; but one has to go backwards but a short distance in time to the Puerco epoch, to find their tri- tubercular ancestors. Between these and Phenaco- dus, comes the quadritubercular genus Euprotogonia Cope, of the Puerco, which has simpler premolar teeth. Between Phenacodus and Hyracotherium there is room for two or more genera with fully facetted car- pals and tarsals, longer feet, and a rudimental first toe on the anterior foot, and first and fifth toes on the hind foot. In Hyracotherium these digits have disappeared. Further, in Hyracotherium the internal cusps of the molars are more or less connected with the external by low and indistinct ridges, which in the superior molars include the small intermediate tubercles or conules. Thus is the lophodont dentition foreshad- owed. Hyracotherium was a contemporary of Phe- nacodus and continued later in Eocene time. Some of its forms developed an increased complexity of the last premolars in both jaws, forming the genus Pliolo- _phus, and foreshadowing the development of molar- like premolars, which is so characteristic of the later members of the horse line. In the genus Epihippus Marsh, of one epoch later in time (the Uinta), two such premolars are developed in each jaw. We have seen very short interspaces next the canine teeth in Phenacodus, and these have become longer in Hyra- cotherium and Epihippus. 148 PRIMARY FACTORS OF OKGANIC EVOLUTION. With the opening of the Neocene age, we have the descendant of Epihippus in Mesohippus Marsh, which differs from its predecessor as follows. There are but three toes on all the feet; three premolars resemble the true molars ; the crests which connect the internal pair of cusps with the external in both jaws, are much more elevated, and soon form on wearing a part of the pattern of the crown. Hyracotherium already walked on the ends of its toes, and the feet of Mesohippus continue the character. The crowns of all the molars are short like those of its ancestors. In the Middle Neocene formations we have the genus Anchitherium Kaup, where the incisor teeth show the addition of a ridge, or cingulum round the inner side, which bounds a cup; forming the cupped incisors so characteristic of the horses. The species have been all the while growing gradually larger. Towards the end of Neocene time important pro- gress was made. In the Loup Fork epoch the three toed horses were very numerous in species, but their lateral toes were all much shortened so that they did not reach the ground. The crests of the molar teeth were much stronger, and in the superior series the conules had assumed a greater importance, extending themselves posteriorly from the transverse crests, and showing crescentic sections, resembling those of the outer cusps, with which they are parallel. The an- terior conule extended so far posteriorly as to join the posterior one, resembling in this respect also the an- terior external cusp. So the crown came to have six modified cusps of which the two inner are the smallest and remain unconnected with each other. The crowns of the moiars vary in length in these later three-toed horses. Some are short like the Anchitheriums, and PHYLOGENY. 149 others are longer, approaching the true horses. In the valleys between these high cusps cement is deposited, as in the true horses and other mammals with long- crowned molars. There are two types of these later three-toed horses. In one the posterior inner cusp is not joined to the conule by a transverse crest (genus Hippotherium Kaup), or it is so joined (genus Pro- thippus Leidy). Plistocene times witnessed the perfection of the horse line. The lateral toes dwindled into splints concealed beneath the skin. The crowns of the molar teeth became very long, and in the upper jaw the inner posterior tubercle, now a column, joined the adjacent conule, and became extended very much in fore and aft diameter. The small anterior premolar disap- peared, and the canines became the mark of the male sex only, in general. The lower molars acquire some additional complications, and the feet are longer than in any of its ancestors. The genus Equus L. is fin- ished, and remains a permanent member of the human epoch, from which its only relatives, the rhinoceros and the tapir, are gradually disappearing. This history may be duplicated in ‘manner and mode, by the lines of the camels, the dogs and bears, the cats, the beaver, etc. Examination of all these genealogical lines reveals a certain definiteness of end and directness of approach. We discover no accessions of characters which are afterwards lost, as would naturally occur as a result of undirected variation. Nor do we discover anything like the appearance of sports along the line, the word sport being used in the sense of a variation widely di- vergent from its immediate ancestor. On the con- trary, the more thorough becomes our knowledge of 150 PRIMARY FACTORS OF ORGANIC EVOLUTION. the series, the more evident does it become that pro- gressive evolution has advanced by minute increments along a definite line, and that variations off this line have not exerted an appreciable influence on the re- sult. 2. The Phylogeny of Man. In man the feet retain the pentadactyle plantigrade type with scarcely grooved astragalo-tibial articula- tion, which characterizes the Mammalia of the Puerco epoch, and most of those of the Lower Eocene.!_ His dentition is not lophodont, but is simply bunodont, like that of the Phenacodontidez of the Lower Eocene. It is only in the structure of the brain and the reproduc- tive system that man shows an advance over the Eocene type. In the former he greatly excels any mammal that has appeared since; a superiority already apparent in one of his early ancestors, the anaptamorphous lemur of the Lower Eocene. In the reproductive system he is about on a par with the higher Artiodactyla, although the male, in the persistent union of the genital and urin- ary efferent ducts is not so much specialized as some of the latter, where they are distinct. It is an interesting fact that man displays in his dentition strong tenden- cies to a greater specialization by simplification beyond the ordinary quadrumanous type, by reduction in the number of the true molarsand incisors. Thus the M.3is not unfrequently absent in the highest races, and some families display a rudimental condition and absence of the 1.2. : Much importance attaches to the composition of the molar dentition. Many years ago, Owen? called 1This fact was first pointed out by myself in the Penn Monthly Magazine, 1875; see Origin of the Fittest, p. 268. 2Odontography, 1840-5, p. 454. PHYLOGENY. I51 attention to the fact that in the dark races the roots of the last superior molar are distinct from each other, while in the Indo-Europeans they are known to be Fig. 38.—Fig «, skull of Anaptomorphus homunculus Cope, natural size. Fig. 4, same, oblique view, displaying the large cerebral hemispheres. Fig.c, superior view of skull, natural size. Fig. @, inferior view, three-halves natural size. Lower figs. 2, 4, and ¢, left branch of lower jaw of Anaptomorphus emu- dus Cope, twice natural size; a, from left side; 4, inner side; c, from above. more or less fused together. These now well-known characteristics of human dentition constitute one of the examples of transition from a simian to a human 152 PRIMARY FACTORS OF ORGANIC EVOLUTION. type. Ihave pointed out a corresponding modifica- tion in the structure of the crown of the superior true molars, viz.: the transition from a quadritubercular to a tritubercular structure in passing from the lower to the higher races. As this point has some interesting im- plications in the earlier phylogeny of man, and as its value has been disputed, I give it a little attention. The facts of the case are as follows: I have dem- onstrated! the fact that all forms of dentition exhibited by the eutherian mammals have been derived from a primitive tritubercular type. Professor Osborn says that he expects to be able to do the same for the multi- tuberculate (? Prototherian) dentition. I have also shown that man exhibits a tendency to revert from his primitive quadritubercular molar to this tritubercular type.? As to the significance of these facts, I have expressed the view that this acquisition of a tritubercu- lar molar is a reversion to the lemurine type. This conclusion is necessary because the lemurs are the last of the families in the line of the ancestry of man which present this dentition. The monkeys and anthropoid apes are all quadritubercular, except a few limited col- lateral branches of the former, which still retain the lemurine type. There are also a few collateral types of lemurs which have acquired one or more quadri- tubercular niolars, but they are not typical. In many tritubercular mammals, a precocious form or two can be found, which has acquired the fourth tubercle. But the further back we go in time, the fewer they become, until, in the Puerco fauna, of eighty-two species of 1 Proceeds. Amer. Philos, Soc., Dec., 1883; Origin of the Fittest, 1887, pp 245, 347, 359- 2American Journal of Morphology, \1., 1883, p. 7. PHYLOGENY. 153 eutherian mammals, but four have true quadritubercu- lar superior molars. ; I take this opportunity of saying, however, that re- version is not necessarily a result of heredity. It may be simply a retrogression on a line of advance already laid down. What influence lemurine heredity may have had in the case of man, it is not easy to know. But it must be borne in mind that various forms of degeneracy of molar teeth are possible other than the resumption of the tritubercular type, yet the normal reduction generally presented is just this lemurine and primitive eutherian condition. The simplicity of the elements involved, has something, but not everything, to do with this reversion. Dr. Paul Topinard has made an investigation ! of the characters of the crowns of the molars in man, and hasreached general conclusions identical with my own. He remarks (p. 665): ‘‘It is demonstrated, in conclu- sion, that the teeth of man are, at present, in process of transformation, and that in some future which is re- mote the inferior molars shall certainly be quadri- cuspid, and the superior molars tricuspid. It will be curious to have the statistics as to prehistoric man; unfortunately, their crania are rare, and their molars generally much worn.” In the details of his examina- tion, there are some divergencies from my results. Thus he finds the quadritubercular second and third superior molar relatively of more frequent occurrence in Europeans than I did. But the absence of Europeo- Americans from his tables reduces the percentage of trituberculars in the Indo-Europeans. He makes no report of Esquimaux. Had he observed this type, he would have found a higher per cent. of tritubercular 2L' Anthropologie, 1892, p. 641 {Nov., Dec.). 154 PRIMARY FACTORS OF ORGANIC EVOLUTION. upper molars than in any race that he has recorded. He confirms my conclusion as to the high percentage of quadritubercular superior molars in the Malays, Polynesians, and Melanesians. The relation of this fact to phylogeny is to confirm Haeckel’s hypothesis of the lemurine ancestry of man. I have advanced the further hypothesis that the An- thropomorpha (which include man and the anthropoid apes) have been derived directly from the lemurs, with- ‘out passing through the monkeys proper. This close association of man with the apes, is based on various considerations. One of them is that the skeleton of the anthropoid apes more nearly resembles that of man in the most important respects than it does that of the monkeys. This is especially true of the verte- bral column, where the anapophyses are wanting in the Anthropomorpha (insignificant rudiments remain- ing on one or two vertebre, as pointed out by Mivart), while they are well developed in the monkeys and lemurs. The molar teeth of the apes and man resem- ble each other more than either do those of the mon- keys, since they lack the crests which connect the cusps, which are general in the latter. The frequent presence of the tritubercular molar in man suggests the superior claim of the lemurs over the monkeys to the position of ancestor. Another signifi- cant fact pointing in the same direction is the existence of large-brained lemurs with a very anthropoid denti- tion (Anaptomorphidz) in our Eocene beds, which have the dental formula of man and the Old World monkeys andapes. This resemblance is very remarkable, much exceeding that lately observed by Ameghino in certain extinct forms of monkeys in Patagonia, which appear to be ancestors of the existing South American mon- PHYLOGENY. 155 keys (Cebidz), and possibly of the Old World monkeys also. The superior molars of the Anaptomorphus are tritubercular, while the premolars, canines, and in- cisors are essentially anthropomorphous, and rather human than simian. Anaptomorphus is probably at the same time the ancestor of the Malaysian lemurine genus Tarsius, and M. Topinard remarks that Tarsius has as good claims to be regarded as ancestral to Homo as Anaptomorphus. But M. Topinard must be aware that in the existing genus the character of the canine and incisive dentition is very unlike that of the Anaptomorphus and Homo. It is specialized in a different direction. The dentition of Anaptomorphus being so generalized as compared with Tarsius, I sus- pect that its skeleton will be found to present corre- sponding characters. Of course, if it be found here- after to have the foot structure of Tarsius (which I do not anticipate), it cannot be included in the ancestry of the Anthropomorpha. It must be further observed that the ancestral line of the Anthropomorpha cannot be traced through any existing type of Lemuride, but through the extinct forms of the Eocene period.! This is on account of the peculiar specialization of the inferior canines, which are incisor-like, and because of the peculiar character of the incisors themselves, in the modern lemurs in the restricted sense. But we have numerous lemurine types of the Eocene of both America and Europe which satisfy the conditions exactly, so far as the den- tition is concerned. These are mostly referable to the family Adapide. Unfortunately, we do not know the entire skeletons 1On the Primitive Types of the Orders of the Mammalia Educabilia, 1873, p. 8. 156 PRIMARY FACTORS OF ORGANIC EVOLUTION. of these Eocene lemurs, but as far as we have them (genera Tomitherium and Adapis) they are monkey- like. But we have what is almost as useful, the skel- eton of their Eocene and Puerco ancestors, the Con- dylarthra. I long since pointed out that the latter order (not the genus Phenacodus, as Lydekker has ae pe Fig. 39.—Tomitherium rostratum Cope, one of the Adapide, mandible, natural size; a, from left side; 4, from above. Original, from Report U. S. Geol. Survey Terrs., Vol. III. represented me as saying) must be the ancestors of the lemurs, basing my views expressly on the general structure of the Phenacodus, Periptychus, and Menis- cotherium. The structure of the ungual phalanges of Periptychus is very significant, and even more so is that in Meniscotherium, as recently shown by Marsh, PHYLOGENY. 157 who adopts (without credit) my hypothesis of lemurine affinities of the Condylarthra (which he renames the Mesodactyla). From Condylarthra back to Creodonta is an easy transition, and I have always assumed that the Creodonta were derived from generalized polypro- Fig. 40.— Tomitherzum ros- tratum Cope, fore arm, five- sixths natural size, Original, 6, ulna; ¢, radius. todont Mursupialia. This view has been entirely confirmed by the recent discoveries of Ame- ghino in Patagonia, where he. has found forms whose remains may be referred with equal pro- priety to the one group or the other. M. Topinard has been rather hasty in reaching the marsupial ancestry in suppos- ing that Phenacodus belongs to that order. All the evidence shows that Phenacodus is a generalized ungulate placental. To return to the more im- mediate ancestry of man. I have expressed,! and now main- tain as a working hypothesis, that all the Anthropomorpha were descended from the Eo cene lemuroids. In my sys- tem? the Anthropomorpha in- cludes the two families Homi- nide and Simiide. The sole difference between these families is seen in the struc- ture of the posterior foot; the Simiide having the lAmerican Naturalist, 1885, p. 467. 2 Origin of the Fittest, 1887, p. 346, from American Naturalist, 1885, p. 344, where the classification of the Taxeopoda should be in a foot-note; oc. cit., 1889, October. 158 PRIMARY FACTORS OF ORGANIC EVOLUTION. hallux opposable, while in the Hominide the hallux is not opposable. This is not a strong character, since it de- pends ona slight differ- ence in the form of the entocuneiform bone. In some vertebrates, as the tree-frogs, the same and similar characters (ge- nus Phyllomedusa) are not regarded as of fam- ily value. It is then highly probable that Homo is’ descended from some form of the Anthropomorpha now extinct, and probably unknown at present, al- though we do not yet know all the charac- ters of some extinct supposed Simiide, of which fragments only remain to us. It cannot now be determined whether man and the Simiidze were both de- scended from a genus with opposable hallux, or without opposable hallux, or whether from a genus presenting an Fig. 41.— Tomitherium rostratum Cope, five-sixths natural size; a, ilium; 4, femur. Original. PHYLOGENY. 159 intermediate character in this respect. This genus was, in arty case, distinct from either of the two existing genera of Simiide, Simia and Hylobates, since these present varied combinations of anthropoid resem- blances and differences, of generic and specific value. Professor Virchow in a late address! has thrown down the gage to the evolutionary anthropologists by asserting that ‘scientific anthropology begins with living races,” adding ‘that the first step in the con- struction of the doctrine of transformism will be the explanation of the way the human races have been formed,” etc. But the only way of solving the latter problem will be by the discovery of the ancestral races, which are extinct. The ad captandum remarks of the learned professor as to deriving an Aryan from a Negro, etc., remind one of the criticisms directed at the doctrine of evolution when it was first presented to the public, as to a horse never producing a cow, etc. It is well known to Professor Virchow that hu- man races present greater or less approximations to the simian type in various respects. Such are the flat coéssified nasal bones of the Bushmen; the qua- dritubercular molars of the Polynesians; the flat ilia and prognathous jaws of the Negro; the flat shin- bones of various races; the divergent hallux of some aborigines of farther India, etc. Professor Virchow states that the Neanderthal man is a diseased subject, but the disease has evidently not destroyed his race characters; and in his address he ignores the important and well-authenticated discovery of the man and wo- man of Spy. These observations are reinforced by recent discovery of a similar man by DuBois at Trinil 1 Popular Science Monthly, January, 1893, p. 373, translated. 160 PRIMARY FACTORS OF ORGANIC EVOLUTION. in the island of Java. To these ancient people I will now devote some space. What had been long suspected is now established, Fig. 42,_Megaladapis madagascariensis Forsyth Major, 4% natural size; lemuroid from Plis- tocene bed of Madagascar, with tritubercular superior molars. From Forsyth Major. through the discovery and descriptions of Messrs. Fraipont and Lohest of Liége; viz. that there dwelt in Europe during Paleolithic times a race of men which PHYLOGENY. 162 possessed a greater number of simioid characteristics than any which has been discovered elsewhere. ‘The important discovery in the grotto of Spy of two skele- tons, almost complete, served to unify knowledge of this race, which had previously rested on isolated frag- ments only. These skeletons proved what had been previously only surmised, that the lower jaws of Nau- lette, and of Shipka, and probably the skeleton of Neanderthal, belong to one and the same race. The Fig. 43.—Skull of the man of Spy. From Prof. G. F. Wright’s Man and the Glacial Period. From a photograph. simian characters of these parts of the skeleton are well known. These are the enormous superciliary ridges and glabella; the retreating frontal region; the thickness of the cranial wall; the massive mandibular ramus with rudimentary chin, and the large size of the posterior molars. Messrs. Fraipont and Lohest have added other characters to these, viz.: the tibia shorter than in any other known human race; the sigmoid flexure of the femur; the divergent curvature of the bones of the fore-arm, and most important, a very 162 PRIMARY FACTORS OF ORGANIC EVOLUTION. peculiar form of the posterior face of the mandibular symphysis, already pointed out by Topinard in the jaw From Fraipont and Lohest. Fig. 44.—Outlines of calvaria; of the Neanderthal man in solid line; of the Spy man No.1 in dots, and of the Spy man No. 2 in broken line. of Naulette. On these characters the following re- marks may be made.! I. The prominent superciliary crests, which are 1 Archives Belges de Biologie, V11., 1886, p. 731, Gand. PHYLOGENY. 163 characteristic of the Neanderthal race. No existing race presents such a development, neither the Papu- ans, Australians, nor Negroes of any race. But we find the superciliary crests and underlying sinuses identical in adult female orangs and chimpanzees and young male gorillas. Inthe female chimpanzees the crests are almost inferior in size to those of the man of Spy. II. The retreating forehead of the two crania of Spy is not found in any existing human race, while it is typical of that of Neanderthal. It is characteristic of female orangs and gorillas and of the young males of both species, and of adult males and female chim- panzees. It appears in existing men in rare and iso- lated cases; [probably as survivals]. III. The prominent transverse superior semicircu- lar crest of the occipital bone is found in existing races among the Fellahs of Africa and the Nigritos. It is characteristic of the Neanderthal skulls, and presents exactly the same characters as the young male and female orang and gorilla and young male and adult female of the chimpanzee. IV. No human race presents the characters of the lower jaw exhibited by those of Spy, Naulette, and Shipka. In this part of their osteology the anthro- poids depart widely from man, the most conspicuous point in the latter being the presence of a chin. Ac- cordingly, the angle formed by the anterior face of the symphysis with the inferior border of the horizontal ramus, is less than a right angle in man, and much more than a right angle in the anthropoids. According to Topinard, this angle in fifteen Parisians is 71.4°; in fifteen African Negroes, 82.2°; in fifteen Neocaledo- nians, 83.9°; in the jaw of Naulette, 94°. In the best 164 PRIMARY FACTORS OF ORGANIC EVOLUTION. preserved jaw of Spy the angle is 107°, if measured from the inferior symphysial border, or go° if meas- A B :) c.. B M~ Cc D E Fig. 45.—Vertical sections of symphysis mandibuli of gorilla (Fig. 4), and orang (Fig. 2), of chimpanzee (Fig. C), of Spy man No.1 (Fig. D), and Spy man No. 2 (Fig. £). From Fraipont and Lohest. ured from the inferior border of the ramus. There is no chin in the jaw of the Spy race, and the large angle approaches without nearly equaling that of the PHYLOGENY. 165 anthropoids. But the posterior face of the symphysis presents the most remarkable peculiarity. In the symphysis of the apes (Fig. 54, 4, B, C) the posterior border is a continuous slope from the alveolar border to the inferior margin, interrupted by a slight concav- ity below the middle. In the human jaw this line slopes backward to near the middle, where are situated the small tuberosities for the insertion of the genio- glossal muscles. (B in the accompanying figures.) The surface then slopes rapidly forward to pass into the narrow inferior border of the chin (Fig. 46, 7, G). In the jaws Naulette and Spy the structure is exactly in- termediate between the two, and quite different from both (Fig. 45, D, £). It commences above with a pos- terior slope similar to that of the apes, exhibiting what is called by Topinard ‘internal prognathism,” as it appears in the lower human races. The surface then descends abruptly, forming a vertical concavity, which is bounded a considerable distance below by another protuberance, the insertion of the genioglossal muscles. This concavity is not present in the human symphysis, while it is less developed in the simian. The surface then slopes forward, as in the human sym- physis, but this portion is shorter than in human jaws generally. It is represented by a convex face in the simian jaw. ‘This character, taken in connection with the others cited, goes a long way toward justifying the separation of the Neanderthal race as a distant species, as has been done by some author under the name of Homo neanderthalensis. This name is objectionable but must be retained. To these observations Messrs. Fraipont and Lohest add the following. V. The curvature of the ulna and radius, which 166 PRIMARY FACTORS OF ORGANIC EVOLUTION. produces a wide interosseous space, is not found in any human race, but is common to the apes. On the con- trary, the shortness of these bones is entirely human. VI. The anterior convexity of the femur, with its round section, is only found among living races among the Nigritos of the Philippine Islands. It is seen ina less degree in femora of Neolithic men, and occasional instances are seen among existing Europeans. It is the normal condition in the apes. VII. The tibia is shorter in its relation to the femur than in any human race, and is more robust than in OA fF G Fig. 46.—Sections of symphysis mandibuli of modern Litgois (Fig. /) and of an ancient Parisian (Fig. G). From Fraipont and Lohest. most of them. This character, with the oval section, while not identical with what is seen in the apes, forms an approximation to it. Messrs. Fraipont and Lohest have pointed out the general characters of the dentition of the man of Spy. They show that the molars increase in size posteriorly to the same extent that they do in the apes, which is the reverse of what is usual in man, where they dimin- ish posteriorly, or, in a few lower races (Australians, etc.), remain equal. They show that the superior mo- lars are all quadritubercular, and that the internal root 4a 2a Fig. 47.—Molar teeth of man. 1-2, man and woman of Spy; 3, Maori; 4, Tahitian; 5-6, Fan, 7, Esquimau. 168 PRIMARY FACTORS OF ORGANIC EVOLUTION. is distinct in all of them. Through the kindness of M. Lohest I received casts and photographs of these teeth, and I give here figures of the former (Fig. 47), which are more satisfactory than those in the memoir from which J have already quoted so fully, where, in- deed, the grinding faces are not represented at all. The figures accompanying! show the large size of the last superior molar, which exceeds in its propor- tions those of the corresponding tooth in the chimpan- zee. The fourth tubercle, or hypocone, is especially large. In the male the crowns are more produced posteriorly than in man generally, and remind one of the character seen in the orang. The strong divergence of the internal root of the last molar is shown in No. 2a, and the corresponding character in a Maori and a Fan from tropical Africa is shown in Nos. 3 and 5a. The quadritubercular crown of the last superior molar of a Tahitian is shown in No. 4.4; and the roots, which are exceptionally fused nearly as much as in the typi- cal Indo-European, are shown in No. 4. Dr. Eugene Dubois of the Army of the Netherlands has recently published in Batavia, Java, in a brochure in quarto, an account of some bones of an interesting quadrumanous mammal allied to man, which were found in a sedimentary bed of material of volcanic ori- gin of probably Plistocene age, near a village called Trinil. The remains consist of a calvarium which in- cludes the supraorbital ridges and a part of the occi- put; a last superior upper molar, and a femur. The tooth was found close to the skull and belongs probably to the same individual as the latter, while the reference of the femur is more uncertain, as it was found some fifty feet distant. 1From The American Naturalist, April, 1893. PHYLOGENY. 169 The characters of the skull are closely similar to those of the men of Neanderthal and of Spy, but the walls are not so thick as those of the former, and more nearly resemble those of the latter. The frontal region is, therefore, much depressed, and it is also much con- stricted posterior to the postorbital borders. The su- tures are obliterated. Dr. Dubois states that the cranial capacity is just double that of the gorilla, and two-thirds that of the lowest normal of man, bridging the gap which has long separated the latter from the apes. Thus the capacity of the former is 500 cubic centime- tres, and the latter is 1500 cubic centimetres. In the Java man the capacity is 1000 cubiccentimetres. The last upper molar has widely divergent roots,.as in apes and inferior races of man, and the crown is large, with the cusps not clearly differentiated, showing a character commonly observed in the lower molars of the gorilla. The femur is long, straight, and entirely human. This discovery of Dr. Dubois adds to our knowledge of the physical characters of the Paleolithic man, and espe- cially to his geographical range. As regards the proper appellation of this being, Dr. Dubois is not happy. He proposes for him a new genus Pithecanthropus (after Haeckel), and even anew family, Pithecanthropide, without having shown that he is not a member of the genus Homo. It is not certain that he is not an individual of the species Homo neanderthalensis. is cranial capacity is less, it is true, than that of the man of Spy, but Virchow has pointed out that some of the Nigritos possess a remark- ably small cranial capacity, as little as 950 cubic centi- metres, and an inhabitant of New Britain only 860 cubic centimetres, a capacity even smaller than that of the man of Trinil. Until we learn the characters of 170 PRIMARY FACTORS OF ORGANIC EVOLUTION. the lower jaw of the latter we shall be in doubt as to whether this individual pertains to the Homo sapiens or the Homo neanderthalensis. The characters of the dentition, cranium, and limbs which have been observed in the Paleolithic man, are not without parallel in existing races, though the char- acters do not generally occur together in the latter. The supposition that all the Paleolithic men so far found are all pathological subjects is not a probable solution of the question, although this type was no doubt subject to pathological conditions such as have been found in the leg-bones of the men of Neanderthal and Trinil. The characters of the symphysis of the lower jaw are quite sufficient to separate the Neander- thal man as a distinct species of the genus Homo.! This character is not pathological but it is zodlogical, and places that species between Homo sapiens and the apes. In conclusion, it may be observed that we have in the Homo neanderthalensis a greater number of simian characteristics than exist in any of the known races of the Homo sapiens, although, so far as known, he be- longs to the genus Homo. ‘The posterior foot, so far as preserved, indicates this to be the case. The foot- character, which distinguishes the genera Homo and Simia still remains. There is still, to use the language of Fraipont and Lohest, ‘‘an abyss’’ between the man of Spy and the highest ape ; though, from a zoélogical point of view, it is not a wide one. The flints which were discovered in the stratum of cave deposit containing the human remains, are of the Paleolithic type known as Mousterien in France, which 1 This view was first insisted on in an article on the Genealogy of Manin the American Naturalist, 1893, p. 331. PHYLOGENY. I7I are of later origin than the Chelléen or older Paleo- lithic. The older Paleolithic man is not yet known. It is interesting to observe that these flints (Mouste- rien) are of the same form as the obsidian implements which I collected at Fossil Lake, in Oregon, with the bones of extinct Ilamas, horses, elephants, sloth, etc. The animals which accompanied the man of Spy are, Celodonta antiguitatis (wooly rhinocetos), Eguus ca- ballus, Cervus elaphus, Cervus tarandus, Bos primigenius, Elephas primigenius, Ursus speleus, Meles taxus, Hyena spelea ; five extinct and four existing species. As the evidence now stands, the most primitive and simian of human races inhabited the Old World. No trace of the Homo neanderthalensis has been found in America. As, however, Paleolithic implements are found in all continents, we may anticipate that this or some similar species of man will be discovered there also. The genealogy of man may be then represented as follows: CLASS & BRANCH ORDER AND FAMILY. GEOL. SYSTEM Hominide Plistocene Simiidze Neocene Adapidz Eocene Mammalia Condylarthra Cretaceous Creodonta Marsupialia polyprotodontia Jurassic Monotremata Triassic Reptilia Theromora Carbonic Batrachia Batrachia Stegocephali Carbonic Pisces Teleostomi Rhipidopterygia Elasmobranchii Ichthyotomi Agnatha q Cephalochorda _Leptocardii Vermes i Ceelenterata 1. Protozoa 1 1 Subordinate type not specified. 172 PRIMARY FACTORS OF ORGANIC EVOLUTION. 3. THE LAW OF THE UNSPECIALIZED. The facts cited in the preceding parts of this chap- ter show that the phylogenetic lines have not been continuous, but that they may be represented by a system of dichotomy. In other words, the point of departure of the progressive lines of one period of time has not been from the terminal types of the lines of preceding ages, but from points farther back in the series. Thus it is not the highly developed or spe- cialized ‘plants which have given origin to the animal kingdom, but the lowest forms or Protophyta, which are not distinguishable from the Protozoa. Among animals it is not the specialized Arthropoda or Mol- lusca which present the closest affiliations with the Vertebrata, but the simple Vermes or Tunicata, from which the origin of the latter may be traced. In the Vertebrata it is not the higher fishes (Actinopterygia) which offer the closest points of affinity to the succeed- ing batrachian class, but that more generalized type of the Devonic period, the Rhipidopterygia, which probably occupies that position. The modern types of Batrachia (Urodela, Salientia) have plainly not fur- nished the starting-point for the reptiles, but the an- cient order of the Stegocephali, which are also fish- like, is evidently their source. The Reptilia of the Permian present us with types with fish-like vertebre (Cotylosauria, Pelycosauria), from which the class Mammalia may be distinctly traced. The later reptiles diverged farther and farther from the mammalian type with the advance of geologic time. The same prin- ciple has been found to be true in tracing the history PHYLOGENY. 173 of the subdivisions of the great classes, in the preceding section. Agassiz and Dana pointed out this fact in taxon- omy, and I expressed it as an evolutionary law under the name of the ‘‘ Doctrine of the Unspecialized.” This describes the fact that the highly developed, or specialized types of one geologic period have not been the parents of the types of succeeding periods, but that the descent has been derived from the less spe- cialized of preceding ages. No better example of this law can be found than man himself, who preserves in his general structure the type that was prevalent dur- ing the Eocene period, adding thereto his superior brain-structure. The validity of this law is due to the fact that the specialized types of all periods have been generally in- capable of adaptation to the changed conditions which characterized the advent of new periods. Changes of climate and food consequent on disturbances of the earth’s crust have rendered existence impossible to many plants and animals, and have rendered life pre- carious to others. Such changes have been often espe- cially severe in their effects on species of large size, which required food in large quantities. The results have been degeneracy or extinction. On the other hand plants and animals of unspecialized habits have survived. For instance, plants not especially restricted to definite soils, temperatures, or degrees of humidity, would survive changes in these respects better than those that have been so restricted. Animals of om- nivorous food-habits would survive where those which required special foods, would die. Species of small size would survive a scarcity of food, while large ones would perish. It is true, as observed by Marsh, that 174 PRIM1RY FACTORS OF ORGANIC EVOLUTION. the lines of descent of Mammalia have originated or been continued through forms of small size. The same is true of all other Vertebrata. It is not to be inferred from the reality of the law of ‘‘the unspecialized” that each period has been de- pendent on the simplest of preceding forms of life for its population. Definite progress has been made, and highly specialized characters have been gradually de- veloped, and have passed successfully through the vicissitudes of geologic revolutions. But these have not been the most specialized of their respective ages. They have presented a combination of effective struc- ture with plasticity, which has enabled them to adapt themselves to changed conditions. In a large number of cases in each geologic age forms have been successful in the struggle for existence through the adoption of some mode of life parasitic on other living beings. Such habits reduce the struggle to a minimum, and the result has been always degen- eracy. In other cases it is to be supposed that ex- tremely favorable conditions of food, with absence of enemies, would have occurred, in which the struggle would have been almost nil. Degeneracy would fol- low this condition also. On the other hand, extreme severity of the struggle cannot have been favorable to propagation and survival, so that here also we have a probable cause of degeneracy. Degeneracy is a fact of evolution, as already remarked, and its character is that of an extreme specialization, which has been, like an overperfection of structure, unfavorable to sur- vival. In general, then, it has been the ‘‘golden mean” of character which has presented the most favorable condition of survival, in the long run. CHAPTER III.—PARALLELISM. T IS now generally recognized that the successive types of organic beings present characters which are traversed in the embryonic life of those which at- tain the greatest complexity of development, and which occupy the highest places in the scale of life. This fact was observed by the early embryologists, as Von Baer and Agassiz, who did not admit its bearing on the doctrine of evolution. But Darwin and Spencer understood its significance, and Haeckel, Hyatt, and the writer applied it directly to the explanation of phy- logeny. At the present time one of the chief aims of the science of embryology is to discover the record of the history of the past, recapitulated in the stages of embryonic life, and to unravel the phylogenesis of plants and animals by this method. The utility of these researches is attested by the results which they have attained, though for obvious reasons, these are not as definite and conclusive as those which are de- rived from paleontology. The general conclusion is however justified, i. e., that the records of embryology and paleontology are closely similar, and that any dis- cordance between them may be explained on compre- hensible principles. 176 PRIMARY FACTORS OF ORGANIC EVOLUTION, A number of illustrations of the parallelism be- tween taxonomy, ontogeny, and phylogeny may now be given. 1. PARALLELISM IN THE BRACHIOPODA. For the following abstract I am indebted to Mr. C. E. Beecher of New Haven, whose excellent work in this field is well known. The parallelism between the ontogeny and phy- logeny in the Brachiopoda has been worked out in numerous instances.! To illustrate these, some more or less familiar genera may be taken as characteristic ex- amples. Lingula has been shown by Hall and Clarke (Pai. New York, Vol. VIII., 1892) to have had its inception in the Ordovician. In the ontogeny of both recent and fossil forms, the first shelled stage has a straight hinge line, nearly equal in length to the width of the shell. This stage may be correlated with the more ancient genus Paterina from the lowest Cambrian. Subsequent growth produces a form resembling Obolella, a Cam- brian and Ordovician genus. Then the linguloid type of structure appears at an adolescent period, and is completed at maturity. Thus, Lingula has ontogenetic stages corresponding to (1) Paterina, (2) Obolella, and (3) Lingula, of which the first two occur as adult forms in geological formations older than any known Lin- gula. Paterina represents the radicle of the brachiopods. 1C. E. Beecher, ‘‘ Development of the Brachiopoda,” Part I., Introduc- tion, Amer, Journ, Scz., Vol, XLI., April, 1891; ‘‘ Development of the Brachio- poda,” Part II., Classification of the Stages of Growth and Decline, Amer. Jour, Scz., Vol. XLIV., August, 1892; ‘‘ Development of Bilobites,’’ Aver. Jour, Sc7., Vol. XLIL., July, 1891; ‘‘ Revision of the Families of Loop-bearing Brachiopoda,”’ 7raus. Conn, Acad, Sci., Vol. 1X., May, 1893. PARALLELISM. 177 It shows no separate stages of growth in the shell, is found in the oldest fossiliferous rocks, and corresponds to the embryonic shelled condition (protegulum) of other brachiopods. The genus Orbiculoidea of the Discinide first ap- pears in the Ordovician and continues through the Mesozoic. The early stages in the ontogeny of an in- dividual are as in Lingula, first a paterina stage, fol- lowed by an obolella stage. Then from the mechan- ical conditions of growth a Schizocrania-like stage follows, and completed growth results in Orbiculoidea. The elongate form of the shell in Lingula, as well as in many other genera, is determined by the length of the pedicle and freedom of motion. The discinoid or discoid of Orbiculoidea and Discinisca among the brachiopods, and Anomia among pelecypods, is deter- mined by the horizontal position of the valves, which are attached to an object of support by a more or less flexible, very short organ, a pedicle or byssus, without calcareous cementation. This mode of growth is char- acteristic of all the discinoid genera, but, as already shown, the early stages of Paleozoic Orbiculoidea have straight hinge lines and marginal beaks, and in the adult stages of the shell the beaks are usually subcen- tral and the growth holoperipheral. This adult disci- noid form, which originated and was acquired through the conditions of fixation of the animal, has been ac- celerated in the recent Discinisca, so that it appears in a free-swimming larval stage. Thus, a character ac- quired in adolescent and adult stages of Paleozoic spe- cies through the mechanical conditions of growth, ap- pears by acceleration in larval stages of later forms before the assumption of the condition of fixation which first produced this character. 178 PRIMARY FACTORS OF ORGANIC EVOLUTION. The two chief subfamilies of the Terebratellide undergo complicated series of metamorphoses in their brachial structure. Generic characters are based upon the form and disposition of the brachia and their sup- ports. The highest genera in one subfamily, which is austral in distribution, pass through stages correlated with the adult structure in the genera Gwynia, Cistella, Bouchardia, Megerlina, Magas, Magasell, and Terebra- tella, and reach their final development in Magellania and Neothyris. The higher genera in another subfam- ily, boreal in distribution, pass through metamorphoses correlated with the adult structures of Gwynia, Cistella, Platidia, Ismenia, Miihlfeldtia, Terebratalia, and Dal- lina. The first two stages in both subfamilies are re- lated in the same manner to Gwynia and Cistella. The subsequent stages are different except the last two, so that the Magellania structure is similar in all respects to the Dallina structure, and Terebratella is like Tere- bratalia. Therefore Magellania and Terebratella are respectively the exact morphological equivalent to, or are in exact parallelism with Dallina and Terebratalia. The stages of growth of the genera belonging to the two subfamilies Dallinine and Magellaniine are further correlated in the accompanying tables. The simplest genus Gwynia, as far as known, passes through no brachial metamorphoses, and has the same structure throughout the adolescent period, up to and including the mature condition. In the ontogeny of Cistella the gwyniform stage, through acceleration, has become a larval condition. In Platidia, the cistelliform structure is accelerated to the immature period, and in Ismenia (representing an ismeniform type of struc- ture in the higher genera), the gwyniform and cistelli- form stages are larval, and the platidiform represents MORPHOGENY FROM GWYNIA TO DALLINA. PERIODS STAGES STAGES STAGES STAGES STAGES STAGES STAGES Larval gwyniform? gwyniform gwyniform gwyniform gwyniform gwyniform gwyniform . cistelliform cistelliform cistelliform cistelliform Adolescent gwyuiform cistelliform cistelliform platidiform platidiform platidiform platidiform ismeniform ismeniform ismeniform miuhlfeldtiform miihlfeldtiform terebrataliform Mature Guwynia Cistella Platidia Ismenia Mihlfeldtia Terebratalia Dallina MORPHOGENY FROM GWYNIA TO MAGELLANIA. PERIODS STAGES STAGES STAGES STAGES STAGES STAGES STAGES STAGES Larval gwyniform|gwyniform |gwyniform | gwyniform gwyniform gwyniform gwyniform gwyniform cistelliform cistelliform cistelliform cistelliform cistelliform Adolescent | gwyniform | cistelliform |cistelliform | bouchardiform] bouchardiform] bouchardiform] bouchardiform| bouchardiform megerliniform | megerliniform | megerliniform | megerliniform magadiform |magadiform {| magadiform magaselliform | magaselliform terebratelliform Mature Guwyntia Cistella Bouchardia | Megerlina Magas Magasella Terebratella | Magellania 180 PRIMARY FACTORS OF ORGANIC EVOLUTION. an adolescent condition. Similar comparisons may be made in the other genera. Progressively through each series, the adult structure of any genus forms the last immature stage of the next higher, until the highest member in its ontogeny represents serially, in its stages of growth, all the adult structures, with the larval and immature stages of the simpler genera. It is evident that in the identification of specimens belonging to the Terebratellida, whether recent or fossil, the strict specific characters must be given first consideration. Species, therefore, must be based upon surface orna- ments, form, and color, within certain limits, and gen- era only upon structural features developed througha definite series of changes, the results of which are per- manent in individuals evidently fully adult. In each line of progression in the Terebratellide, the acceleration of the period of reproduction, by the influence of environment, threw off genera which did not go through the complete series of metamorphoses, but are otherwise fully adult, and even may show re- versional tendencies due to old age; so that nearly every stage passed through by the higher genera has a fixed representative in a lower genus. Moreover, the lower genera are not merely equivalent to, or in exact parallelism with the early stages of the higher, but they express a permanent type of structure, as far as these genera are concerned, and after reaching ma- turity do not show a tendency to attain higher phases of development, but thicken the shell and cardinal pro- cess, absorb the deltidial plates, and exhibit all the evidences of senility. (‘1ayoo0gq worg) ‘erueljeseW jo AuoZoysyd pue £ue8000 ‘ epodomprig ur wsr[aTereq—"gh “Sty i) Vamarecane Bouckar do, Re iy" tS, PN. @ Megerlina Cistella neopolitora Nigflevestens waokwoypboyy washyyeyoagaray, warhypsoboy wah yoboy, waohrurpaba yl wo rp soyonog warohrypysrg 182 PRIMARY FACTORS OF ORGANIC EVOLUTION, 2z. PARALLELISM IN THE CEPHALOPODA. Among Mollusca it is well known that the Cepha- lopoda form a number of series of remarkable regular- ity, the advance being, in the first place, in the com- plication of the folds of the external margins of the septa, and, in the second place, in the degree of invo- lution of one or both extremities of the shell to the spiral ; third, in the position of the siphon. Alpheus Hyatt, in an important essay on this sub- ject, points out that the less complex forms are in many cases identical with the undeveloped conditions of the more complex. He says: ‘‘ There is a direct connection between the position of a shell, in the com- pleted cycle of the life of this order, and its own de- velopment. Those shells occupying the extremes of the cycle” (in time), ‘‘the polar forms, being more embryonic than the intermediate forms. The first epoch of the order is especially the era of rounded, and, in the majority of the species, of unornamented shells with simple septa; the second is the era of or- namentation, and the septa are steadily complicating ; in the third the complication of the septa, the orna- mentation, and the number of species, about twice that of any other epoch, all combine to make it the zenith of development in the order ; the fourth is dis- tinguishable from all the preceding as the era of re- trogression in the form, and partially in the septa. ‘‘The four periods of the individual are similarly arranged, and have comparable characteristics. As \Memoirs of the Boston Society for Natural History, 1866, p. 193. Hyatt was followed by Wirtenberger in Ausland, 1873, who entirely confirmed his conclusions, PARALLELISM. 183 has been previously stated, the first is rounded and smooth, with simple septa; the second tuberculated, and the septa more complicated; the third was the only one in which the septa, form, and ornamentation simultaneously attained the climax of individual com- plication; the fourth, when amounting to anything more important than the loss of a few ornaments, was marked by a retrogression of the whorl to a more tabu- lar aspect, and by the partial degradation of the septa.” I am indebted to Professor Hyatt for the following more detailed account of the results of his researches in this interesting field. The evidence as to the na- ture of evolution derived from the Cephalopoda is more complete than that obtained from any other source. - ‘Every group of nautiloids passes through, during its evolution in time, either a part or the whole of a certain series of changes. These modifications con- sist: first, of a straight or nearly straight cone, ortho- ceran; second, a curved cone, cyrtoceran; third, a coiled cone, gyroceran, which does not come in con- tact at any point; fourth, a coiled-cone, nautilian, which does come in contact at the termination of the first vo- lution and then during further growth remains in about the same condition, all of the internal whorls being visible as ina flat coil of rope; fifth, a coiled cone, involute-nautilian, which ‘also comes in contact like the fourth but then the whorl growing with greater rapidity spreads internally, covering up more or less of the internal volutions sometimes to such an extent that even the centre is concealed from view. The ex- amples which I have myself seen of the fifth kind range from the Silurian to the Nautili of the existing fauna, some being present in every period, and of other kinds, the first, second, and third kinds die out gradually, 184 PRIMARY FACTORS OF ORGANIC EVOLUTION. diminishing in the Devonian and Carboniferous and ultimately ceasing their existence altogether in the Trias. The fourth ceases in the Cretaceous, and the fifth alone survives in the Tertiaries and is still living in the Mautilus umbilicatus and pompilus, and two other species. ‘«Wherever found, the young of shells of the fifth kind are at first orthoceran or cyrtoceran like the first and second kind, then gyroceran in curvature like the third class, and then they become more or less rapidly nautilian like the fourth class in succeeding stages. In Silurian, Devonian, and Carboniferous forms this suc- cession is so marked that about all of the young shells of the fifth class may be described as palingenetic, that is as cyrtoceran, gyroceran, nautilian, and then involute-nautilian in their individual or ontogenetic development. In the Trias, Jura, Cretaceous, Ter- tiary, and present, as the fifth class increases in num- bers, there is a decided tendency to shorten and su- persede the gyroceran or third stage and introduce the fourth kind or the tendency to spread by growth in- wards, at earlier stages. ‘‘The characteristics of. the sutures are correlative with these stages of development, and it may be said in a general way, that all other characteristics corre- late more or less when studied comparatively in dif- ferent series with the differences in the curvature and coiling of the whorls. - The curvature and amount of involution is therefore the most important single char- acteristic of the nautiloids,so far as the comparative study of change by evolution is concerned; whether the whole order be considered statistically as above, 1. e. with reference to the existence or non-existence of certain forms orthoceran, cyrtoceran, etc., or gen- PARALLELISM. 185 etically, i. e. with sole regard to the evolution of dis- tinct series which may be traced from their origin to their termination in time. ‘‘Of these last there are some in every period trace- able with more or less completeness by gradations of adults back to orthoceran or cyrtoceran ancestors. Of these series of adults, some pass through only the or- thoceran and cyrtoceran modifications, others have the orthoceran, cyrtoceran, gyroceran, and nautilian, but those having the latter and the nautilian-involute are of extreme rarity until the Carboniferous is reached. After this the nautilian shells begin to predominate in every series, ultimately becoming the sole representa- tives of genetic series. ‘«Such series are, of course, frequently so closely parallel that it is possible to follow them, and show they are distinct only by means of certain genetic char- acters, the apertures, the structure of the siphuncle, the sutures and septa, and sometimes, although very rarely, all of these internal characters may show dif- ferences peculiar to some one genetic series in which the regular gamut of forms is passed through in the usual succession. Neglect of the comparative study of the stages of development and decline, and of the obvious parallelisms between these and adults of an- cestral forms, have caused naturalists, notably Bar- rande, to make artificial classifications in which about all straight forms, with the exception of some in which the siphuncles were notably distinct to be classed as Orthoceras, most of second kind as Cyrtoceras, most of the third kind as Gyroceras, most of the fourth and fifth kinds as Nautilus. ; ‘“*To such authors the involute-nautilian forms of the Silurian and the existing fauna were considered to 186 PRIMARY FACTORS OF ORGANIC EVOLUTION. be only specifically distinct, although any prolonged study and comparison of the young would have shown that they were widely separated in development and really only morphic equivalents evolved from entirely distinct ancestors. “A good example of this is the Eudoceratide? in- cluding the Silurian and Devonian Eudoceras and Trip- teroceras, and probably gyroceran form Edaphoceras of the Carboniferous and the close-coiled nautilian shells of Endolobus of the Carboniferous. The pe- culiar forms of this series and their remarkable sutures enable the observer to follow the line both in the grada- tions of the adults and by means of the parallelisms of the development. ‘‘Another good series easily distinguished by the re- markable sculpture of the shells is Zittelloceras of the Silurian with cyrtoceran forms, and the gyroceran and nautilian Halloceras of the Devonian. “One of the best is Thoraceras, a rough spinous cone of the Silurian, Devonian, and Carboniferous, which has straight and cyrtoceran shells; the gyroceran Triboloceras of the Carboniferous, and the nautilian shells of Vestinautilus and its allies in the same period. «There is no possible explanation of the parallel- isms of development of these nautilian shells and the adult stages of others except heredity in the same gen- etic series. It is useless to waste time in discussion unless the facts are specifically denied after having been properly reéxamined. ‘When the ammonoids are taken up, it is easy to demonstrate? by the study of the young of the Gonia- 1‘ Genera of Fossil Cephalopods,’’ Proc. Bost. Soc. Nat, Hist,, p. 287. 2 See ‘‘ Genera of Fossil Cephalopods,”’ Proc. Bost, Soc. Nat. Hist., XXII, 1883, p. 303. PARALLELISM. 187 titinee that they had straight forms among their ances- tors and that these forms have a central siphuncle and suture as among nautiloids. The Devonian Goniati- tine and some of the Carboniferous forms had also gyroceran forms and loosely coiled nautilian forms, in- dicating an ancestry with similar cones, but at these stages the siphuncle is invariably ventral as in the adults. The young of all of the Ammonitinz, however involute the shell may afterwards become, have an in- variably straight or curved cyrtoceran cone in the apical part, and when they come in contact by growth, the first whorl or whorls are equally invariably open coils like the coils of the fourth grade in nautiloids. The fifth kind of shell, the involute-nautilian, follows in precisely similar succession to what it does in the ontogeny of nautiloids. Farther than this the degree of involution increases according to the species, with age, and the amount of this involution is often an im- portant part of the specific diagnosis. “Among Ammonitine one finds at once that there are no orthoceran or cyrtoceran shells except among the large group designated by the author as Bactrites. This genus begins early in the Silurian with shells that are not distinguishable from true Orthoceras except by having the siphuncle in adults and later stages close to the venter. Some of these forms have no bulb or protoconch and have a large scar on the apex as in true Orthoceras, others have a calcareous bulb or protoconch on the apex as in true Ammonitine. There are also open or gyroceran shells in the adults of the genus Mimoceras which are repeated in the young of Anarcestes and other genera of Goniatitine figured in my ‘Embryology of Fossil Cephalopods.’} 1Bull, Mus. Comp. Zodl., UI. 188 PRIMARY FACTORS OF ORGANIC EVOLUTION. The shells of the Ammonitine, however, are of the fourth and fifth kinds almost exclusively, and in xol- lowing out the separate genetic series one has to dis- tinguish the progressive gradations by means of the greater or less amount of involution even in the Go- niatitine of the Devonian and Carboniferous. «« There are also some very remarkable facts show- ing that the coiling is closer in the Mesozoic than in the Paleozoic forms as a matter of hereditary derivation. The young of the Silurian and Devonian forms have the open, slowly coiling whorls figured by Sandberger and Barrande and repeated by myself as referred to above, but the young of all Mesozoic forms are close coiled so far as known. This is shown in the centre of the umbilicus by the perforation or central opening which is extremely large in most of the Paleozoic Go- nititinea but becomes almost obliterated in the true Ammonitine of the Jura. ‘In tracing parallels between development of the individual and the series among Ammonitine it has been found by Branco and the author, that in orna- mented shells the young are first like a nautilus in the sutures, then have a goniatitic stage like the first rep- resentatives of the order of ammonoids in the Paleo- zoic, and that during these stages it is invariably smooth and similar in general form to these same an- cestors. After this nepionic stage is passed through the sutures and the characteristics alter with greater or less rapidity, but the stages show decisive parallel- isms with the immediate ancestors of the same genetic series. Some of the best examples of palingenetic de- velopment of this kind, where the later stages of growth present parallels with proximate ancestors, are cited in PARALLELISM. 189 my ‘Genesis of the Ariétidaz’! and others have been given by Buckman and Wiirtenburger. ««Some of the most remarkable occur in the least expected quarters. As usual, when one has a true law, it leads him into conclusions that are, perhaps, more surprising to himself than to his readers, or to any subsequent investigator. This was certainly my own case in being led to recognise the perfect examples of parallelisms in retrogressive series. Quenstedt and all students since his time agree that the so-called genera Crioceras, Hamites, Ancyloceras, Baculites, forms that are successively more and more uncoiled until in Baculites they are absolutely straight cones, were derived from normal, close-coiled, involute-nau- tilian shells of the Ammonitine. Their young have been repeatedly shown to be close-coiled and they grade into the normal progressive shells by all of their adult characters. «The ultimate fact in this demonstration has been added by Dr. Amos Brown in the discovery of a close coiled nepionic stage in the straight Baculites of the Cretaceous, the only form whose development had not been ascertained and whose exact ‘relations had not been determined. “It is now admitted by all students of Ammonitinz that these retrogressive groups are not true genera; but that as first demonstrated by Quenstedt, Baculites, Crioceras, etc., are retrogressive stages in the evolu- tion of distinct genetic series and that they do not ex- ist as natural groups of species. In other words, dif- ferent genetic series of the Ammonitine die out by passing through a series of modifications which are parallel and which are just the reverse of the parallel 1 Smithson, Contrib., 673, p. 41 et seq. 190 PRIMARY FACTORS OF ORGANIC EVOLUTION, series of the orthoceran, cyrtoceran, gyroceran shells with which each distinct complete genetic series of nautiloids arose in time. ‘‘While the nautiloids coil up in their progressive evolution and the Ammonitine increase this coiling up tendency in the primitive and progressive forms of each genetic series, the latter in becoming retrogres- sive reverse the processes of progressive evolution. They become more and more uncoiled, each complete retrogressive series ending with a straight cone. All other characters correlate with this uncoiling and in a general way may be said to degenerate in greater or less proportion to the amount of the uncoiling. To make this extraordinary picture complete it is only necessary to add that these retrogressive series followed out to their ultimate development are distinctly parallel with changes or stages of modification taking place in the senile stages of individuals of the same genetic group. . In old age the highly ornamented shell gradually parts with its spines and other ornaments, the whorls slowly diminish, the involution decreases and even- tually in extreme age it becomes separated from the spiral and completely rounded and smooth. The aper- ture becomes correlatively modified, and also the su- tures. If an old ammonite could have its life pro- longed, it would become Baculites, and the full-grown part of the shell would, in some forms of Lytoceratinz be very similar to the minute nepionic shell of the Ba- culites as described and figured by Dr. Brown. If now the coiled adult part of this imaginary shell were broken off and lost, the straight senile fragment would be re- ferred to the old genus Baculites. The morphic char- acters of the gerontic or old-age stage of ontogeny are PARALLELISM. 1gI therefore parallel with the forms evolved in the para- plastic or retrogressive stage of evolution of the phy- lum. -In other words, the morphic modifications which may occur as permanent, specific, and generic charac- ters in the adults of retrogressive descendants of any progressive individuals may be predicted from the study of the similar changes that take place in the senile stages of the progressive individuals. As it has been stated by the writer on several occasions, the embryonic, nepionic, and later stages of development up to the adult repeat with greater or less clearness in proportion to their removal in time and organization from the point of the origin of the genetic group to which they belong the permanent characteristic of their ancestors; the adult gives the existing essential differentials acquired by its own species, genus, and group, being the index according to the time of its oc- currence of the progression or retrogression of its group; the old, in its invariably retrogressive course, indicates the path that must be followed by degraded series after the acme of the group to which the indi- vidual belongs has been reached. This, of course, is a generalized statement of the correlations of the on- togenic cycle and the phylocycle when they occur as in the Ammonitine, but it will be found eventually that this law is true of all animals to some degree. It is obvious from all past experience that every law of correlation of structures cannot be true in any one group without being found more or less in all organ- isms. I have therefore ventured upon the basis of this and Beecher’s, Clarke’s, and Schuchert’s re- searches among Brachiopoda, corals, and trilobites, Dr. Jackson’s among pelecypods, and after the con- firmations by the independent researches of Wiirten- 192 PRIMARY FACTORS OF ORGANIC EVOLUTION. burger and Buckman among Ammonitine, and those of Bather among crinoids, to designate the complete study of the correlations of the ontecycle and phylo- cycle as Bioplastology. Bioplastology is easily sepa- rable from the study of growth, and from that of hered- ity, for which last I have proposed the term Genesi- ology, and from that of Ctetology or the study of the origin of acquired characteristics. By properly defin- ing these different branches of research it is practicable to see that bioplastology includes the results of the ac- tion of growth, the laws of growth, as well as those of genesiology and ctetology, but has a field entirely dis- tinct from all of these in so far as it deals essentially with the study of parallelism in all its phases.” The parallelism of the gyroceras with an early stage of all coiled Cephalopoda is represented in Fig. 119 page 410, as illustrative of the inheritance of an ac- quired character. 3. PARALLELISM IN THE VERTEBRATA. Parallels between the ontogeny and phylogeny are well known in the Vertebrata. The primary relations of the Vertebrata are discernible in the successive types of structure of the nervous system, and of the skeleton, but most clearly in those presented by the circulatory system. It is well known that the central organ—the heart, is, in the amphioxus, a straight tube. In the next higher group, the Marsipobranchii (lampreys), it is a bent tube, with a constriction which divides it into two chambers. In the Pisces (fishes) the heart is composed of two chambers related to each other in a reversed longitudinal direction. In the Ba- 1 See Proceedings of the Boston Society of Natural History, 1893, p. 59. PARALLELISM. 193 trachia and Reptilia the cephalad (auricular) division is divided into two chambers by a septum; while in the birds and Mammalia the caudad division (ventricle) is also so divided, making four chambers in all. ‘The sources of the great vessels which distribute the blood to the body and return it to the heart, dis- play the same successional relation of types. In the Acrania (amphioxus), the Marsipobranchii, and most of the fishes, the vessel (truncus communis) which receives the blood from the central organ, gives off several branches on each side, which are distributed to skeletal bars or arches which are in immediate con- tact with water, which aerates the blood. They then return, and, first sending the carotids anteriorly, unite dorsad to the heart, and form the aorta posteriorly. In the Batrachia, where aerial respiration succeeds to an aquatic one during the life of the animal, the number * of the vessels contributing to form the aorta is re- duced from five to three in the sucessive types. One of the arches is aborted as an arch, and sends the cir- culating fluid to the modified swim-bladder of the fish, or lung, where it is aerated. This aerated blood is returned to the heart with non-aerated blood from other organs, and the mixture is sent throughout the body. In the reptiles we have essentially the same system, but the aorta-origins are reduced to two, and one, on each side. Next a division of the truncus com- munis ensues, which corresponds functionally with that in the ventricle, so that the impure blood from one auricle is sent into the ventricle (right) which com- municates with the lung; and the aerated blood is then returned to the other auricle, which pours its contents into the left auricle, which drives it into the aorta, and thus throughout the body. Thus pure or Fig. 49.—Circulatory systems; 1-2, fish; 3-4, batrachian; 5, reptile; 6, bird; all from Gegenbaur. Figs, 7-8, human foetus, from His, PARALLELISM. 195 aerated blood is distributed to the organs, and all but one of the old roots of the aorta have ceased to function as such. This evolutionary succession is preserved with much fidelity in the ontogeny of the respective classes of Vertebrata. The representatives of each class pass through the stages which are permanent in the classes below them in the series. The Mammalia, as the highest class, pass through all the stages. (Fig. 49.) This series coincides also with phylogenetic succes- sion. The order of appearance in time of the Verte- brata is, first Agnatha, then Pisces, Batrachia, Rep- tilia, and Mammalia. In all the details of structure the same relation may be observed. Referring to the illustrations of phylogeny and variation of character described in the preceding pages, many of the characters definitive of natural divisions have been observed to appear in the course of the embryonic life of those types which pos- sess them. Those of greater systematic significance appear earlier, and those of less importance in a tax- onomic sense, later. I select some illustrations of this principle. I have shown that the primitive type of superior molar in the placental Mammalia is tritubercular, the fourth tubercle being added internally and posteriorly in the later forms. Dr. Taeker has recently observed that in the development of the superior molars in the horse, at an early stage the crown is tritubercular, and that the fourth cusp or hypocone is subsequently added, as in the phylogenetic history. As the horse presents the most complex molar among Mammalia, this sur- vival of the record is interesting. In the Artiodactyla and Edentata which lack su- 196 PRIMARY FACTORS OF ORGANIC EVOLUTION. perior incisor teeth, rudiments of them can be found in the early stages. We now know early extinct forms of both of these types where these teeth are permanent throughout life. In the toothless whalebone whales the same phenomenon has been observed. It is well known that the highest deer (Cervide) add an antler to the simple spike horn in the third year, and an additional antler with each successive year for several years. Also they develop a basal snag of the antler (see Cuvier, Ossem. Fossiles; Gray, Catal. Brit. Mus.) at the third year. Nowa majority of those of the New World (genera Cariacus, Coassus) never develop it except in ‘‘abnormal”’ cases in the most vigorous maturity of the most northern Cariacus (C. virginianus); while the South American Coassus retains to adult age the simple horn of the second year of Cervus. Among the higher Cervide, Rusa and Axis never assume characters beyond an equivalent of the fourth year of Cervus. In Dama the characters are on the other hand assumed more rapidly than in Cervus, its third year corresponding to the fourth of the latter, and the development in after years of a broad plate of bone, with points, being substituted for the addition ot the corresponding snags, thus commencing another series. Returning to the American deer, we have Blasto- cerus, whose antlers are identical with those of the fourth year of Cariacus. The oldest known deer (Palzomeryx) have no horns, or they are undivided. Among Batrachia excellent illustrations are fur- nished by the two series of Salientia, the Arcifera and the Firmisternia. PARALLELISM. 197 The firmisternial structure is a modification of the arciferous, which comes later in the history of growth, and probably also in geological time. During the early stages the Firmisternia have the movable shoul- Fig. 50.—Succession of horns of Cervus elaphus L. from Gaudry. a, second year represented by permanent horn of Coassus; 4, third year represented by permanent horn of Furcifer; c, fourth year represented by permanent horn of Rusa; 4, fifth year; é, sixth year, der girdle which characterizes those of the arciferous division, the consolidation constituting a modification superadded in attaining maturity, Furthermore, young Salientia are toothless, and one section of the species of Arcifera never acquire teeth. In these (the Bu- 198 PRIMARY FACTORS OF ORGANIC EVOLUTION. fonide) we have a group which is imperfect in two points instead of one. The genera of these salientian suborders exhibited on a preceding page as forming indentical series in five different families (pp. 66-67) are related to each other as developmental stages in the history of the genera that attain the extreme development on each line. For example we select the family Hylidz of which the terminal genus is Trachycephalus. Nearly allied Fig. 51.—Shoulder girdles of Anura; a, of the arciferous type (Phyllomedusa bicolor) ; b, Rana temporaria, tadpole with budding limbs; ¢, do., adult, firmis- ternial type; 4 and c from Parker, to it is the genus Osteocephalus, which differs in the normal exostosis of the cranium not involving the derm, as it does in the former. Close to this is Scy- topis, where the fully ossified cranium is not covered by an exostosis. Next below Scytopis is Hyla, where the upper surface of the cranium is not ossified at all, but is a membranous roof over a great fontanelle. Still more imperfect is Hylella, which differs from Hyla in the absence of vomerine teeth. Now, the genus Trachy- cephalus, after losing its tail and branchie, possesses PARALLELISM. 199 all the characters possessed by the genera Hylella and Hyla, either at or just before the mature state of the latter, as the ethmoid bone is not always ossified in advance of the parietals. It soon, however, becomes a Scytopis, next an Osteocephalus, and finally a Tra- chycephalus. It belongs successively to these genera, for an exhaustive anatomical examination has failed to reveal any characters by which, during these stages, it could be distinguished from these genera. The same succession in development of the genera of the other families is well known, the genus Otaspis of the Bu- fonidez attaining a point beyond any of the others, in the enclosure of its membranum tympani posteriorly by dermoéssification. Finally reaching in our review the relations of spe- cific characters, the readers will call to mind that the species of the lacertilian genus Cnemidophorus (page 41) are either striped, spotted, or cross-banded, and that the Lacerta muralis agrees with them in this re- spect. It was also shown that the young of all the species are striped, and that the cross-banded forms pass through not only a striped, but an intermediate spotted stage, before attaining the adult coloration. The young of spotted salamanders are without spots (genera Amblystoma and Salamandrae g.); so that unspotted species resemble the young of the spotted. In many species of birds of more or less uni- form patterns of coloration, the young are spotted. In some of these the females remain spotted throughout adult life. In some other species both sexes retain the spotted coloration of the young. The young of most deer are spotted. In the fallow-deer (Axis) the adults retain the spotted coloration, thus resembling the young of most of the species. 200 PRIMARY FACTORS OF ORGANIC EVOLUTION. 3. INEXACT PARALLELISM OR C/ENOGENY. When the early or transitional stage of a higher form is exactly the same as a permanent lower form, the parallelism is said to be ‘‘exact.” Such is the re- lation of a Cnuemidophorus gularis scalaris to a Cnemido- phorus gularis gularis as to color characters ; and of an Amblystoma tigrinum to a permanent breeding Siredon lichenoides in characters of higher structural value. When the transitional stage of the higher only re- sembles the lower form in some one or more features, but not in all, the parallelism is said to be ‘‘inexact.” It is evident that ‘‘exact parallelism ” can only exist between ancestor and descendant in the same re- stricted line, and can be therefore only demonstrated in the case of the nearest relatives, between which a perfect phylogeny is known. So soon as new subordi- nate characters are assumed, or a change in the order of appearance of characters supervenes, the parallel- ism becomes ‘‘inexact,’’ and such is the kind of paral- lelism usually observed. And it is more inexact the more widely removed in relationship are the forms compared. ‘Thus the parallelism between the embryo man with five branchial slits, and the adult shark, is very inexact; but that between a true fish and a shark is much less inexact. That between a higher and a lower shark is still more exact, and so on. Exact par- allelism in growth is called by Haeckel palingenesis or palingeny. The growth which has, through changes introduced subsequent to the origin of a line of de- scent, become inexact, or ‘ falsified,” is termed by the same author cenogenesis or cznogeny. PARALLELISM. 201 The superposition of characters which constitutes evolution, means that more numerous characters are possessed by the higher than the lower types. This involves a greater number of changes during the on- togenetic growth of each individual of the higher type. In other words, characters acquired during the phylo- genetic history are continually assumed by the pro- gressive form at earlier and earlier periods of life. This process has been metaphorically termed by Pro- fessor Alpheus Hyatt and myself ‘‘acceleration.” All progressive organic evolution is by acceleration, as here described. Retrogressive evolution may be ac- complished by a retardation in the rate of growth of the taxonomic characters, so that instead of adding, and accumulating them, those already possessed are gradually dropped; the adults repeating in a reversed order the progressive series, and approaching more and more the primitive embryonic stages. This pro- cess I have termed ‘‘retardation.”” Retardation is not however, always exact, even in retracing a true phylo- genetic line, whence in such instances the process may not be correctly described as retardation. Professor Hyatt has applied to such types the term ‘‘senile,” and gerontic ; and to the resulting condition, the term ‘¢senility.” His observations on this subject have been made on Mollusca, and principally on the Cephalo- poda, and are of fundamental importance in this con- nection. The history of a type which has passed through a full cycle of life, from its earliest appearance to its ex- tinction, is divided by Haeckel into three stages, viz.: those of its rise; full vigor, as displayed by predomi- nance of variations and numbers; and decadence. For these stages he uses the expressions Anaplasis, 202 PRIMARY FACTORS OF ORGANIC EVOLUTION. Metaplasis, and Cataplasis. For the processes which bring about the first and last of these conditions, Pro- fessor Hyatt has used the terms Anagenesis and Cata- genesis. Catagenesis is equivalent to degeneracy and has played an important part in organic evolution. I had used the term previously to Professor Hyatt for the same process, but with a wider application; ex- tending its use to inorganic nature as well.! (See Chapter IV. of this book.) Embryology has, however, revealed another series of phenomena which in many instances obscure the simplicity of the problem of ontogeny as presented in the preceding pages. It was the merit of Haeckel to generalize from the facts brought to light by this sci- ence, so as to present the relations which subsist be- tween the primitive stages of all multicellular animals. This is known as the Gastrewa theory. He showed that the primitive gastric cavity of all such animals is produced by an invagination of a portion of the sur- face of a primitive sphere or morula, which results from the segmentation of the odsperm. This hollow half-sphere he termed the gastrula, and the theoretical primitive animal which corresponds to it he called the Gastrea. Marine animals very similar to this Gas- trea have been discovered. Haeckel showed, how- ever, that gastrulas are not all alike, since they differ in the extent to which the segmentation of the odsperm may be carried, and the rate of segmentation of differ- ent parts of it. Thus early do inexact parallelisms arise. From this point onwards special peculiarities of the various developmental lines appear, some of which have especial reference to the necessities of em- bryonic life. Hence the trochosphere stage of so many lOrigin of the Fittest, p. 422. PARALLELISM. 203 invertebrate forms, and the nauplius and zoza of the Crustacea. Such are the statoblasts which are resting-stages for the embryos of fresh-water sponges and Polyzoa, and the glochidia of the Unionide, which are wanting in the marine forms of the same orders. Such are the amnion and allantois of certain Vertebrata and the placenta of certain Mammalia, which have no refer- ence to any structures but their own residua, found in the adults of those animals. A remarkable instance of this state of things ap- pears in the history of the evolution of the insects. It is quite impossible to understand this history without believing that the larval and pupal states of the high- est insects are the results of a process of degeneracy which has affected the middle periods of growth, but not the mature results. The earliest insects are the Orthoptera, which have active aggressive larve and pup, undergoing the least changes in their meta- morphosis (Ametabola), and never getting beyond the primitive mandibulate condition at the end. The meta- morphosis of the jawed Neuroptera is little more marked, and they are one of the oldest orders. The highest orders with jaws undergo a marked metamorphosis (Coleoptera, Hymenoptera), the Hy- menoptera even requiring artificial intervention in some instances to make it successful. Finally, the most specialized orders, the suctorial Diptera and Lepidoptera, especially the latter, present us with very unprotected more or less parasitic larval stages, both active and inactive. These animals have evidently de- generated, but not so as to prevent their completing a metamorphosis necessary for purposes of reproduction. Asis well known, many imagines (Saturniidz, CEstridz) 204 PRIMARY FACTORS OF ORGANIC EVOLUTION. can perform no other function, and soon die, while in some Diptera the incomplete larve themselves repro- duce, so that the metamorphosis is never completed. This history is parallel to that proposed by Dohrn to account for the origin of the Ammoceetes larval stage of the Marsipobranchii. He supposes this form to be more degenerate than the corresponding stage of its probable ancestral type in the ancestral line of the Vertebrata. An inactive life in mud is supposed by Dohrn to have been the effective cause. An inactive life on the leaves of plants, or in dead carcases, has probably been the cause of the same phenomenon in the Lepidoptera and Diptera. Thus we have developed an ontogeny within an ontogeny, and a phylogeny within a phylogeny. These facts do not, however, affect the general result in the least. They only show us that the persistent larve of those animals which possess them, have a history of their own, subject to the same laws of evolution as the adults. It results that in many cases the phy- logeny can only be determined by the discovery and investigation of the ancestors themselves, as they are preserved in the crust of the earth. In all cases this discovery confirms and establishes such definite con- clusions as may be derived from embryology. It is also clear that on the discovery of phylogenetic series it becomes at once possible to determine the nature of defective types. It becomes possible to ascertain whether their rudimental parts represent the begin- nings of organs, or whether they are the result of a process of degeneration of organs once well devel- oped. An excellent illustration of inexact parallelism is to be found on comparison of man with the lower Ver- PARALLELISM. 205 tebrata. I have pointed out! that in the structure of his extremities and dentition, he agrees with the type of Mammalia prevalent during the Eocene period (cfr. Phenacodus). Hence in these respects he re- sembles the immature stages of those mammals which have undergone special modifications of limbs and ex- tremities, such as Ungulata in which cenogeny has not obliterated the early stages from the embryonic record. These forms are probably extinct. I have also shown? that in the shape of his head man resem- bles the embryos of all Vertebrata, in the protuberant forehead, and vertical face and jaws. In this part of the structure most Vertebrata have grown farther from the embryonic type than has main, so that the human face may be truly said to be the result of a process of retardation. Nevertheless, in the structure of his ner- vous, circulatory, and for the most part, of his repro- ductive system, man stands at the summit of the Ver- tebrata. It is in those parts of his structure that are necessary to supremacy by force of body only, that man is retarded and embryonic. 5. OBJECTIONS TO THE DOCTRINE OF PARAL- LELISM. An objection to the theory of parallelism in its full sense has been recently put forward by Mr. C. Her- bert Hurst. He says, ‘‘My object now is to show that in neither case can a record of the variation at any one stage of evolution be preserved in the ontog- eny, much less can the ontogeny come to be a series of 1‘ The Relation of Man to the Tertiary Mammalia,’ Penn Monthly, 1875; Origin of the Fittest, 268. 2 The Developmental Significance of Human Physiognomy,”’ American Naturalist, June, 1883; Origin of the Fittest, 1887, p. 281. 3 Natural Science, 1893, p. 195- 206 PRIMARY FACTORS OF ORGANIC EVOLUTION. stages representing in proper chronological order some of the stages of adult structure which have been passed through in the course of evolution.” Again: ‘The early stages of the fish embryo are very like those of the bird embryo. These two do correspond to each other. The statement that the embryonic structure of a bird follows a course which is from beginning to end roughly parallel with, but somewhat divergent from, the course followed by a fish, is borne out by the actual facts. A bird does not develop into a fish and then into a reptile, and then into a bird. There is no fish-stage, no reptile-stage, in its ontogeny. The adult resembles an adult fish only very remotely. Every earlier stage resembles the corresponding earlier stage of the fish more closely. There is a parallelism be- tween the two ontogenies. There zs no parallelism be- tween the ontogeny and the phylogeny of either g bird or any other animal whatever. A seeming parallelism will fall through when closely examined.” <“ The promise that this theory gave of serving as the guide to knowl- edge of past history without the labor involved in pale- ontological research, was indeed tempting : and where the royal road to learning has been shown by it, itis not surprising that some zodélogists should have entered for the race along this road. To what goal that road has led may be learned by a comparison of the nu- merous theories as to the ancestry of the ‘Chordata’ which have been put forward by those who have adopted the theory without enquiring as to its valid- ity.” I have made this quotation as showing the point of view from which the doctrine of parallelism when in- correctly stated may be assailed. There is truth in the author’s accusation that embryologists who have PARALLELISM. 207 not used their results with proper caution, have been frequently led to incorrect and even absurd results. The errors of this class of biologists are mainly due to their ignorance of species in the adult state, and their ignorance of systematic biology or taxonomy. They profess to regard this branch of the science as only suitable for beginners, and as comparatively unim- portant, as compared with their own; yet one might as well attempt the study of philology without a knowl- edge of alphabets, as to study phylogeny without the knowledge of natural taxonomy. The correct discrim- ination of species, genera, etc., imposes much greater burdens on the faculty of judgment, than does any- thing to be found in any science which includes obser- vation and record only. But Mr. Hurst’s statement is somewhat overdrawn, and he does not give embryolo- gists the credit which is due to their theory of recapitu- lation. I think he will find the following, which I wrote in 1872! to be a correct statement of the facts, and a fair induction as to principles. ‘<] [22 ][ oo } srofe| arene a> ee Pelt, “weer Pollex Indeg Med? Ann? Minim? Index Med? Ann’ Fig. 78.—Diagram of carpus of a Taxeopod (A) and (8) of a diplarthrous ungulate. From Osborn. wards, so that the limb, were it free from the ground, would be twisted or rotated on its long axis from within, forwards and outwards. As the foot rests on Fig. 79.—Raccoon pacing, showing right fore foot just before recovery. From H. Allen. the ground, the limb experiences a torsion strain in the directions mentioned. This throws the weight on the interior bones of the lower legs, the radius and the tibia. Thus these bones have acquired a great supe- . 300 PRIMARY FACTORS OF ORGANIC EVOLUTION. riority in dimensions over the external elements (ulna and fibula) in all the Diplarthra. The bones of the inner side of the first carpal and tarsal rows have thus transmitted an ever-increasing share of the impact, as the radius and tibia have developed, and have grown se. (U.S Gu make ( Va) Fig. 80. Fig. 81. Fig. 80.—RAznocerus unicornis carpus. Arrow ending in P, line of impact in plantation; do. ending in X, line of strain in recover. Fig. 81.—Z£guus cabalius fore foot. Sc, scaphoid; Z, lunar; Cz, cunei- form; 7Yoz, trapezium and trapezoideo; Um, unciform; mg., magnum. with their growth at the expense of the external ele- ments, the cuneiform in the carpus, and the calcaneum in the tarsus, which have become very narrow elements in the higher Diplarthra. As the pressure has been obliquely from within outwards, the growth of the proximal elements, the scaphoid and lunar in front, From Brehm. Fig. 82.--Gazella dorcas, gazelle. 302 PRIMARY FACTORS OF ORGANIC EVOLUTION, and the astragalus behind, has been in the same direc- tion. It has been shown by Dr. H. Allen that just before the recover of the foot, the latter is directed outwards from a line parallel with the axis of the body, so that the weight falls on the inner part of the sole of the former. This naturally causes the bones of the foot to press inwards on the heads of the metapodials, so that the latter tend to grow outwards on the second tarsal row. In this way were produced the facets on the external side of the heads of the metapodials. Thus is accounted for, on simple mechanical princi- ples, the phenomenon of carpal and tarsal displace- ment exhibited in its highest development, by the Dip- larthra. It is significant that diplarthrism has not appeared in mammals which possess an elastic pad of connec- tive tissue on the soles, as in Unguiculata generally, and,especially in the Carnivora. Diplarthrism is pres- ent in the camels, which have a pad, but I have shown that this pad did not appear until a comparatively late geologic epoch, and long after diplarthrism had be- come established in the camels’ ancestors, the Poébro- theriidz. The faceting of the head of the astragalus as the result of impacts, is seen on comparison of the astra- gali of Phenacodus and Hyracotherium in Ungulata (Figs. 33-35), and of Dissacus and Mesonyx among the Creodonta. In this last genus we have the only faceted astragalus among carnivorous mammals, but this genus is at the same time subungulate. b. The Forms of Vertebral Centra. The mutual articulations of the vertebral column are those of the centra and of the zygapophyses. Many KINETOGENESIS. 303 important modifications in these articulations are to be seen in Vertebrata, the Reptilia presenting the great- est variety, excepting in the zygapophyses, which are tolerably uniform in that class. In the Mammalia. modifications of the central articulations are not more striking than are those of the zygapophyses. The forms of central articulation are four, viz.: the amphiccelous, the ball-and-socket, the plane, and the saddle-shaped. The first type is only seen in a very imperfect degree in Mammalia and in but very few vertebrz, where it is indeed but a modification of the plane. The ball-and-socket is chiefly found in the neck of the long-necked Mammalia, as the higher Diplarthra, and to a less degree in their lumbar re- gions, while the dorsal vertebre present an approach to the same type in the same groups. The saddle- shaped centrum is only found in Mammalia in the necks of certain genera of monkeys. The majority of Mammalia present the plane articulation of all the ver- tebral centra. In Mammalia in which movement of the vertebre on each other has become impossible, the centra co- éssify, as for instance in the sacrum. In this region the number of vertebrz coésified is directly as the length of the iliac bone, which supports and holds them im- movable. Such is their condition throughout the dorsal region in the extinct Edentata of the family Glyptodontide, where the carapace is, as in the tor- toises, inflexible, and which therefore limits the possi- bility of motion of the vertebral column. Another illustration is seen in the necks of the balenid Ceta- cea, and to some degree in the Delphinide and Physe- teride. The lack of present mobility of this part of the column is due to its extreme abbreviation, a char- 304 PRIMARY FACTORS OF ORGANIC EVOLUTION. acter which has been gradually developing during Ce- nozoic time ; since the earliest Cetacea had consider- ably longer necks than the later ones, and had their vertebral centra distinct. It appears to me probable that the shortening was the result of disuse. This disuse would arise from gradually increasing powers of locomotion through the water, a progress, which, judging from the character of the limbs of the Zeuglo- don, was evidently made after the time of the Eocene. The increase of speed would enable the animal to over- take and capture its prey, without the necessity of using a long prehensile neck in seizing it in the pur- suit. The ball-and-socket articulation of the vertebre is well known to be the predominant condition in the Reptilia, and the fact that it is necessarily associated with the flexibility of the column is equally well under- stood. The flexibility is directly as the weakness of the limbs, for in the large and long-limbed terrestrial Reptilia of the order Dinosauria, the vertebral articu- lations of the dorsal region, at least, are plane. That it is chiefly confined to, and best developed in, the most flexible regions, i. e., the cervical and lumbar, of the column of the Mammalia, also shows this neces- sary connection. There can be no doubt but that the ball-and-socket vertebral articulation has been pro- duced by constant flexures of the column in all direc- tions, as has been suggested by Marsh. lll. INCREASE OF SIZE THROUGH USE. Under this head I enumerate examples where the mechanical causes in operation are less self-evident than those included under the preceding section. They KINETOGENESIS, 305 are, however, probably due to the same process, viz., impact and strain. a. The Proportions of the Limbs and of Their Segments. The length of the legs of terrestrial Mammalia has increased with the passage of time. The inferior types of Mammalia now existing, as Marsupialia, Glires, Insectivora, Edentata, have short legs, with a few cases of extreme specialization as exceptions, such as kangaroos, rabbits, and jerboas (hind legs only), the Dolichotis patachonica, the Rhynchocyonide, and the sloths. In the orders which stand at the summit of the series, as the Diplarthra, Proboscidia, Carnivora, and Anthropomorpha, the legs are much increased in length, and this is especially marked in certain forms which stand in all respects at the summit of their re- spective orders. Thus in Diplarthra, the deer, ante- lope, and horse are distinguished for length of limb; in the Proboscidia, the elephant; in the Carnivora, the large cats and hyenas; in the Anthropomorpha, the fore limbs are long in all, the hind ones especially so in man. The cause of this elongation is apparently use. It is the hind legs that are elongated in a straight line in animals that walk on them, as man; and both, in those that walk on both, as the elephant. In animals that leap with the hind legs these are still more elongated, and are folded when at rest, and rapidly extended when in motion. In animals that climb with the fore legs, these are elongated, as in the Anthropomorpha, except man. In those that climb with all fours, all are elongate, as in the sloths. It must be remembered that these elongations are the sum of increments added one to the other through long ages of use in geologic . 306 PRIMARY FACTORS OF ORGANIC EVOLUTION. time. The mechanical character of that use has not been identical. It is of two principal kinds, viz.: im- pact and longitudinal strain. These two forms of en- ergy move in directions opposite to each other; the one as compression in the direction of the length of the bone; the other, as a stretching in the direction of the length of the bone. Both processes alike appear to have stimulated growth in the direction of the length of the bone. The increase in the length of the legs has not been always due to increase in length of the same segment. In a majority of the higher mammals, the increase has been principally in the foot, and especially in the meta- podials and digits, producing digitigradism. In the forms which have remained plantigrade, the femur (Proboscidia), or femur and tibia (Quadrumana), or all three segments (Tarsius), have been the seat of the elongation. We can again trace these especial elonga- tions to special uses. In animals which leap, the dis- tal segments of the limbs are elongated; in those which do not leap, but which merely run or walk, it is the proximal segments of the limbs which are elon-: gated. Animals which run by leaping are divided into those which run and leap with all fours, as Diplarthra ; and those which run and leap with the posterior limbs only, as the jerboas and kangaroos. In both types, the distal segments of the hind limb are elongated, and in the Diplarthra, those of the fore limb also. Animals which do not leap in progression (ele- phants, Quadrumana, bears) are always plantigrade, and have very short feet, but elongate thighs, and, mostly, tibias. These facts show that those elements which receive e KINETOGENESIS. 307 the principal impact in progression are those which increase in length. In digitigrade animals it is the feet which receive the impact of the repeated blows on the Fig. 83.—Pes of (A) Merychocherus montanus from Scott; (B) Bos taurus, much reduced. Ca, Calcaneum; As, Astragalus; Va, Navicular; Med, Na- viculocuboid ; Cz, Mec, Ecto-mesocuneiform ; /¢, Metatarsals (cannon bone); Enc, Entocuneiform. earth while in progression, while supporting the weight of the body at every stage of the process. In planti- grade animals it is the soles of the feet, and the bones of the leg in line with them. which receive the impact, 308 PRIMARY FACTORS OF ORGANIC EVOLUTION. Fig. 84.—Anterior feet of primitive Ungulata, reduced. A, Phenacodus primevus. B, Coryphodon elephantopus, Hyracotherium venticolum, G KINETOGENESIS. 309 while the feet beyond this point receive none, and do not support the body, except very partially at the mo- ment of leaving the earth. b. The Number of the Digits. The reduction in the number of toes is supposed to be due to the elongation of those which receive the greater number of strains and impacts in rapid pro- gression, and the complementary loss of material available for the growth of those not subject to this stimulus. This is rendered probable from the fact that the types with reduced digits are dwellers on dry land, and those that have more numerous digits are inhabitants of swamps and mud, or are more or less aquatic. That this inequality is due to these mechan- ical causes is still further indicated by the fact that in those forms where the soles are thickly padded (Car- nivora, Proboscidia) the reduction has either not taken place, or has made little progress, amounting to the loss of only one digit. (An apparent exception in the case of the camels will be mentioned later.) A still more important body of evidence which shows that the inequality in size and number of digits is due to impacts and strains unequally distributed, has been brought forward by Ryder. He points out that defi- nite results are to be observed in those limbs of a given type of animal which experience correspondingly defi- nite influences; while in the limbs where the strains are equal, the modifications do not appear. Examples of this kind are to be found in the unguiculate Mam- malia and in the Marsupialia. Thus in the jerboas which use the hind limbs in leaping, these only dis- play reduced digits, the fore limbs remaining of prim- 310 PRIMARY FACTORS OF ORGANIC EVOLUTION. itive character. The same is true of the kangaroos. In digging genera, the fore limbs experience the modi- Ni Hi | Hi ai rh } i Fig. 85.—A, Right posterior foot of Prothippus sejunctus Cope, from Colo- rado, about one-half natural size. From U.S. Geological Survey of Territo- ries, F. V. Hayden, IV. 8, Right posterior foot of Poébrotherium labiatum Cope, from Colorado, three-fifths natural size. From Hayden's Report, IV., Plate CXV. fications, while the hind limbs are more normal, as in Chrysochloris and various Edentata. KINETOGENESIS. 3Ir Ryder sums up the evidence in two propositions, as follows :1 «I. The mechanical force used in locomotion dur- ing the struggle for existence has determined the digits which are now performing the pedal function in such groups as have undergone digital reduction. “IT. When the distribution of mechanical strains has been alike upon all the digits of the manus or of the pes, or both, they have remained in a state of ap- proximate uniformity of development.” The application of the impact, or strain, or both, in progression, is easily understood. In recover (see Pp. 299), the leg is bent on the foot as it rests on the ground, and those digits which then leave the ground last, sustain greater strain than those which leave it sooner. In replacing the foot on the ground (planta- tion), those digits which strike it first experience greater force of impact than those which strike it later. Sup- posing the five primitive digits to have been of equal length, the distribution of the impact and of the strain will depend on the angle at which the foot is directed - with reference to the direction of motion. If the feet are pointed forwards, the middle digits will experience strain and impact ; if outwards, the inner digits bear the weight; if inwards, the external digits receive it. Observation on five-toed plantigrade: mammals shows that their feet are turned neither inwards nor outwards in progression, but straight forwards. It is probable that the primitive Mammalia moved in the same manner. ‘This is also to be inferred from the fact that they were plantigrade, so that the leverage transversely in or out which results from the elevated heel of the digitigrade leg was very much less in them. lAmerican Naturalist, 1877, p. 607. 312 PRIMARY FACTORS OF ORGANIC EVOLUTION. In progression of this type, the middle digits of course leave the ground last and strike it first. Thus the middle toes have been stimulated at the expense of the lateral ones, so that in the Diplarthra, either the mid- ssmiztay ‘S Ssnz0jzay ‘b sens‘ ‘snuvjog -odpy ‘t ‘snuevjogodgery ‘t ‘AYSABTEMOY UIOIJ | paonpal YONur ‘e[AjOepory Jo snueW—'9g “BIg dle one (Perissodactyla) alone remains, or the middle two (Artiodactyla). In the kangaroos the external toes have been chiefly used, so that the fourth and fifth digits have been principally developed. In man, who KINETOGENESIS. 313 now turns his feet out when using them as bases of re- sistance to muscular labor, the inner digit has become most robust. The mechanical history of the human great toe is however yet unknown. As regards the equal development of the third and fourth digits in the Artiodactyla, as distinguished from the development of the middle digit of the Perisso- dactyla, I have advanced the following hypothesis. I have supposed that the primitive members of this for- mer division sprung from pentadactyl plantigrades who dwelt in swamps and walked on very soft ground. ‘The effect of progression in mud is to spread the toes equally in all directions and on each side of the me- dian line. Such feet remain in the mud-loving hippo- potamus, and to a lesser degree in the true pigs. From such ancestry the cloven-footed Diplarthra derived their characters. The Hyracotheriinz, the ancestors of all Perissodactyla, display on the other hand evi- dence of a life on harder ground, especially in the pos- terior foot, where articulations are already rigidly de- fined, and the third digit is longer than the others. Some of their descendants love swamps, as one or two species of tapirs and rhinoceroses, but others live on the dryest ground, as the Andean tapir and the African rhinoceros. As to the highest members of both even and odd toed groups, the Bovide and the Equida, their habitat is in the vast majority of cases the dry land (Figs. 80-81). Continued and excessive prehensile strain with weight on the longest digits, must be assigned as the cause of the especial elongation; and disuse as the cause of the loss of the external and shorter digits, of the sloths; so that there remain but two and three (Choleepus and Bradypus), and in the climbing ant- 314 PRIMARY FACTORS OF ORGANIC EVOLUTION. eater (Cycloturus) but one principal toe and two rudi- ments. The excessive strain and impact experienced by certain digits in leaping, accounts for the digital reduction in the hinder foot of the kangaroos and jer- boas, precisely as in the perissodactyle ungulates. c. The Horns. Horns are developed in Mammalia and other Ver- tebrata on similar parts of the skull, principally on the posterior lateral angles, as in various Batrachia, Rep- tilia, and Mammalia, and on the nose, as in a few Mammalia and several reptiles, recent and extinct. These parts are the ones which are especially brought into contact with resisting bodies; the nose in pushing a path or way for the head and body; the lateral occi- pital region in defence and assault, when the sensitive nose and eyes are protected: by being held near the ground. In the latter position the posterolateral an- gles, when present, receive more frequent collision with, and vigorous stimulation from, a body attacked or resisted, and in accordance with the observed re- sults of irritation on dermal and osseous tissues, addi- tional matter has been deposited. In Lacertilia and Batrachia Salientia there are distinct posteroexternal cranial angles; in Batrachia Urodela such angles are less prominent. In unguiculate Mammalia and in all others with a sagittal crest there are no such angles; hence this type of skull has never developed posterior horns. The rhinoceros has developed the dermal nasal horn, and the Elasmotherium, a median osseous horn, since posterolateral angles of the skull are want- ing or close together. Inthe Dinocerata and the Ar- tiodactyla, where the temporal crests are lateral, leav- ing a wide fronto-parietal plane with posterior lateral KINETOGENESIS. 315 angles, horns are developed. In members of both groups horns have been developed over the orbits also (Fig. 87), and in the Dinocerata on the extremities of the nasal bones as well. These growths are all on parts which are subject to especial irritation by contact with other bodies, animate and inanimate. Among Artiodactyla, the deer (Cervide) are espe- cially distinguished by the periodical shedding of all but the bases of their horns. Extinct forms found in the Upper Miocene of the United States and France (the Loup Fork series) furnish the explanation of the origin of this remarkable peculiarity. In the genus Cosoryx we find that the horns may or may not pos- sess a burr near the base of the beam, like that of the deer; the same species being indifferently with it or without it. This observation has been made on three species,—the C. necatus, C. furcatus and C. ramosus. The following explanation of these facts has been pro- posed by myself.1 “From the facts of the case the following inference may be derived, premising that it is very probable that a genus allied to the present one has given origin to the family of the deer. It is ob- vious that the horns of (Dicrocerus) Cosoryx did not possess a horny sheath as in the Bovide, from the fact of their being branched. As the sheath grows by ad- dition at the base, the presence of branches which necessarily obstruct its forward movement, would be fatal to the process. There is much to be said in favor of the view that the horns were covered with an integu- ment, probably furred, as in the giraffe and young stage in the deer. Thus there are grooves in the sur- face of the beam for superficial blood-vessels, which 1U.S. G. G. Survey West of the rooth Mer., G. M. Wheeler: IV., Paleon- tology, 1877, Pp. 348. 316 PRIMARY FACTORS OF ORGANIC EVOLUTION, must have been protected by skin (I do not observe these grooves on the beam of C. ¢eres). The retention of the broken extremity of an antler, so as to be re- united, as described (Fig. 87, C), could not have been accomplished without an integument. The presence of the burrs cannot be accounted for on any other supposition, as there are no foramina to give exit to nutrient vessels at’ the point where they exist; the irregularity of those positions also forbids the latter idea, and adds to the probability that the arteries which furnished the deposit of phosphate of lime were contained in a super- ficial dermal coating. The supposition is also strengthened by the fact that the only existing ruminants (the giraffes) Fig. 87.—A, B, Cosoryx necatus Leidy; A, without, &, with, burr on antler; C, D, Cosoryx ramosus Cope; C, antler broken With permanent horns and reunited; D, beam with burr; one- : half natural size; original; from Report without horny sheaths U.S. Gov. Geol, Expl. rooth Mer, G.M» have them covered with eee hairy skin. ‘“‘It appears that in the antlers of the Cosoryx the deposit of a burr was immediately associated with the death of the portion of the horn beyond it, so that it disintegrated and disappeared. This was not the case with the beam in the specimens observed. Neverthe- KINETOGENESIS. 317 less it is probable that the death of the horn would be associated with the deposit of the burr in this case also, were the conditions the same. What those con- ditions were we can only surmise. It was very prob- ably the death of the integument which invested and nourished the horn that produced that result; and this would more readily occur in the exposed antlers than in the more protected basal portion of the beam. It is very probable that this result would follow blows and laceration of the surface received during combat, or accidental contact with hard substances. The in- tegument would be stripped up to near the junction of the antlers with each other, or of the beam with the cranium, and the arteries would be constricted or closed at those points. It is near these junctions that all of the burrs are found. But as such lesion would be necessarily less complete at the point where the ° horn has greatest circumference, so the entire death of the horn might be less usual than that of the branches. Should such lesions have occurred for a long period at the breeding season, nature’s efforts to repair by redeposit of bony tissue might as readily be- come periodical as the increase in size and activity of the reproductive organs and other growths which char- acterize the breeding season in many animals. The subsequent death of the horn would be at some time followed by its shedding by the ordinary process of sloughing.” Cosoryx is not the true ancestor of the Cervida, as its teeth have already attained the prismatic type of the higher Bovide. But Blastomeryx is most prob- ably the ancestor of the deer. The remains of this genus occur with those of Cosoryx, but the burr has not yet been observed on its horns. 318 PRIMARY FACTORS OF ORGANIC EVOLUTION. lv. THE MECHANICAL ORIGIN OF DENTAL TYPES. In investigating the origin of dental types it is necessary to become acquainted with the nature of the mutual movements of the series of the opposing jaws. I have classified them as follows :1 I. Inferior molars work within superior molars, but not between them. Psalidodect mastication. 1. The inferior molars shear on the interior side of the su- perior : Triconodontide. II. Part or all of inferior molars work alternately to and between superior molars. Amoebodect mastication. 2. The inferior molar shears forwards on the superior mo- lar. Proterotome mastication: Creodonta; Carnivora. 3. The inferior molars shear posteriorly against the superior molars. Opisthotome mastication : Coryphodontide, Uintatheriide. III. Molar teeth of both jaws oppose each other. Antiodect mas- tication. 4. The movement of the lower jaw is vertical. Orthal mas- tication : Suotdea, Tapiride, 5, The movement of the lower jaw is from without inwards. Ectal mastication : many Perissodactyla. 6. The movement of the lower jaw is from within outwards. Ental mastication : most Artiodactyla; some Perrissodactyla, 7. The movement of the lower jaw is from before backwards, Proal : some Monotremata Multituberculata and most Glires. 8. The movement of the lower jaw is from behind forwards. Palinal : Proboscidea (Ryder). The distinction of teeth into incisors, canines, and molars appears independently at various points in the line of Vertebrata. Incisors and molars are distin- 1Mechan, Origin Hard Parts of Mammalia, 1889, p. 226. KINETOGENESIS. 319 guished in sparoid fishes, and in placodont and dia- dectid reptiles. Canine-like teeth, or pseudo-canines, appear in clepsydropid and crocodilian reptiles, and in saurodont fishes. Canine-like incisors appear in the Clepsydropide. The variety of character in these structures presented by the Mammalia to be consid- ered is great, and the principles deduced from obser- vation of them are applicable to the Vertebrata in general. As mechanical causes of the origin of dental modi- fications, I have enumerated the following : 1. Increase of size of a tooth, or a part of a tooth, is due to increased use, within a certain maximum of capacity for increased nutrition. 2. The change of direction and use of a tooth take place away from the direction of greatest, and in the direction of least resistance. 3. It follows, from their greater flexibility, that crests of crowns of teeth yield to strains more readily than do the cusps. 4. The increase in the length of crests and cusps in all directions, and therefore the plications of the same, is directly as the irritation from use to which their apices and edges are subjected, to the limit set by the destructive effects of such ‘use, or by the re- cuperative energy of nutrition. 5. The direction of growth of the branches of a V, or of the horns of a crescent, will be the direction of movement of the corresponding parts of the opposite jaw. Before giving a review of the various dental types of Mammalia, I wish to describe some special exam- ples where the effect of mechanical causes is most ob- vious. I therefore first repeat the observation of Ryder 320 PRIMARY FACTORS OF ORGANIC EVOLUTION. as to the origin of the selenodont molars of the Artio- dactyla; and my own as to the origin of a similar structure in the molars of certain multituberculate Pro- totheria. In the former the mastication is ental; in the latter it is proal, as shown by Osborn. In the accompanying figure from Ryder the move- ments of the lower jaw in mastication of lophodonts, are diagrammatically represented. @ represents the movement in Carnivora, and in the orthal bunodonts, as the pigs. 4 shows a slight lateral movement be- lieved by ee to exist in the wart hog (Phacoche- MNOS Fig. 88.—Diagram of excursion of lower jaw in mastication; from Ryder; a-b, orthal; c-/, ental. rus). ¢ represents the movement in kangaroos, pha- langers, and tapirs. Ind a theoretical intermediate movement is represented, such as Ryder supposed to have characterized the Anchitherium. In ¢ the usual movement among ruminants is depicted, as is seen in the deer, etc. In / the wider excursion of the jaw is that seen in the giraffe, camel, and ox. In these move- ments from 4 to f, the lower jaw is moved transversely across the upper jaw from one side to the other. Some of the Diplarthra masticate on one side of the jaw when performing this movement, and some on the other. That is, in passing the lower jaw across the face of the upper, some masticate the food on the side KINETOGENESIS. 321 where the external face of the lower jaw crosses the upper jaw from within outwards (ental); while in other types the food is masticated on the side where the lower jaw passes the external edge of the upper jaw from without inwards (ectal). While masticating with one side of the jaws, the opposing dental series of the other side are not in contact. All mutual effect of the teeth of one jaw on the other could therefore appear on the side temporarily used for mastication only. Among recent Un- gulata the rumi- nants present the ental mastication ; the rhinoceros and horses, the ectal ; and rodents, the proal. Ryder is of the opinion that the mastication of the Proboscidia is pali- nal, but I have not Fig. 89.—Cervus, molars: a, superior, exter- been able to satisfy nal view ; 4, do. inferior view; c, inferior molars, superior view; from Ryder. myself of this. When the crests of the inferior molars were devel- oped, their relation to the crests of the superior molars was always anterior in mastication. That is, the in- ferior crest, in the closing of the jaw, collides with the crest of the upper molar, with its posterior edge against the anterior edge of the latter. This is because: first, as to position, the two anterior cusps of the lower molar are the remains of the anterior triangle which fit originally between two superior molars, and: be- cause, in the closing of the jaw, these cusps continue to hold that position ; and second, as to function, be- i 322 PRIMARY FACTORS OF ORGANIC EVOLUTION. cause the canine in the ungulate series diminishes in size, and does not, therefore, draw the inferior molars forwards by wedging on the superior molar, as in the Carnivora, but allows free scope to the posterior trac- tion of the temporal muscle in its exercise on the lower jaw. In those forms which masticate from the inside out- wards (the ental type), the cusps of the inferior molars, passing between those of the superior molars, would tend to flatten the sides on which they exerted fric- tion, and to extend those sides outwards beyond the median apex of the cusp. (Fig. 90.) The result would be, and taking into view the yielding of the tissue to Fig. g90.—Cusps of superior premolars and molars: a, external cusps of molar of Sarcothraustes; 4, of Phenacodus; c, of Anthracotherium; d, of Oreodon; ¢, half of inferior molar of Cervus; /, superior premolar of Cory- phodon; from Ryder. such strain, has been, to modify the shape of the cusp by pushing its side walls, so that a horizontal section of it would become successively more and more cres- centic. The effect on the inferior teeth would be to produce the same result in their external cusps, but in the opposite direction. The sides of the cusps would be pushed inwards, past the apex, giving a crescentic section more or less perfect, as the operation of the cause had been of long or short duration. The result of the lateral movement in mastication may be under- stood by reference to the accompanying cut, Fig. gr. The external crescents of the inferior molars (4) are seen to pass between the internal crescents of the KINETOGENESIS. 323 superior molars (2). The mutual interaction and effect on the form of the crescents may be readily under- stood. In Fig. go the successive stages of this effect Fig. 91 —Two true molars of both jaws of a ruminant: @, superior molars, their inner crescents; 4, inferior molars, their external crescents; the arcs show directions of motion of jaws in mastication; from Ryder. on one or two cusps may be seen, beginning with a cone (@) and terminating with crescents (¢/). Thus is the origin of the selenodont dentition of the highest Fig. 92.Transverse vertical sections of superior molar teeth, showing transition from bunodont (4) type to lophodonts (8, C). A, Sus erymanthius. B, Ovis amaltheus. C, Bos taurus, From Gaudry. Letters: d, dentine; e, enamel; c, cementum. artiodactyle explained by Ryder, and, I believe, cor- rectly. Kowalevsky and I have shown that the types with selenodont (crescent-bearing) molars, have descended 324 PRIMARY FACTORS OF ORGANIC EVOLUTION. from tubercle-bearing (bunodont) ancestors. This de- scent has witnessed an increased depth of the infold- ing of the crown, as represented in the accompanying figure. Now, in the table of masticatory types above given it is shown that in the bunodont type of the Suoidea the mastication is orthal, and a gradual in- crease in the length of! the lateral excursions of the lower jaw has been shown to have resulted in the ental mastication. Thus has structure kept pace with func- tion in the evolution of the selenodont dentition. Fig. 93.— Chzrox plicatus Cope, palate and molar teeth from below, three- halves natural size. From Puerco bed of New Mexico. From American Nat- uralist, 1887, p. 566. The general structure of the dentition in the Proto- theria Multituberculata is similar to that of the Glires. The incisors in the Plagiaulacide, Chirogide, and Poly- mastodontide have structure and functions generally similar to those of that order. The result in the form and function of the molar dentition has been sim- ilar to that observed in the Glires. The postglenoid process is probably absent in these animals; in any case the mandibular condyle is rounded, and is not transverse. Prof. H. F. Osborn has pointed out to me that mastication was performed by a fore-and-aft KINETOGENESIS. 325 movement of the inferior molars on the superior in Plagiaulacide. This was no doubt the case in the other families named. The molar teeth of the lower types, as Tritylodon from the Trias, present conical tubercles in longitudinal series, two in the lower and three in the upper jaw. The two series of the lower jaw alternate with the three in the upper jaw, moving in the grooves between the latter, while the three se- ries of the upper molars reciprocally embrace the two of the lower molars. This is demonstrated by the mutual wear of the tubercles seen in Ptilodus and Chirox (Fig. 93). The trituration was probably the same in Tri- tylodon, but in Polymastodon the increased thickening of the tubercles prevented their interlocking action in masti- cation. In this genus the tu- bercles slid over each other, Fig. 94.—4, Meniscobssus conquis- and truncated the apices i eae a oT until in old specimens they twicenatural size. B, Meniscoéssus, were entirely worn away. In second species, from Osborn. Meniscoéssus (Fig. 94) and Stereognathus we have an interesting illustration of the effect of the action of cusps on each other when under prolonged mutual lateral thrust. Their external sides have been drawn out into angles in the direction of thrust, converting their transverse sections from circles to crescents. As the thrust is in the Multituberculata longitudinal, the crescents are transverse to the axis of the jaw. Inthe selenodont Artiodactyla, where the thrust is transverse to the line of the jaw, the crescents are longitudinal. That similar effects should accompany similar move- 326 PRIMARY FACTORS OF ORGANIC EVOLUTION. ments in two groups of Mammalia so widely separated as these two is strong evidence in favor of the belief that the two facts stand in the relation of cause and effect. I now present an example of the effect of strain, as shown by the direction of the inferior incisors of the lemurine Quadrumana. These teeth project horizon- tally from the extremity of the mandible, so as not to oppose the superior incisors, in consequence of which they are useless as organs of prehension. But they Fig. 95.—Lemur collaris, dentition from below and above; natural size; original. are used by their possessors as a comb for the fur, drawing them from below upwards when thus employ- ing them. The strain is always in one direction, and must have resulted in developing the procumbent posi- tion which they now display (Fig. 95). This is a di- rect deduction from the fact that the incisor teeth are similarly displaced by the pressure of the tongue in cases of the abnormal enlargement of that organ in man. I now describe the general character of mammalian dentition, with the view of pointing out how strong, in KINETOGENESIS, 327 the light of the facts already cited, is the evidence of their origin through mechanical strains and impacts. a. The Origin of Canine Teeth. The origin of canine, pseudo-canine, and canine- like incisor teeth is due to the strains sustained by them on account of their position in the jaws at points which are naturally utilized in the seizing of prey, or the fighting of enemies. In some reptiles (Dimetro- don) the end of the muzzle has been utilized ; in croco- diles, the side of the jaw; while the intermediate position has been most used by Mammalia. The rea- son why the canine instead of the incisor teeth have been selected by carnivorous Mammalia for prehensile purposes is not at present clear to me. In accordance with Rule I., its increased size has been due to the especial and energetic strains to which it has been subjected while in use as a prehensile or offensive weapon, when buried in the body of its prey or enemy. The superior canine would acquire larger size earlier in time than the inferior canine, since it bears the greater part of such strain, as attached to the more fixed head and body of its possessor. The anterior teeth of the lower jaw would be less available for use, since they offer weaker and less fixed resistance to the opposing body. That the first tooth behind the canine was not generally enlarged is (under I.) due to the fact that its posterior position prevents it from having the same amount of use, and experiencing the strain that a tooth more anteriorly placed necessarily re- ceives. It is excluded from considerable use by the projecting muzzle above and in front of it. That it was not drawn out into a horizontal position was due to the presence of teeth anterior to it. 328 PRIMARY FACTORS OF ORGANIC EVOLUTION. That the increased size of canine teeth is due to strains is strongly indicated by the huge development of these teeth in the walrus. This animal uses its ca- nines for the breaking of ice, and for lifting itself from the water on to the edge of strong ice. The fact that canines and not incisors have been thus developed is a necessary result of the fact that the walrus is a de- scendant of a line of animals which had already re- duced incisors and larger canines. Fig. 96.—Esthonyx burmeisterii Cope, dentition: @, profile; 4, superior; c, inferior dentition, grinding faces. Reduced. 6. Development of the Inctsors. The history of the incisor teeth of the Mammalia exhibits three processes, viz.: hypertrophy (e. g. Glires), specialization (e. g. Galeopithecus, Lemur- ide), and atrophy (e. g. Bodidea, Phacochcerus, Glos- sophaga, etc.). KINETOGENESIS. 329 Of hypertrophy we have two types: the first repre- sented by the Glires, Multituberculata, Tillodonta and their ancestors; and second, by the Proboscidia, the narwhal and certain Sirenia. As the uses of the inci- sors present two types corresponding with their struc- ture, we have ground for believing the uses in question to have been the efficient agent in producing the lat- ter. Esthonyx furnishes us with an example (Fig. 96) where all the incisors are present in the lower jaw, and Fig. 97.—Psittacotherium multifragum Cope, mandibular ramus, one-half natural size; «, profile; 4, from above. where the function of one pair of them (the second) has evidently been partially rodent in character; that is, it has served as a scraper and gouger of food sub- stances. Persistent use has apparently developed the size of this pair of teeth, until we find in Psittacothe- rium (Fig. 97) they have reached a greater efficiency, and that the external incisors of the lower jaw have disappeared. This disappearance can be accounted for on the ground of disuse, a retirement from service due to position, and the increased growth of incisor 330 PRIMARY FACTORS OF ORGANIC EVOLUTION. No. 2. In Calamodon the first incisor has become ru- dimentary from the same cause, and in Anchippodus it has disappeared altogether, leaving a truly rodent incisor dentition, consisting of the second incisors only, in the lower jaw. Continued use as chisels has developed these teeth to the great proportions seen in such Glires as Castoroides, etc. (Fig. 107). The use which the Proboscidia and Sirenia (Hali- core) give their incisors, is, from a mechanical point of view, like that which the Carnivora give their ca- nines ; that is, it consists of impacts in the long axis, and strains transverse to the long axis of the tooth. The elephants use their tusks for prying up the vege- tables on which they feed, or for pushing aside the vegetation through which they wish to pass. The an- cestors of the Proboscidia are not certainly known, but they possessed incisors of enlarged proportions, such as we find in the Toxodontia and other late rep- resentatives of some of the primitive Ungulata. Use of such teeth in the manner referred to, without oppo- sition from the inferior incisors, will account for the tremendous proportions which they ultimately reached in some of the species of Elephas. The use made by the narwhal of its single huge superior incisor, that of an ice-breaker, indicates the origin of its large dimensions. So with the straight incisors of the hippopotamus; use as diggers has straightened them to a horizontal from their primitive vertical direction, a change which is also partially ac- complished in the true pigs (Sus). In the Sirenian genus Halicore the upper incisors have been used in excavating vegetable growths from the banks and bottom of shallow seas. The transition from three incisors (Prorastomus) to two (Dioplothe- KINETOGENESIS. 331 rium), and to one (Halicore), is identical with what has taken place in the Proboscidia and Glires, and has resulted in the production of an effective digging- tool. In other genera it may be supposed that their habits of browsing on soft growing materials did not necessitate the use of digging incisors, hence these teeth became atrophied, as in the manatee and Rhy- tina. c. Development of Molars. In fishes and reptiles where teeth occasionally pre- sent very primitive conditions, the theory of the origin of particular types of molar teeth is more simple than in the case of Mammalia. The observations of Hiiter on the action of osteoblasts under stimulus show that under moderate irritation osseous tissue is deposited, while under severe pressure osseous tissue is removed. Koelliker has shown that the action of these bodies is the same in dentine as intrue bone. Hence modifica- tions of dental structure must stand in close relation to the uses to which they are put. Thus severe pressure on a simple tooth crown would, if long continued, cause it to expand laterally, or in the direction of least re- sistance, and to grow but little in its vertical axis, i. e., in the direction of greatest resistance. The molar teeth have been subjected to much more severe direct irritation from use than any others in the jaws, and this will account for their increased diameters. In the case of the eutherian Mammalia, molar teeth are not traceable back to ancestral types of reptilian mo- lars, but to simple conic (haplodont) reptilian teeth. The process of the evolution of the complex mamma- lian molars from these, forms the subject of a paper in the American Journal of Morphology for 1889, from which I quote extensively in the present work. 332 PRIMARY FACTORS OF ORGANIC EVOLUTION. I have there shown that the greater number of the types of this series have derived the characters of their molar teeth from the stages of the following succes- sion. First a simple cone or reptilian crown, alternat- ing with that of the other jaw (haplodont type). Sec- ond, a cone with lateral denticles (the triconodont type). Third, the denticles to the inner or outer side of the crown, forming a three-sided prism, with tritu- bercular apex, which alternates with that of the oppo- site jaw (tritubercular type). Fourth, development of a heel projecting from the posterior base of the lower jaw, which meets the crown of the superior, forming a tuberculosectorial inferior molar. From this stage the carnivorous and sectorial dentition is derived, the tritubercular type being retained. Fifth, the develop- ment of a posterior inner cusp of the superior molar and the elevation of the heel of the inferior molar, with the loss of the anterior inner cusp. Thus the molars become quadritubercular, and opposite. This is the type of many of the Taxeopoda, including the Quadrumana and Insectivora as well as the inferior Diplarthra. The higher Taxeopoda (Hyracoidea) and Diplarthra; add various complexities. Thus the tu- bercles become flattened and then concave, so as to form V’s in the section produced by wearing ; or they are joined by cross-folds, forming various patterns. In the Proboscidia the latter become multiplied so as to produce numerous cross-crests. @. Origin of the Carnivorous Dentition. The anterior cusplet of the triconodont crown is (Fig. 98 A), in the upper jaw, the paracone, and in the lower jaw the paraconid ; and the posterior cusplet is the metacone or metaconid, respectively. As the prin- KINETOGENESIS. 333 cipal cusps, or protocone and protoconid, alternate with each other, the cusplets stand opposite to them in the closing of the jaws, and a certain amount of interfer- ence results. As the lesser cusps are the less resistant to the wedging pressure of such contact, their position would change under its influence, rather than the large central cusps. The lower jaw fitting within the upper, the effect of the collision between the major cusps of the one jaw, and cusplets of the other, would be to emphasize the relation still more ; that is, the cusplets of the upper jaw would be wedged outwards, while those of the lower jaw would be pressed inwards, the major cusps retaining at first their original alternate position. With increase of the size of the teeth the cusps would soon assume in each jaw a position more or less transverse to that of the other jaw, producing, as a result of the crowding, a crown with‘a triangular section in both. The process may be rendered clear by the following diagram : _: 2Q@okR Wc OQ'‘on A ses BUAY B Fig. 98.—Diagrammatic representations of horizontal sections of tricuspi- ‘ date molars of both jaws in mutual relation; the shaded ones represent those of the upper jaw: A, Triconodon; &, Menacodon; C, ideal tritubercular mo- lars, approached by Menacodon, B. It is supposed on the contrary by Rése and Kiken- thal that mammalian molars which support more than one cusp have been formed by the fusion of several simple reptilian cones. So far as regards the higher Mammalia this hypothesis is in opposition to all the facts of paleontology and is not worthy of discussion. 334 PRIMARY FACTORS OF ORGANIC EVOLUTION. The only question that can arise is with reference to. ’ the origin of the multituberculate molar of the'Proto- theria. It is further questioned by Forsyth-Major, whether the tritubercular molar has been derived from the tri- conodont. He believes, on the contrary, that it is de- rived from the multitubercular type by reduction. There are two objections to this view: (1) the cones of the tritubercular tooth or trigon should be subequal, were they derived from a multitubercular source. On the contrary, the two external cones of the upper, and the two inner cones of the lower series are in the earliest (Jurassic), as well as most of the tritubercular types, smaller than the single opposite cusp or proto- cone, precisely as are the anterior and posterior cones of the triconodont molar. (2) No paleontologic series from the multitubercular to the tritubercular types has been traced, while the series from the.triconodont to the tritubercular is well known. Forsyth-Major’s evi- dence that such a transition exists in the Glires, is better explained by tracing the moderate complexity he describes to a tritubercular origin. It is also alleged by Allen and Scott that the inter- nal cusps of the premolars, when present, originate by the development of internal cingula, and have no prim- itive tritubercular ancestry. The evidence at our dis- posal from paleontological sources is in favor of this view; hence it is reasoned that the history of the mo- lar teeth must have been identical. This however does not follow, especially as the paleontologic evidence points the other way. The history of the two series has been different. In the first place the premolars have been subjected to much less use than the true molars; hence they retained the primitive reptilian KINETOGENESIS. 335 simplicity for a much longer period, a simplicity which they retain in the Carnivora, except the 1, which be- came the sectorial. Secondly, the premolars, instead of increasing in size, have in many types decreased ; the Diplarthra alone presenting an exception to this rule. That the internal cusps of the premolars may have arisen by growth of cingula in this order, is by no means improbable. We seem to have here an excel- lent illustration of the origin of two identical struc- tures by different evolutionary routes. The first modification of the tritubercular molar of the lower jaw is the addition of a low cingulum at the posterior base. This is seen in a rudimentary condi- tion in various living species of the Centetide and Chrysochloridz of the insectivorous order (Fig. 100); but in these existing forms the superior molar has added a posterior cingulum also, which widens inter- nally, or towards the palate (Fig. 101). In the evolu- tion of the dentition, the inferior posterior cingulum, or ‘‘heel,” was developed first, as in the Deltatherium, Centetes, and Stypolophus (Figs. 9g, 100, 102), where it is quite large; while the superior cingulum is want- ing in Stypolophus and Didelphodus, but is present in a very rudimentary condition in De/tatherium fundamt- nis. Inallof these genera the external cusps of the superior series have been pressed inwards, and more or less together, and are therefore removed in this re- spect from the primitive condition. The more primi- tive state of the superior cusps is seen in some species of Mioclenus, where, however, a posterior cingulum may be developed. The primitive type of tritubercular superior molar is that of Sarcothraustes, and in the same genus the inferior molar only differs from the primitive type in having a well-developed heel. Among 336 PRIMARY FACTORS OF ORGANIC EVOLUTION. recent Mammalia the carnivorous and insectivorous Marsupialia generally have the tritubercular lower mo- lar with heel. In the Chiroptera and many Insectivora the heel is largely developed, and supports two cusps, as it does in some Creodonta. eo780 Set tteey, . a ; mee, Fig. 99.—Deltatherium fundaminis Cope, fragmentary skull, two-thirds natural size; from the Puerco bed of New Mexico. a, 4, c, from one individ- ual; @, from a second animal; a, right side of cranium; 4, palate from be- low; c, mandible, part from above; d@, left ramus, outer side; from the Re- port of the U.S. Geol. Surv. Terrs., Vol. UI. From this point the evolution of the tritubercular molar must be considered from two standpoints. The first is the mechanical cause of the changes of its form ; and the second is the mechanical cause of its definite KINETOGENESIS, 337 location in a particular part of the jaw. For it has been already stated that in the evolution of the secto- rial dentition of the Carnivora, the number of molars and premolars has considerably diminished, while those that remain have become relatively much larger. In the tritubercular dentition the crowns proper of one jaw alternate with those of the other (Fig. 100); but when heels are added in either jaw, they will op- pose such part of the crowns of the teeth in the oppo- Fig. 100,—Centetes ecaudatus: A, skull, side seen obliquely from below; B, superior molars from below; C, inferior molars from above. site jaw as comes in contact with them when in use. The development of the heel in the inferior molars produced a type which is known as the tuberculosec- torial. This type characterizes the Creodonta and a few Carnivora. In the former there are generally three such teeth, in the latter but one. In the tuberculosectorial type of inferior molar the primitive tritubercular part of the crown (trigonid of Osborn) stands principally anterior to the posterior 338 PRIMARY FACTORS OF ORGANIC EVOLUTION. root of the tooth. It appears that the posterior root has been extended backwards, so as to occupy a posi- tion below the middle of the superior molar, while the tritubercular crown has been confined to the space be- tween the crowns of the superior molars. This would follow of necessity from the alternating action of the crowns of the opposite series, in connection with a general increase in size of the teeth. In the opening of prt Fig. 101.—Series of inferior molar crowns representing the transition from the simple (haplodont) to the quadritubercular, From Osborn. the jaws in a Creodont, the elevated portion of the in- ferior crown shears by its posterior face against the anterior face of the superior molar, thus restraining its extension posteriorly. The stimulus of use, however, develops a low extension posteriorly, or a heel, which covers the posterior root, and opposes in mastication the internal extremity or tubercle of the crown of the superior molar above it. Thus a molar element in KINETOGENESIS. 339 mastication is added to the sectorial in some Creodonta, and in Canide and Urside, etc., among Carnivora. This function predominates over that of the anterior triangle in the Lemuride. (Fig. 95.) I have already pointed out the successive modifica- tions of form which have resulted in the existing spe- cialized single inferior sectorial tooth of the Felide. They consist in the gradual obliteration of the poster- ior-internal cusp, and of the heel, and the enlargement of the external and anterior internal tubercles of the primitive triangle. The modification in the character of the dentition taken as a whole was shown to consist Fig, 102.—Stypholophus whitie Cope; diagram representing the apposition of the inferior and superior molars. The superior are in light, the inferior in heavy lines. The numbers represent the molars and premolars: C, canine; poc, protocone; fac, paracone; wc, metacone; POC, protoconid; PAC, para- conid ; C, metaconid ; 4c, hypocone; HC, hypoconid. in the reduction of the number of the teeth, including the sectorials, until in Felis, etc., we have almost the entire function of the molar series confined to a single large sectorial in each jaw. The genesis of the superior sectorial tooth has been explained as follows. In consequence of the fact that the lower canine tooth shuts anterior to the superior canine, the result of the enlargement of the diameters of those teeth will be to cause the crowns of the inferior teeth to be drawn from behind forwards against those of the superior teeth when the jaw is closed (Fig. 102). Thus a shearing motion would re- 340 PRIMARY FACTORS OF ORGANIC EVOLUTION. sult between the anterior external edge of the lower triangle and the posterior internal edge of the superior triangle. Now the characters of the true sectorial teeth consist in the enormous extension of these same edges in a fore and aft direction, the inferior shutting inside of the superior. To account for the development of these blades we must understand that the oblique pres- sure of the front edge of the lower tooth, on the hind edge of the superior tooth, has been continued for a very longtime. We must then observe that the inter- nal tubercle of the superior triangle has been pushed continually forwards and been reduced to a very small size. Why should this occur? Why should not the corresponding tubercles of the inner side of the lower crown have been pushed backwards, since action and reaction are equal? The reason is clear: The superior tubercle is on the internal apex of the trigon, and is supported by but one root, while the resistant portion of the inferior crown is the base of the trigonid, and is supported by two, thus offering twice the resistance to the pressure that the superior does. But why should the anterior part of the inferior tooth move forwards? even if it be in the direction of least resistance? This is due to the regular increase in size of the teeth them- selves, an increase which can be traced from the be- ginning to the end of the series. And this increase is the usual result of use (Fig. 102). The mechanics of the above proposition I believe to be correct, but I have had occasion to modify the statement as to the initiatory cause of the process. In many primitive Ungulata the canines have been as well developed as in the Carnivora, yet the forward pressure of the inferior molars on the superiors has not resulted, or has not been sufficient to produce sec- KINETOGENESIS. 341 torial molars in those types. In the Amblypoda, the lower molars even shear backwards on the upper ones. It seems then that this growth of the canines is not in all instances sufficient to cause a proterotome masti- cation. I suspect that the more usual cause is to be found in the voluntary effort of the primitive flesh- eater, to masticate flesh by the manipulation of his lower jaw and the body to be divided. The presence of the inferior canine forbids a posterior shearing move- ment of the molars, so that the anterior shear is the only one possible to most of the Creodonta. The ab- Fig. 103.—Cynodictis geismarianus Cope; skull one-half natural size: w, right side; 4, left side from below. sence of preglenoid crest in primitive Creodonta will permit a manipulation such as we observe in various ungulates to-day. The formation of a habit of a pro- terotome mastication would result, and the structural results would succeed as above pointed out. The excess of the forwards pressure of the inferior teeth against the superior over any backwards pres- sure, has left the posterior internal cusp of the triangle of the inferior molar (metaconid) without contact or consequent functional use. It has, consequently, grad- ually disappeared, having become small in the highest 342 PRIMARY FACTORS OF ORGANIC EVOLUTION. Canidz, and wanting in some Mustelide, and all Fe- lide. The heel of the same tooth has had a similar history. With the diminution in size of the first supe- rior tubercular, with which it comes in opposition in mastication, its functional stimulus also diminished ; and it disappeared sometimes a little sooner (Felide) and sometimes a little later (Hyznidz) than that tooth. The specialization of one tooth to the exclusion of others as a sectorial, appears to be due to the follow- ing causes. It is to be observed in the first place that when a carnivore devours a carcass, it cuts off masses with its sectorials, using them as shears. In so doing Fig. 104.—Aelurodon sevus Leidy; diagram representing coadaptation of crowns of superior and inferior molars in mastication; lines and lettering as in Fig. 102. it brings the part to be divided to the angle or canthus of the soft walls of the mouth, which is at the front of the masseter muscle. At this-point the greatest amount of force is gained, since the weight is thus brought immediately to the power, which would not be the case were the sectorial situated much in front of the mas- seter. On the other hand, the sectorial could not be situated farther back, since it would then be inacces- sible to a carcass or mass too large to be taken into the mouth. The position of the sectorial tooth being thus shown to be dependent on that of the masseter muscle, it re- mains to ascertain a probable cause for the relation of KINETOGENESIS. 343 the latter to the dental series in modern Carnivora. Why, for instance, were not the last molars modified into sectorial teeth in these animals, as in the extinct Hyznodon, and various Creodonta. The answer ob- viously is to be found in the development of the pre- hensile character of the canine teeth. It is probable that the gape of the mouth in the Hyznodons was very wide, since the masseter was situated relatively far posteriorly. In such an animal the anterior parts of the jaws with the canines had little prehensile power, as their form and anterior direction also indicates. They doubtless snapped rather than lacerated their enemies. The same habit is seen in the existing dogs, whose long jaws do not permit the lacerating power of the canines of the Felidz, though more effective in this respect than those of the Hyznodons. The use- fulness of a lever of the third kind depends on the ap- proximation of the power to the weight ; that is, in the present case, the more anterior the position of the masseter muscle, the more effective the canine teeth. Hence it appears that the relation of this muscle to the inferior dental series depended originally on the use of the canines as prehensile and lacerating organs, and that its relative insertion has advanced from be- hind forwards in the history of carnivorous types. Thus it is that the only accessible molars, the fourth above and the fifth below, have become specialised as sectorials, while the fifth, sixth, and seventh have, firstly, remained tubercular as in the dogs, or, sec- ondly, have been lost, as in hyenas and cats. The reduction of the number of molars in relation to the increase in the size of the canines commenced as early as the Jurassic period. It is seen in the gen- era Triconodon (Owen) and Paurodon (Marsh), where KINETOGENESIS. 345 \ the canines are large and the molars few. In the Plagiaulacide a similar relation is: seen between the development of the incisors and the reduction in num- ber of the molars. This is the modification of relation From Gaudry. Fig. 106.—Sections of crowns of molars of Ungulata from Dinotherium to Eiephas primigenius, A,B, Dinotherium; C, D, Mastodon; Z, /, G, Elephas. observed in existing Mammalia of the orders Probos- cidia and Glires, which will be mentioned later, under the head of proal dentition. e. Origin of the Dental Type of the Glres. The peculiarities of the rodent dentition consist, as is well known, in the great development of the incisors ; 346 PRIMARY FACTORS OF ORGANIC EVOLUTION. the loss of the canine and of all but one, or rarely of two, of the premolars, which leave a wide diastema ; and the posterior position of the molar teeth, as relates to the rest of the skull. A peculiarity which belongs to the highest types of the order is the prismatic form of the molars, and the deep inflection of their always transverse enamel folds both laterally and vertically. A peculiarity of the masticating apparatus, which is the basis of distinction from the bunotherian order, is the lack of postglenoid process, and the consequent freedom of the lower jaw to slide backwards and forwards in mastication. Appropriately to this motion the condyle of the mandible is either subglobular, or is extended anteroposteriorly, and the glenoid cavity is a longitudi- nal instead of a transverse groove. The mechanical action of the development of the rodent dentition has been as follows. The first factor in the order of time and importance was the increasing length of the incisor teeth. Those of the lower jaw closed behind those of the upper in the progenitors of the Glires (e. g. Psittacotherium) as in other Mamma- lia. Increase of length of these teethin both jaws would tend to keep the mouth permanently open, were it not for the possibility of slipping the lower jaw backwards as it closed on the upper. This backward pressure had undoubtedly existed, and has operated from the earliest beginning of the growth of the rodent incisors. The process has been precisely the opposite of that which has occurred to the Carnivora, where the pres- sure has been ever forwards owing to the development of the canines. The progressive lengthening of the incisors through use has been dwelt on by Professor Ryder (2 ¢.). The posterior pressure on the lower jaw, produced by its closing on the upper, has been KINE TOGENESIS. 347 increased directly as the increase in the length of the incisors, especially those of the lower jaw. The first effect of this posterior pressure will have been to slide the condyle of the mandible posteriorly be ae : ng ramus: c, ; e, foramen infraorbitale; from Hall and Wyman. over the postglenoid surface, if any were present, as is probable, in the bunotherian ancestor of the rodent. Continued repetition of the movement would probably push the process backwards so as to render it ineffec- tive as a line of resistance, and ultimately to flatten it 348 PRIMARY FACTORS OF ORGANIC EVOLUTION. out against the otic bulla, and atrophy it. The lower jaw would thus come to occupy that peculiarly pos- terior position which it does in all rodents. The anteroposterior (proal!) type of mastication becoming necessary, an appropriate development of the muscles moving the lower jaw, with their inser- tions, follows, parz passu. As a result we see that the insertion of the temporal muscle creeps forward on the ramus, until in the highest rodents (Cavia) it extends along the ramus to opposite the first true molars. The office of this muscle is to draw the ramus backwards and upwards, a movement which is commenced so soon as the inferior incisor strikes the apex of the su- perior incisor on the posterior side. By this muscle the inferior molars are drawn posteriorly and in close apposition to the superior molars. Connected with this movement, probably as an effect, we find the co- ronoid process of the mandible to have become grad- ually reduced in size to complete disappearance in some of the genera, e. g. of Leporidz. In these gen- era the groove-like insertion of the temporal muscle develops as the coronoid process disappears. As third and fourth effects of the posterior position of the lower jaw, we have the development of the in- ternal pterygoid and masseter muscles and their inser- tions and origins. The angle of the ramus being forced backwards, these muscles are gradually stretched back- wards at their insertions, and their contraction be- comes more anteroposterior in direction than before. The internal pterygoid becomes especially developed, and its point of origin, the pterygoid fossa, becomes much enlarged. The border of the angle of the man- dible becomes more or less inflected. In their effect 1Page 318. KINETOGENESIS. 349 on the movements of the ramus they oppose that of the temporal muscle, since they draw the ramus for- wards. They are the effective muscles in the use of the incisor teeth ; that is, in the opposition of the in- ferior incisors against the superior from below and posteriorly. Hence the great development of the in- ternal pterygoid, and, in a less degree, of the masse- ter. Both muscles tend also to close the jaws, but at a different point in the act of mastication from that at which the temporal acts. If we suppose the mouth to be open, the action of the masseter and internal pterygoid muscles draws the mandible forwards and upwards until the incisors have performed their office, or the molars are in contact with each other or with the food. They then relax, and their temporal muscle continues the upward pressure, but draws the ramus backwards to the limit set by the adjacent parts, caus- ing the act of mastication. A fifth effect of the development of the incisors and of the proal mastication, is seen in the position of the molar teeth. The indefinitely repeated strain and pressure applied to the superior molars from forwards and below has evidently caused a gradual extension of the maxillary bone backwards, so that the last molars occupy a position much posterior to that which they do in other orders of mammals. This is especially the case in such forms as Bathyergus, Arvicola, and Cas- toroides (Figs. 107-108), where the last molars are be- low the temporal fossa, and posterior to the orbit. A sixth effect of the causes mentioned has been re-: ferred to by Ryder.!. This is the oblique direction of the axes of the molar teeth. These directions are op- posite in the two jaws ; upwards and forwards for the 1 Proceedings Philadelphia Academy, 1877, Pp. 314. 350 PRIMARY FACTORS OF ORGANIC EVOLUTION. lower, and downwards and batkwards for the upper. The mechanics of this change of direction from verti- cal in the primitive forms (Sciuridz) to oblique in the Fig. 108 —Castoroides ohioensis Foster ; two-thirds natural size; skull from below. a, incisine foramen; 4, pterygoid fossa; c, internal pterygoid plates; d, fossa in basioccipital bone; ¢, external auditory meatus; /, mastoid pro- cess; g, occipital condyles; 4, tympania bulla, after Hall and Wyman. genera with prismatic molars, is simple enough. The inferior crowns when closely appressed to the supe- rior, and drawn posteriorly in the direction of the long axis of the jaw, press and strain the teeth in the two KINETOGENESIS, 351 directions mentioned. The development of the long prismatic crowns which has proceeded under these circumstances, has been undoubtedly affected by the pressure and strain, and the direction we find has been the result. The seventh effect is in the detailed structure of the teeth themselves. Beginning with short crowns with simple transverse crests (Psittacotherium and Sci- uride, Figs. 106, 109), we reach through inter- mediate forms, crowns with vertical lamine of enamel, which some- times divide the crown entirely across (Chinchil- lide, Caviide, Castoroid- ide) or appear only on the side of the crown, through the - continued coalescence of the prisms of which each molar crown is composed (Arvi- Fig. 109.—Jschyromys typus Leidy, from 3 the White River beds of Colorado; orig- cola). In many instances inal; from the Report U.S. Geol. Surv Terrs.: b, cranium from below; @, man- the crowns increase iN 4:10 from above. transverse at the expense of their longitudinal diameter (Castor, Lepus). The vertically laminated structure is evidently due to the crowding together of transverse crests by the same pressure which has given the crowns their oblique di- rection. In many genera the lengthening of the crown has included the lengthening of the longitudinal con- nection between the transverse crests, as in Arvicola, Castor, and Hystricide generally. In others this con- nection has not been continued, so that the crown is ‘ 352 PRIMARY FACTORS OF ORGANIC EVOLUTION. composed of prisms which are separate to near the base, as in Amblyrhiza and Geomyidz. In others, con- nection between the prisms has been lost by cenogeny, as in Chinchillide and Caviide generally. The latter families display also the greatest amount of crowding. Vv. DISUSE IN MAMMALIA. Modifications of structure of the mammalian skel- eton accompany the disuse of parts, no less distinctly than in other divisions of animals. That these modi- fications are the direct consequences of this disuse may be reasonably inferred as the antithesis to the thesis of development of structure through use, main- tained in the preceding pages. The evidence is more convincing from the fact that the same structures are observed to be related to similar dynamic conditions in groups of different taxonomic position. I select four illustrations from the Mammalia, from types in which the phylogeny is known, so that there is no question as to the degeneracy of the parts described. a. Natatory Limbs. The limbs have undergone great modifications of form in their gradual adaptation to aquatic habits. The stages of this process are to be observed first in the sea-otter (Enhydra), then in the seals, then'in the sirenians, and last in the Cetacea. This succession of groups is not given here as a phylogeny, for paleon- tology does not warrant any such history, but the phy- logeny of the limbs has been similar in the order of succession. The use of a limb as an oar for propulsion in the water requires that it shall be, so far as the blade is concerned, inflexible. Such a structure has existed in KINETOGENESIS. 353 all thoroughly aquatic Vertebrata. This implies the immobility of the articulations, which is due to the loss of their condylar surfaces. This may be traced to disuse of such articulations. This disuse would be at first voluntary, the limb being held stiffly while used as an oar in the act of swimming. Loss of power of extension and flexion. is well known to result from disuse. It is well known that the flex- ors and extensors of the manus have become atrophied in the Cetacea. Not so, however, with the flexors and extensors of the humerus, which become those of the entire limb. In the whales the first segment of the fore limb is enclosed within the integument of the body, so that its motion being much restricted, the insertional crests are reduced in size. In the eared seals (Otariidz) the hind limbs are somewhat free from the body integument, so that they can be turned for- Fig. 110,.—Balena mysticetus ward when on land. They are oo from Cuvier, Oss, Fos further enclosed in the true seals (Phocidz) so that their motion is very slight and they cannot be used for progression on land, and are available only for swimming. b. Abortion of Phalanges in Ungulata. In the heavy Ungulata the longitudinal diameter of the phalanges is greatly reduced in relation to their 354 PRIMARY FACTORS OF ORGANIC EVOLUTION. The successive increase in depression in the bones of the feet with the advance of time is to be transverse. a b ¢ Fig. 111.—a, Pantolambda bathmodon, digit of posterior foot. 4, Right posterior foot of a species of Cory- phodon from New Mexico, one-half natural size. From Cope, Report Expl. W. of rooth Mer., G. M. Wheeler, IV., Pl. LIX. c, Zobastleus mirabilis, right posterior foot; from Marsh, Dinocerata. most readily seen in the order Amblypoda, where we pass from Pantolambda to Coryphodon and Uintathe- KINETOGENESIS. 355 rium (Fig. 111.) A similar successive reduction is to be seen in the lines of the Perrisodactyla, as we pass from the smaller and lighter to the heavier and more bulky types. Such series are those which commence in the Lophiodontidz, and terminate in the Menodontidz on the one hand, and the rhinoceroses on the other. The elephants display the end of a similar line, which com- mences in the Condylarthra. In all of these bulky mammals the weight in progression is borne on the extremities of the metapodial bones, and the phalanges take but little share in it. They are turned forwards and are nearly useless. Their great reduction in di- mensions in these forms appears to me to have fol- lowed disuse, and this is then the probable cause of it. c. Atrophy of the Uina and Fibula. Successive atrophy of the ulna and fibula has been already referred to (p. 135). This is coextensive with reduction of the number of the digits in the ungulate Mammalia, and with the development of the digital patagium in the bats. This is in broad contrast to the subequal development of the ulna and radius in the Cetacea, where the fore limb functions as the blade of anoar. The cause of the reduction of the two ele- ments in the Ungulata is the restriction of the func- tions of the fore and hind limb to the radius and tibia respectively. The distal extremities of the ulna and fibula are supported by the external bones of the carpal and tarsal series respectively. The reduction of the external digit deprives the external bones in question of their share in the support of the general weight, and consequently relieves them of the impact which passes through the longer median digits which remain. The median digits, on the other hand, support the radius 356 PRIMARY FACTORS OF ORGANIC EVOLUTION. and tibia through the medium of the carpus and tar- sus, and it is these elements therefore which function in the use of the limb. We have here an evident illustration of. the effect of disuse in effecting the atrophy of an element, and of use in increasing the size and complexity of an adjacent element, of the same organism. No other explanation seems possible, for the elements which are reduced and those which are enlarged, are subjected in every other respect to the same conditions. ad. Atrophy of Incisor Teeth. This is complete in both jaws of existing Edentata ; the upper jaw of Dinocerata and many Artiodactyla, and is partial in the upper jaw in various Chiroptera and Lemuride. We have already seen (p. 326) that the superior incisors of certain Lemuride are without utility, owing to the conversion of the inferior incisors into a horizontal comb. I have ascribed the reduction of the superior incisors of bats to disuse consequent on the adoption of a frugivorous diet.1_ Further reason, which is common to the living members of the orders mentioned, is to be found in the disuse which has fol- lowed the use of the tongue as an organ for the pre- hension of food. The fruit-eating bats with most re- duced incisors (Glossophaginz) carry the soft parts of fruits into the mouth with the tongue. The Edentata use the tongue for the collection of both insect and vegetable food, projecting it far exterior to the mouth. The Artiodactyla without superior incisors however, combine the prehensile use of the tongue with a use of the lower incisors, which bite off the grass thus seized, 1Mechan. Origin, etc., Mammalia, 1889, p. 224. See also Dr, H. Allen, Proceeds, Academy, Philadelphia, 1891, p. 451. KINETOGENESIS, 357 while it is pressed against the pad which replaces the superior incisors. Why the superior incisors should have disappeared in this group is not yet clear to my mind. In this connection Dr. Allen (/. c.) reminds us that in hypertrophy of the tongue in man, the inferior in- cisors are thrown forward and are widely separated from each other. He considers it reasonable to infer that in lower animals where the tongue is used for pre- hension, the similar change which takes place in the teeth, from a vertical to a horizontal position, is induced by this cause. Vi. HOMOPLASSY IN MAMMALIA. The direct evidence in favor of kinetogenesis above adduced is greatly strengthened by corroborative tes- timony presented by distinct phyla of animals. Re- stricting myself here to Mammalia, I will enumerate a number of cases where the same structures have ap- peared in distinct lines of descent under similar me- chanical conditions, a phenomenon already referred to on page 72 under the name of Homoplassy. Before reviewing the subject, I cite what is the most remarkable example of homoplassy in the Mam- malia which has yet come to the knowledge of paleon- tologists. Ameghino has discovered in the Cenozoic formations of Argentina a group of Ungulata which he calls the Litopterna, and which I regard as a suborder of the Taxeopoda, allied to the Condylarthra (p. 128). Ameghino placed the group under the Perissodactyla, but the tarsus and carpus are of a totally different char- acter, and indicate an origin from the Condylarthra quite independently of that division. The carpal and tarsal bones are in linear series, or if they overlap, it is in 358 PRIMARY FACTORS OF ORGANIC EVOLUTION. Fig. 112.Feet of Proterotheriide, from Ameghino. 4, fore foot of Proterotherium cavum Ameghino; B, C, fore and hind feet of Diadiaphorus majusculus Ameghino ; D. crepidatum Ameghino, Much reduced. £, fore and hind feet of Thoatherium KINETOGENESIS. 359 a direction the opposite of that which characterizes the order Diplarthra (= Perissodactyla and Artiodactyla). But the Litopterna present a most remarkable paral- lelism to the Perissodactyla in the characters of both the feet and the dentition. No genus is known as yet which possesses more than three toes before and be- hind, and these are of equal length in Macrauchenia Owen. In this genus the teeth are not primitive but are much modified. The most primitive dentition is seen in the genus Proterotherium (Ameghino) where the superior molars are tritubercular as in many Con- dylarthra. In this genus (Fig. 112, 4), there are three toes, but the lateral ones are reduced, about as in the Equine genus Anchitherium (p. 148). In the next genus, Diadiaphorus Ameghino, the superior molars are quadritubercular and crested, while the lateral toes are reduced still more, being quite rudimental (Fig. 112, B, C), as in the equine genera Hippotherium and Prothippus (p. 149; Fig. 70). The superior molars have not progressed so far as in these genera, but are not very different from those of Anchitherium. Inthe third and last type (Thoatherium Ameghino), the lat- eral digits have disappeared from both fore and hind feet (Fig. 112, C, D), so that the condition is that of the genus Equus (Fig. 81), but the splints in the 7hoa- therium crepidatum Ameghino are even more reduced than in the known species of horse. The superior molars have not assumed the pattern of the genus Equus, but resemble rather those of Macrauchenia, and could have been easily derived from those of Dia- diaphorus. Here we have a serial reduction of the lateral digits and their connections with the leg, and increase in the proportions of the middle digit and corresponding in- 360 PRIMARY FACTORS OF ORGANIC EVOLUTION. crease in the proximal connections, exactly similar to that which took place in the horse-line, in a different order of Mammalia. In review I now cite as examples of homoplassy: First, as regards the development of the tongue- and-groove ankle-joint. This has been developed in- dependently along four distinct phyla, viz., in the lepo- rid Glires, the Carnivora, and the even and odd toed Diplarthra. Second, the wrist-joint. The faceting of the radial surface has appeared independently in the perissodac- tyle and artiodactyle lines, but is best developed in the latter. Also it appeared independently in the separate suoid and bo@id lines in the latter suborder. Third, the trochlear crest of the elbow-joint ap- peared independently in the perissodactyle and artio- dactyle Diplarthra, and in the leporid Glires (the rab- bit family). Fourth, the round head of the radius appeared in- dependently in the lines of the Edentata (ant-eater) and Quadrumana, under the stress of supination of the hand. Fifth, the development of cusps with crescentic sec- tion out of cusps with round section has occurred in the widely different groups of the multituberculate Prototheria, and the selenodont Artiodactyla. In the former the crescents are transverse, since the thrust of the teeth in use is longitudinal; in the latter they are longitudinal, since the thrust of the jaws is transverse. Sixth, the deep plication and hypsodénty of molars appeared independently in the Glires, Tiliodonta, Pro- boscidia, Sirenia, Perissodactyla, and Artiodactyla; and probably in the Edentata and Toxodontia. Seventh, increase in the length of the legs has en- KINETOGENESIS. 361 sued in the Marsupialia, Glires (Lepus, Dolichotis, Dipus), Carnivora, Ungulata, Quadrumana. : Eighth, reduction of digits has occurred under sim- ilar conditions in Marsupialia, Glires, Insectivora, Carnivora, Ungulata. Ninth, the atrophy of the ulna and fibula occur in the distinct lines of the Perissodactyla and Artiodac- tyla, and the atrophy of the fibula in the leporid Glires ; all in limbs which function in the most rapid progression. Further confirmation of the law of kinetogenesis is to be found in those cases where different structures appear in different parts of the skeleton of the same individual animal, in direct correspondence with the different mechanical conditions to which these parts have been subjected. Examples: the diverse modifi- cations of the articulations of the limbs in consequence of the uses to which they have been put, in mammals which excavate the earth with one pair of limbs only; as in the anterior limbs of the fossorial Edentata, In- sectivora, and Glires. The reduction of the number of the digits in the posterior limb only when this is extensively used for rapid progression, as in leaping: this is seen in the kangaroo and jerboas, in the orders Marsupialia and Glires. The development of a dental structure of premolars identical with that of the molars, from a different struc- tural origin, in the Perissodactyla. From the preceding facts I have inferred that in biologic evolution, as in ordinary mechanics, édentical causes produce tdentical results. 362 PRIMARY FACTORS OF ORGANIC EVOLUTION. vil.A HYPOTHESIS OF THE ORIGIN OF THE DIVISIONS OF THE VERTEBRATA. In order to estimate the part which has been played by the movements of the Vertebrata in chang- ing their environment in past geologic ages, we have to rely principally on inferences derived from the present physical characteristics of the earth. Formerly, as now, conditions of temperature, humidity, soil, shel- ter, food, etc., were avoided or appropriated by ani- mals, through their capacity for moving from place to place. What concerns us chiefly here, is the effects on their structure produced by the movements of Ver- tebrata. In examining this question I will take it up in systematic order, so as to observe whether kineto- genesis has been the principal or only a subordinate agency in the evolution of this branch of the animal kingdom. The most conspicuous index of the serial succes- sion of the vertebrate classes, is, as has been already remarked, the circulatory system. The modifications of this system have been immediately connected with those of the method of respiration, which the exigen- cies of the environment induced in vertebrates. The existence of branchial arteries and veins dates from the earliest vertebrate, if not from prevertebrate life. They are already established in the Tunicata, and con- tinued throughout the rising scale in diminished num- bers, so long as Vertebrata were exclusively aquatic in their modes of life. When at the close of the De- vonian system the land masses assumed great propor- tions in both the Eastern and Western Hemispheres, it is probable that many fishes were entangled in shal- KINETOGENESIS. 363 low water, which rapidly freshened, and ultimately were desiccated, and respiration by the swallowing of air into the alimentary canal began to take the place of respiration by gills. It is well known that respira- tion by this means may be carried on by fishes of va- rious genera, e. g. Cobitis; and Professor Gage has shown that the same habit exists in Batrachia and in certain tortoises (Tronychide). In the middle Car- boniferous shales tracks of land animals occur, and the bones of Batrachia abound in the coal measures. Already in the Permian these Batrachia are accompa- nied by numerous Reptilia, and air breathers of ter- restrial habits had become numerous on the earth. The habit of holding in the esophagus large quan- tities of air while engaged in seeking food in foul water, or on land, on the part of vertebrates which normally oxygenated the blood by means of gills, was probably the mechanical cause of the development of a pouch, and afterwards of a diverticulum of the cesophagus, which became ultimately a swim-bladder oralung. In vertebrates in which a return to aquatic life became necessary, it became the former; in those which remained for a shorter or longer period of time on land it became the latter.! It is noteworthy that among fresh-water fishes generally, the swim-bladder is more complex than among marine forms, showing’ that the varying conditions of shore and fresh-water life have been mainly responsible for its development. The development of a lung at once produced a change in the uses to which the various branchial arches were put. The posterior, which supply the lung, would be subjected to greater pressure owing to the increased blood supply demanded by the lung, 1This view is adopted by C. Morris, American Naturalist, 1892, p. 975. 364 PRIMARY FACTORS OF ORGANIC EVOLUTION, and a correspondingly diminished pressure would be experienced by the now unused branchial portions of the bows. The first would retain the importance of its basal portion, as the source of the carotids, while the middle arches would continue their existence as the bases of the central dorsal aorta. The loss of the right aorta-root in Mammalia was probably due to the fact that the great arteries which supply the digestive sys- tem are primitively branches of the left aorta-root, as they are to-day in the crocodiles and in many of the Batrachia. The right aorta-root disappeared through disuse. Probably in the immediate ancestors of the birds, as in the crocodiles, the right aorta-root gave off the carotides and the subclaviz. As the birds de- mand an excessive blood-supply for the fore limbs, we have here probably the reason why the right root re- mained in this subclass. The next index of successional development in Ver- tebrata is the brain. Our belief that use under stimu- lus has been the cause of its successive growth, can only be based on the analogy of our own experiences in the matter of education. No part of the human organism is so susceptible to stimuli as the nervous system, and the marvellous effects on faculty of con- tinued exercise are well known to everybody. Since the changes of mental states are necessarily due to corresponding structural changes no one will find in ignorance of the mechanics of brain-evolution a serious obstacle to believing that it has taken place under the influence of the innumerable stimuli always present to animal life. It is in the skeleton that we have the actual record and evidence of the effect of movement on structure. It must be remembered in this connection that skeletal KINETOGENESIS. 365 and dental tissues exhibit the phenomena of nutrition and waste (metabolism), common to all living organic matter. Hence even the hardest osseous tissues are plastic and are subject to mechanical influences to a degree which is not possible to dead matter of equal density. Fundamental differences between Vertebrata are displayed by their organs of movement, but before specially considering these I will refer briefly to cer- tain other fundamental characters displayed by the skull. In advancing from the fishes to the Mammalia we observe a successive consolidation of the mandibu- lar arch, and of its mode of connection with the cra- nium. The mandibular arch in its entirety displays in the fishes a segmented condition, generally compar- able to that which characterizes the branchial arches. Among Batrachians and Reptilia various degrees oi fixation of its suspensor (hyomandibular, quadrate) to the cranium exist, and in some of them it is closely united by immovable suture. The complete fusion with the squamosal seen in the Mammalia is its final status. The segmentation of the mandibular portion of the arch seems, from the discoveries of Ameghino, to have continued among some of’ the Lower Eocene mammals, but that finally disappeared, so that in the modern mammals the movable mandibular arch con- sists of a single element on each side. In this history we see an instance of the progressive codssification of parts, which results from the constant strain of use, of which many other normal and abnormal examples are known. This use is the act of mastication. Where there is no mastication, and the jaws are used only as prehensile organs, this codssification does not occur, as, for instance, in the snakes. In this most special- 366 PRIMARY FACTORS OF ORGANIC EVOLUTION. ized and modern type of Reptilia, the segmentation is complete. The segmentation of the limbs in the Vertebrata is a simple mechanical problem. Paleontology and em- bryology concur in proving that the limbs originated in primitive folds in the external integument, and that their connection with the internal skeleton was of later accomplishment, has been shown by Wiedersheim. At first free, they sought points of support on the skeleton, but did not lose their free mobility when this contact was attained. Appropriately to the me- chanical conditions of rigidity and flexibility neces- sary to their use in a fluid medium, they were orig- inally composed of slender rods which were segmented by interruptions at suitable points. The articulations of the fin-rays of fishes have been made the subject of an interesting research by Ryder, who finds them to be fractures, due to flexures during motion in the water medium.! The limb of land vertebrates (the chiropterygium) was derived from one of the forms of fins (rhipidopterygium) of water vertebrates. This is the simple type of primitive fin displayed by the Pale- ozoic Teleostomi of the superorder Rhipidopterygia. Whether the subdivisions of the chiropterygium, the propodial, metapodial, and phalangeal bones, etc., were divided from the primitive branches of the archi- pterygium, as held by Gegenbaur, or whether they have developed by sprouting from a simple axial series of segments, as held by Baur, or whether, as I have suggested, it is a derivation from the rhipidopterygian type of paired fin, is not yet decided. In either case, the limbs of the first land animals were segmented and flexible at the joints between the segments. The ne- 1 Proceedings of the American Philosophical Society, 1889, p. 547. & KINETOGENESIS. 367 cessities of such limbs are twofold: first, to serve as supports when at rest or in progression ; second, to be applied to the body in protection from enemies, or in aiding the functions of feeding, reproduction, etc. The first function requires principally mobility at the point of connection with the body. The second, flexi- bility at some point on the shaft of the limb. The two kinds of movements in question would conserve two principal points of flexure, and these would be for the fore limb, just what we find, the shoulder and elbow joints; and for the hind limbs, the hip and knee joints. The two median joints are directed in opposite ways, the elbow backwards and the knee forwards. This diversity is clearly due to the diverse positions of the functioning regions. The opposite extremities of the alimentary canal, the posterior including the exits of the urogenital organs, requires that the fore limbs should bend forwards, and the posterior limbs back- wards. And the constantly recurring necessity for the exercise of these flexures must necessarily have devel- oped the appropriate articulations in preference to all others. The terminal flexure, that of the wrist or ankle, has been evidently due to a similar mechanical cause, viz., the flexure due to pressure of the weight of the body on the terminal segments when in contact with earth. The distal segments are the most slender in all types, and least able to maintain a linear direc- tion under pressure, hence, they have flexed easily and thus the line of separation between leg and foot had its origin. The ankle and wrist in the Batrachia Uro- dela is still a mere flexure. Mr. Herbert Spencer has endeavored to account for the origin of the segmentation of muscles into myo- tomes, and the division of the sheath of the notochord 368 PRIMARY FACTORS OF ORGANIC EVOLUTION. into vertebre, by supposing it to be due to the lateral swimming movements of the fishes, which first exhibit these structures.! With this view various later authors have agreed, and IJ have offered some additional evi- dence of the soundness of this position with respect to the vertebral axis of Batrachia,? and the origin of limb articulations.’ It is true that the origin of segmenta- tion in the vertebral column of the true fishes and the Batrachia turns out to have been less simple in its pro- cess than was suggested by Mr. Spencer, but his gen- eral principle holds good, now that paleontology has cleared up the subject. The Echinodermata, Mollusca, Arthropoda, and Vertebrata possess external or internal calcareous or chitinous skeletons for the most part. The lower forms of all these branches, however, are more or less deficient in this kind of protection, and embryology indicates that all of them are the descendants of the Vermes or worms, which are mostly without such hard supports and protections. Whether this be demonstrated or not, we have plenty of evidence to show that the prim- itive Vertebrata were without hard skeletons, and that their bodies were composed internally and externally of perfectly flexible tissues. If we now imagine that either the integuments, or an axial rod, of a worm-like animal has become the seat of a calcareous or chitinous deposit, it is evident that the movements of the animal in swimming or creeping must have interrupted the deposit at definite points of its length. The lateral flexure of the body would be restricted to certain points, and the intervening spaces 1 Principles of Biology, 1873, pp. 198-204. 2 Origin of the Fittest, 1887, p. 305. 3 Mechanical Causes of Origin of Hard Parts of Mammalia, 1889, p. 163. KINETOGENESIS. 369 would become the seat of the deposit. At the lines of interruption joints would be formed, and if the move- ments were habitually symmetrical, these interruptions would be equidistant. In this way the well-known segmentation of the external skeletons of Arthropoda, and the internal skeletons of Vertebrata would be formed. We have more detailed evidence that this has been the case. Thus the segmentation of the os- seous sheath of the chorda dorsalis in both primitive fishes and batrachians has been accomplished in wedge- shaped tracts precisely as may be observed in the fold- ing of a tolerably stiff sleeve of a coat which ensheathes the arm, under the influence of lateral flexures. The wedge-shaped tracts are superior and inferior, the apices directed towards each other. Seen from the side they form two wedges with their apices together, and their bases one up and the other down. Now, if a person who wears a coat of rather thick material will examine the folds of his sleeve as they are produced on the inner side of his arm, he will see a figure nearly like that of the segments of the vertebral column de- scribed. The folds will correspond to the sutures, and the interspaces to the bony segments. He will find that the spaces are lens-shaped, or, when viewed in profile, wedge-shaped, with the apices together. This arrangement results from the necessary mechanics of flexure to one side. In flexure of a cylinder like the sleeve, or like a vertebral column, the shortest curve is along the line of the greatest convexity of the cylin- der. Here is the closest folding of the sheath, and here, consequently, the lines of fold in soft material, or interruption in hard material, will converge and come together. That is just what they do in both the 370 PRIMARY FACTORS OF ORGANIC EVOLUTION. sleeve and the rhachitomous vertebral column, the only difference being that in the animal it is exhib- ‘zzz2svSv snaazgoyvy J “qii‘« ‘stsXydoderp ‘py ‘umajusor1ajur ‘7 2 fumsyueo ‘97g ‘9 tstshydodeimou 'w ‘ourds yeineu ‘¥> *sntpogoxr2ue sngoukyjngy ‘Dd fsoysy snop{puodsozsm JO eiqa}19A [esIop Jo suorsog—EIr ‘31d *‘[PHIZ Wory ‘sngvoanf SNANJDD ‘VY ited on both sides, and on the sleeve on only one side. This difference is, of course, due to the fact that the animal can bend himself in both di- rections, while the arm only bends in one direction. It results from the above obser- vations that the structure of the rhachitomous ver- tebral column Aas been produced by the movements of the body from side to side, as in swim- ming, during the process of the de- posit of mineral ma- tertalin and around the chorda dorsa- lis.? , See figures 113 to 114@ where the coat-sleeve is com- pared with the ‘“‘rhachitomous” vertebrz of primitive 1See American Naturalist, January, 1884. e 2 ‘stsAqdod -eineu ‘w pue ‘wnIyUa.0IMe]d Jo UINAyUAD ‘¢ | IN=}U9019}UL syuasaideiz ‘snyL *¥1qa}10A SHOMIO}IWOeTI & JO S}UsTAdES eq} 0} JepIMs seoedsiazur saavay Yor ainxey [esoze] Aq peonpoid spjoy Surmoys ‘Wwoo e yo eAda[S—‘wbII ‘sty ‘sexay yo qooda reTunl9g ay} WoIg “Mojaq Wors ‘g ‘eTgoid ‘py ‘azis [enyeu YyyINOJ-2u0 ‘adap snyoygarnsam sgohag JO UUIN[OD [¥1qe}10A showoiqoeqy— P11 “31g 372 PRIMARY FACTORS OF ORGANIC EVOLUTION, fishes (Merospondyli1) and Batrachia (Rhachitomi). Such was the origin of the segmentation in the primi- tive sharks, Pleuracanthus, whose structure has been pointed out by Sauvage, and Hybodus, whose charac- ters have been demonstrated by Smith Woodward. The segmented (or rhachitomous, as I have termed it,) condition may be then regarded as the primitive one of the osseous column in the Vertebrata. From the rha- chitomous column two divergent lines have arisen as already remarked (pp. 89,209). The inferior segment has been retained in the fish-batrachian line, whence I have termed their vertebre ‘‘intercentral,” while these bodies have disappeared or become rudimental in the higher Vertebrata.. The pleurocentra (Figs. 113-1142, c, pl.) have, on the other hand, developed downwards, and, mesting below, have formed the effective centrum of the vertebra. Hence, in the Monocondylia and Mammalia the vertebre are ‘‘central.” The Reptilia display a greater variety of vertebral articulation than any of the classes of Vertebrata. After the primitive biconcave (amphiccelous) type was abandoned, the two principal types assumed are the ball and socket (proccelous and opisthoccelous), and the plane (amphiplatyan). In those families in which the body is more or less in contact with the ground, owing to the absence, shortness, or position of the limbs (Lacertilia, Ophidia), the vertebral bodies ex- hibit the ball-and-socket articulation, while in types with longer limbs which supported the body in pro- gression, so that the latter never reached the ground (Dinosauria), the articulations are plane. The ball-and- socket articulation may be inferred to have been pro- 1Zittel, Handbuch der Paleontologie, I11., p. 138, 1887, where this charac- ter is first clearly pointed out in fishes. KINETOGENESIS. 373 ‘ duced by vermiform movements which utilize points of resistance on the earth as aids to progression, while the plane articulation has probably resulted from the per- sistence of the fixed relation which is appropriate to a body which should be relieved by the legs of all share in movements necessary to progression. That this position is correct is sustained by the fact that the cer- vical vertebrz of various reptiles and mammals which have plane dorsal vertebre have the ball-and-socket structure. This is probably due to the constant flex- ures to which that part of the column has been sub- jected, as compared with the fixity of the dorsal re- gion. Owing to the comparatively advanced state of our knowledge of the phylogeny of the Mammalia (Chap- ter II.), this class furnishes especial opportunities for the study of kinetogenetic evolution. But our knowl- edge is not yet sufficiently complete to enable us to account on mechanical grounds for the origin of all the characters which distinguish all its subdivisions. This being the case, I have not presented the subject in taxonomic order, but have contented myself with offering it in the order of evidential value. I have first described certain cases where the action of kinetogen- esis is self-evident. This has been followed by the presentation of cases where the evidence amounts to a high degree of probability. Referring now to the table of definitions of the or- ders of Mammalia on pages 127-128, I will go over the characters serzatim, and show how far our knowledge warrants us in giving a kinetogenetic explanation of . their origin. No mechanical cause can be at present assigned for the loss of the coracoid and episternal bones in the Eutheria. In the Eutheria the presence 374. PRIMARY FACTORS OF ORGANIC EVOLUTION. of the marsupial bones of the Didelphia is a survival, and it may be that their absence in the Monodelphia is due to disuse, on the withdrawal of strain on the abdominal walls which followed the abandonment of the oviparous habit of the Prototheria, and the young- bearing habit of the Marsupialia. The very consider- able weight actually borne by existing forms and prob- able weight carried by extinct forms will probably account for the development of these bones through strain on the prepubic cartilage. The perforate palate is a reptilian survival, and the closure of the fonta- nelles may have been due to the increased strain due to the increased energy of mastication which early en- sued, owing to the increased size and specialization of the molar teeth. Of the three divisions of the Mono- delphia, the Mutilata, Unguiculata, and Ungulata, the Unguiculata possess modifications of a character of ungual phalanges inherited from the Reptilia, while the other two groups have experienced still greater modification. The limbs of the Mutilata display the result of disuse as to the posterior ones, and special use as to the anterior, as I have already pointed out. The Sirenia display a less degree of modification than the Cetacea. The hoofs of the Ungulata may well have assumed their laterally expanded and transverse forms by the extreme pressure and impact on the earth, incident to their function as supports. The origins of most of the dental characters which characterize the orders of Mammalia have already been referred to mechanical causes, and have been already treated of. The same is true of tarsal and carpal characters, which are of so much importance among the Ungulata. The characters enumerated on page 139 as indicat- KINETOGENESIS. 375 ing progressive modification in time have also been mainly accounted for on mechanical grounds. 5. OBJECTIONS TO THE DOCTRINE OF KINETOGENESIS It has been objected that Neo-Lamarckians are self-contradictory and illogical in their defense of the doctrine of the development of structures by use, or by motion. It is asserted that they believe that stimuli of different kinds produce similar results, and that stimuli of the same kind may produce different re- sults. The charge that Neo-Lamarckians hold those views is correct, but it is not correct to suppose that they are illogical or self-contradictory. This criticism is one of those generalities which will not bear exam- ination, while the doctrine of kinetogenesis will bear examination. Thus ‘it has been experimentally shown that bone irritation will produce both bone deposit and bone ab- sorption, according to the degree of irritation. Mode- rate irritation produces deposit, and greater irritation produces absorption. Hence it is that both impact and strain, or pressure and stretching, will elongate a bone, by stimulating growth, if not excessive. We have the illustrations in the elongation of ligaments and cartilages and their ossification under stretching, and the shortening of both in absence of use, from which we may infer their lengthening under use. The continued lengthening of the limbs and teeth of the higher Mammalia, in the course of geologic time, is an illustration of the effect of continued impact and transverse strain; while the lengthening of the limb bones of the sloth, and of the tarsal bones of many bats, is a consequence of longitudinal strain. 376 PRIMARY FACTORS OF ORGANIC EVOLUTION. The differing directions of the elements entering into the articulations of the limbs of Mammalia may be cited in illustration of the supposed inconsistency of supposing them all to be the result of impact and strain. Thus most of the condyles are directed distad, but the heads of the humerus and femur, and the prox- imal surfaces of the carpus and tarsus are directed proximad (except the trochlear groove of the astraga- lus when present). So far as regards the distal ends of the radius and tibia (fore arm and leg) I have pointed out that here dense layers disposed longitudi- nally impinge on a similar layer disposed transversely, with the natural consequence the latter has yielded to the excess of impact so produced. The direct relation of the sculpture of the surfaces concerned, to the lengthening of the foot and increase of speed of the animal, and hence increase of force of impact, leads irresistibly to this conclusion. As regards the con- vexity of the heads of the humerus and femur, or rather the concavity of the corresponding surfaces of the scapula and acetabulum the explanation has been al- ready given. Henke was apparently the first to call attention to the fact that the concavity and convexity of the articular surfaces is directly related to the posi- tions of the insertions of the muscles which move them. He shows that a concave surface is developed at the extremity which is nearest to the muscular insertion, while the convex surface is developed on the extremity which is most remote from its muscular insertion. Thus is accounted for the apparently contradictory evidence of the limb articulations mentioned. In some cases at least, as those of the glenoid cavities of the scapula, ilium, and phalanges, the muscular insertions are so near to their borders, as to suggest that the KINETOGENESIS, 377 growth of the latter is due to a pulling strain on them, as well as to the greater mobility of the element which becomes convex, as supposed by Fick. It is claimed in the preceding pages, that impacts on the extremities of a bone or tooth, gradually in- crease its length. It may be hastily supposed that in this assumption I derive elongation of the shaft of a bone from the same stimulus which produces excava- tion and therefore abbreviation of its extremities. In the gross this charge is correct; but the position I have assumed is defensible, because in detail it is easy to perceive that effect of the use of a limb on an ar- ticular surface of a bone is quite distinct from that which it has on the shaft. At the articular faces we have discontinuity; and therefore friction ; in the shaft we have only the concussions produced by impact, to- gether with some torsion strain. That the former movement stimulates the development and activity of the osteoclasts has been shown by Koelliker; that the latter may stimulate the activity of osteoblasts is ren- dered highly probable from the facts of pathological anatomy. These show that a very slight modification of stimulus is sufficient to change the building cells into the absorbent cells and back again. For the same reason belief in the elongation of bones under stretch- ing strain may not be inconsistent with belief in an elongation under impacts. Cary makes specific objections against the kineto- genesis of the articulations of the mammalian skele- ton.! After a study of the carpus of the Eocene peris- sodactyle genus Palzosyops he concludes that the trapezoid bone is too small to express properly the di- rect result of purely mechanical causes. He says that lAmerican Journal of Morphology, 1892, p. 305. 378 PRIMARY FACTORS OF ORGANIC EVOLUTION in reaching this result he has applied geometrical meth- ods. ‘First, the volume of the bones was got at. Next the area of the bearing surfaces and their inclina- tion to the digits were measured. Then giving to the thrust of each metacarpal a value proportional to its volume, the distribution of that thrust can by resolu- tion and composition of forces, be traced through the foot, and the pressure on each surface and bone ap- proximately obtained.” Further than this the author does not explain how he reached the result that the trapezoid is too small. It is quite essential that this demonstration should be given if we are expected to accept his conclusion. An essential part of the prob- lem is, however, unnoticed by Mr. Cary; and that is the condition of the trapezoid in the reptilian ances- tors of the Mammalia. The phylogeny of an element must be known, since it furnishes the ‘‘ physical basis” of the problem. The fact is that the trapezium, trape- zoides, and the magnum owe their small size to their being the only carpal elements which have not been produced by the fusion of two or more primitive ele- ments of the batrachian and reptilian carpus. The trapezoides moreover occupies a place in a longitudinal series of three elements in the primitive carpus, while the trapezium forms one of a series of only two ele- ments. For similar reasons the cuneiforms are the smallest elements of the tarsus. Mr. Cary then proceeds to criticize the explana- tions offered by Professor Osborn and myself, in ac- counting for the origin of certain structures. He finds our explanations to be self-contradictory, and that we also contradict each other. Osborn has supposed that the conules of the molars are produced by friction of the molars of opposite series on each other. I have KINETOGENESIS. 379 expressed the opinion that the shear of the sectorial teeth of Carnivora was produced by lateral friction during vertical movement of the lower tooth on the upper. Ihave also asserted that the forms of facets of limb articulations are due to pressure. Mr. Cary sees here the attempt to explain the origin of totally different structures through identical mechanical pro- cesses, and believes that the attempt is a failure. Were the conditions of the problems alike, as Mr. Cary thinks them to be, he would have good reason for his opinion. But the conditions in the three cases are entirely different, and our author’s conclusion is due to neglect of the elementary facts of the proposi- tion. The development of conules at the points indicated by Professor Osborn, has been supposed by him to be due to friction between existing ridges of enamel which cross each other when in action, at the points in question. In the case of the development of the sectorial shear, the faces between which the shearing motion takes place are smooth, and without ridges or crests. Hence the entire surface receives a homo- geneous friction. In the third case, that of the foot articulations, there is no friction, but there is pressure which when abruptly applied in movement becomes impact. There is really no parity between the three cases. The author of this paper also thinks that the ex- planation of the elongation of bones through use of different kinds is not a permissible hypothesis. He cites my attempt to account for the elongation of the leg bones of higher mammals through impact-stimulus ; and of other limb bones of other mammals through stretching. But he does not prove that similar results 380 PRIMARY FACTORS OF ORGANIC EVOLUTION. may not flow from mechanical stresses applied in dif- ferent ways. I suppose that any mechanical stress which determines nutritive processes to a part, will in- crease its size, ceteris paribus; and the stretch as well as the impact has this effect. I have in fact shown, in the observations already cited (pp. 277-279), on the production of artificial elbow joints, that osseous de- posit is stimulated by pulling strain as well as by push- ing or impact. In concluding, Mr. Cary admits one of the two con- tentions of the Neo-Lamarckians in his two closing propositions. He says ‘‘ Plasticity of bone, using the word plasticity not in a physical sense merely, but to include absorption under pressure, will probably ac- count for much structure in the foot and elsewhere, especially the connection with the joints, and in the fields of variation and correlation.” In the second proposition he says that facts have been adduced by him which are inconsistent with the theory that the size of bones has been increased by the stimulus they receive, and with the theory that regions of growth are determined by regions of pressure and strain. ‘‘The testimony of the literature as to the latter point,” he says, ‘‘is conflicting.” I have shown that the supposed conflict is due to a misunderstanding on the part of the author of this paper. The proposition that pres- sure does not affect growth is in contradiction to the admission made by the author in his first proposition, where he admits that pressure determines structure ; for in such change of structure there is always growth. Finally Mr. Cary remarks ‘‘That race changes follow those produced in the individual life, or that they are directly caused by their mechanical surroundings, I do not think it has been satisfactorily shown.” The KINETOGENESIS. 381 fact that the characters of bone structure admitted by Mr. Cary to have had a mechanical origin appear in the young before birth, is evidence that race characters are produced, in other words, that they are inherited. Another objection proposed by Tomes, and quoted by Poulton and Wallace with approval, has reference to the kinetogenesis of teeth of Mammalia as described by Ryder and myself. Tomes asserts that it is quite im- possible that the crowns of the teeth could have been altered by mechanical impacts and strains, since their form is determined in the recesses of the dental grooves, entirely removed from all the mechanical influences which affect the external surfaces of the jaws. But the observations of Koelliker and others show that osteoclasts are as active in dental as in ordinary osse- ous tissue. It is altogether probable that the modifi- cations of dental structure have been produced by strain and friction under use in the adult, precisely as in the skeleton, and that the share that the unerupted crowns have in the process is that of inheritance only, as in the case of the skeleton. That teeth deposit den- tine as process of repair in adult mammals is well known, and this repair is in direct relation to use. That the effects of dental wear are inherited is proven by the fact cited by Tomes and Wallace. Another objection to the doctrine of kinetogenesis which has been made by some of the Neo-Darwinians is, that if growth under stimulus be true, how can it have limits, so long as the stimulus of use exists. “In other words, what is to prevent, in the case of the vertebrate skeleton, of an indefinite increase in the length of the legs, of the teeth, and of their cusps, etc. The answer to this objection will vary more or less with the part of the structure considered. In general, 382 PRIMARY FACTORS OF ORGANIC EVOLUTION, however, it may be assumed that stimulus is stress due to a want of harmony between an organism and its environment, and that kinetogenesis is the result of the effort of the organism to adjust itself. So soon as equilibrium is attained, the stress of stimulus ceases or is much reduced, and evolution in this direction ceases. Such equilibrium is attained when the mechan- ism of an animal is sufficient for the satisfaction of its needs. When this is the case severe exertion is no longer necessary, and a period of easy use follows which is sufficient to maintain the mechanism in work- ing condition. In the case where circumstances should become so favorable for the easy satisfaction of the necessities of life, as to call for little or no exertion, degeneracy of the organism is sure to follow. The well-known phenomena of degeneracy from disuse show that a large part, and in some cases all, of the stimulus of use, is only sufficient for the maintenance of the organism in working condition, and that there is no surplus to be expended in progressive evolution. It is however true that some organs are stimulated to excessive growth by active use. Such are some of the teeth, which, if not worn at the crown by the op- position of those of the opposite jaw, soon grow to an inconvenient length. This occurs in the hypsodont molars of horses and artiodactyles, and in the pris- matic molars and incisors of Glires. Hypsodonty in general is an illustration of continuous growth induced by long-continued stimulus in those orders of Mam- malia, and in the Toxodontia and Proboscidia. The excessive growth of the canines in the South American saber-tooth tiger, and of the incisors of the mammoth, are cases where the energy of growth has not subsided in time to prevent excess. KINETOGENESIS. 383 Several years ago Prof. August Miiller and I called attention nearly simultaneously to the probability that many of the forms of the reproductive organs of plants, especially the flowers, are due to the strains and other effects produced by their use by insects. Rev. George Henslow has written a book in which this subject is set forth in detail.! It is impossible to demonstrate this point with the same certainty as the kinetogenetic origin of the articulations of the vertebrate skeleton and their characters, owing to the absence of paleon- tologic evidence. Henslow, however, says: ‘‘ When we find innumerable coincidences all tending in one direction, coupled with an indefinite capacity for vary- ing in response to forces in all parts of plants, I still maintain that [this] theory does not utterly break down,” as asserted by Mr. Wallace.* Wallace argues that since many regular flowers have been subject to the irritation of insects and have not become irregu- lar, there is no reason to suppose that this is the cause of the irregularity in question. To this Henslow re- plies:? [Mr. Wallace] ‘will see that existing regular flowers being mostly terminal, have no lower petals at all, but are so situated as to offer access to insects from all points of the compass. Moreover, when a plant with normally irregular flowers (which are always situated close to the axis, so that insects can only en- ter them in one way) produces a blossom in a terminal position (as foxglove, larkspur, horse-chestnut, etc., often do), it at once becomes quite regular.” This change may be brought about artificially, for, says 1The Origin of the Floral Structures by Insect and Other Agencies. Inter- national Science Series, Vol. LXIV. 2Natural Science, 1894, p. 178. 38Natural Science, 1894, p. 262. 384 PRIMARY FACTORS OF ORGANIC EVOLUTION. Henslow, ‘flowers normally irregular in nature often revert to their ancestral regular forms under cultiva- tion in the absence of insects, and then come true from seed, as the Gloxinias.” CHAPTER VII—NATURAL SELECTION. ATURAL SELECTION is the process of dis- crimination of variations, by which those which are most in harmony with the environment survive. It is, in short, as expressed by Spencer, ‘‘the survival of the fittest.” Fitness is of various kinds, and is only determined by the nature of the environment. The success of a variety which appears in aquatic sur- roundings will depend on characters different from those which bring success in a forest. Variations which favor survival and increase among carnivorous animals differ from those useful to the life of herbivo- rous forms. So survival in human society depends on characters different from those which secure the same result among the lower animals, etc. It is thus evi- dent that natural selection is of many kinds and that forms survive for various reasons ; and it is hence of universal application. The reasons for survival may be divided into those which depend for survival on the relations of a type to the non-living environment, and those which depend on the living. The former may be divided into those which are passive and those which are active. The particular influences may be imper- fectly enumerated as follows: 386 PRIMARY FACTORS OF ORGANIC EVOLUTION. Non-living environment. a, Passive. Isolation of areas ; continuity of areas; building material ; food ; place of concealment; temperature; humidity. aa, Active. Pressure of earth, water, and air. II. Living environment. Food ; reproductive potency; sexual selection; digestive and other physiological power; muscular strength; su- perior weapons and other special mechanisms ; intelli- gence. I have already (page 4) quoted the language of Darwin, where he states that the supposition that nat- ural selection is a cause of the origin of variations is a mistake. From the nature of selection, Darwin’s po- sition thus expressed, is self-evidently sound. The attempt has, however, been made to apply the term selection to the efficient cause of all variation, and to divide its exhibitions into two kinds, natural and arti- ficial selection. If the primary assumption involved in this position is illogical, the dual division proposed is absurd. As may be readily seen in the table in the preceding paragraph, in which the factors of natural selection are enumerated, the conditions necessary to selection are mostly identical, whether imposed by na- ture or by the hand of man; i. e., whether natural or artificial. The physiological effects of food, tempera- ture, exercise, etc., do not differ, whether due to nat- ural conditions, or to the influence of man. The ob- servation of man’s influence is indeed especially in- structive in increasing our knowledge of the effects of natural causes, since in the former case we have the process in action within our control, while in ‘the latter case it is not. The subject of natural selection has been ably NATURAL SELECTION, 387 treated by Darwin, Wallace, and other writers, and it is one on which much further research may be profita- bly expended. It is the science of adaptations, and the name Chorology has been framed for it by Haeckel, but the earlier term CEcology is now generally used. It was not overlooked by biologists prior to Darwin and Wallace, and is stated in general terms by Lamarck in his Philosophie Zoologigue, but it was reserved for the two authors just mentioned to create the science. I shall here only refer to a few aspects of the subject. Isolation naturally tends to emphasize any pecu- liarities of structure which may harmonize with the conditions of the environment, by the barrier which it sets up against the entrance and mixture of forms from other localities where the environments are more or less different, and where the characters are correspond- ingly proportionately diverse. Isolation conversely prevents the emigration of forms, and the consequent mixture with the differing forms of other regions. Breeding in and in is produced on a large scale. Geo- graphical isolation is a result of the formation and population of islands, whether this be accomplished by submergence below or by elevation above sea level. A noteworthy illustration of the former case is seen in the West Indian Islands, which represent the elevated regions of a former continent. Here the faunz of the respective islands have been separated from each other since late Pliocene time. We find that while most of the genera of land Vertebrata are generally distributed, each island possesses peculiar species. This is even true of the birds, whose powers of migration are quite sufficient to enable them to pass from island to island. The restriction of land mollusca is still greater, several islands having genera peculiar to them. 388 PRIMARY FACTORS OF ORGANIC EVOLUTION. Isolation is also produced by inequalities of land- surface, resulting from the elevation of mountain ranges, plateaus, etc. This is well seen in Mexico and Central America where the number of species of Vertebrata is large, owing to the fact that many of them are restricted to very narrow areas bounded by impassable barriers. The restriction of species of land Mollusca to each of the numerous valleys of the Ha- waiian Islands has been made the subject of an espe- cial study by Dr. Gulick, who treats of them with especial reference to the evolution of their forms. The assimilation of inorganic matter necessarily preceded that of organic matter, so that this function characterized the first organic beings, whether animal or plant-like in other respects. Among animals we may regard the vegetable feeders as having by a little preceded the carnivorous forms. Omnivorous forms must have come into existence soon after, and from these, both classes of feeders have been from time to time recruited ever since. The primitive Vertebrata were probably carnivorous, and most of the fishes and Batrachia have always been such. Herbivorous forms have arisen from time to time among Reptilia, and of granivorous birds there are many. The early Mamma- lia were divided between omnivorous (Multitubercu- lata) and insectivorous types (Protodonta, Pantothe- ria); while the higher Mammalia of all kinds were de- rived from more or less omnivorous forms (primitive Condylarthra and some Creodonta). We may account primitive insects to have been largely herbivorous, even more than they are at the present time, while Carnivora predominate in marine invertebrate life. It is not diffi- cult to understand that circumstances of the environ- ment would determine the food of animals, and would NATURAL SELECTION. 389 divert omnivorous forms into carnivorous or herbivo- rous habits as abundance of food and competition of rivals might dictate. Sexual selection is of two general classes, that in which the male selects, and that in which the female determines the result. In the former case the most vigorous males, or those in which the mechanisms for seizing the females are most effective, propagate their kind most successfully. It is well known that in many animals, especially the Arthropoda, the males are fur- nished with especial organs of prehension. In verte- brates similar organs are especially conspicuous in some of the Batrachia Salientia. (See p. 65.) The species of Arcifera exhibit peculiar structures during the breeding season ; either an extension of the natatory membrane, or the development of corneous plates or spurs, as aids to prehension. There is much variety and efficiency displayed in this point (except in Bufonide), in especial contrast to the apparent ab- sence of all but the weakest modifications among the Ranidz. This is in compensation for the structure of the sternum, whose lateral halves, being movable on each other, offer a slighter basis of resistance for the flexor and extensor muscles of the fore limbs of the male. In the Firmisternia the halves of the shoulder girdle do not overlap below on the middle line, but abut against each other, thus preventing compression (Fig. 51, page 198). While no appendages of the season have been ob- served in some Cystignathide, in several genera two acute spurs appear.on the superior aspect of the thumb and more rarely spur-like tubercles on the breast ; the body is sometimes shielded with hardened points on the rugosities, or the lip surrounded by an arched 390 PRIMARY FACTORS OF ORGANIC EVOLUTION. series of corneous ruge. In the Leptodactylus penta- dactylus Laur. a huge acute process of the metacarpal of the thumb, projects inwards. Its apex is covered by a horny cap, and it is a formidable grappling-hook to aid the male in retaining his hold. There is added to this in the same species a horny plate on each side of the thorax of the male, from which project three acute points. With these fixed in her back and the thumb spikes in her breast the females cannot escape. Struc- tures like this do not appear in the Firmisternia. Here the inferior elements of the scapular arch abut against each other, so that the thoracic cavity does not contract on pressure, and the possibility of the male retaining a firm grip on the female is thereby greatly increased. In the Cystignathus pachypus the males exhibit a permanent enlargement of the bra- chium, dependent on largely developed anterior and posterior ale of the humerus. (Vide Ginther, Ann. Mag. N. H., 1859.) Another kind of male selection is accomplished by the combats of males for the possession of the females. This is usual in polygamous birds and Mammalia, and in some promiscuous species of both. Of the birds the Galline form the best known example; and of the mammals, most Ungulata, and the eared seals (Otari- ide), are illustrations. In this way the weak males are eliminated either by death, or by exclusion from the opportunity of reproduction. The males in such species are armed with spurs, horns, or large teeth, except in some of the Perissodactyla, which have neither. Female selection is seen in another direction. Here the male attracts by the superior brilliancy of his colors or peculiarity of physical appearance, as well as by NATURAL SELECTION. 391 his notions. The available growth-energy of the male being superior to that of the female in most animals, his structure is more liable to excess of development in useless directions. In many of the Arthropoda, especially the Insecta, the males possess processes of the head and thorax besides the especially useful pre- hensile peculiarities of the limbs. Among Vertebrata the male generally possesses the more brilliant colors. This is especially noteworthy in fishes and birds. It is also frequently the case in lizards, although in one genus (Liocephalus) the female has the brighter hues. The selection (taking no account of the origin of these characters) acts in the probable preference by the females for the most brilliant colors and most impres- sive forms, thus propagating both, and in the neglect of those males in which these characters are not so well developed. The plainness of the females aids in their concealment and enables them to perform their ma- ternal functions in safety. The desire of the males to attract the favor of the females leads to many peculiar performances among birds. The males display their plumage by spreading their wings, tail, tail-coverts, etc., and strut and go through many antics in the presence of the females. Familiar examples are seen in our barnyards in the turkeys, peafowls, and pigeons. In the paradise-bird the most remarkable exhibitions occur, according to Wallace. In song-birds thé male is frequently the only or the best songster, and the development of the vocal powers resulting from the sexual impulse is most re- markable. Among Mammalia the female selection is less common than male selection. In the case of some of the old world monkeys (Macacus, e. g.) the female presents the greater physical indication of ex- 392 PRIMARY FACTORS OF ORGANIC EVOLUTION. citement in the extraordinary development of the nates at the season of heat. Mankind, appropriately to the high development of the mental powers, is selected with reference to qualities of mind as well as with re- gard to the physical attractions. Mental advantages being equal, beauty is preferred, but beauty is often neglected in favor of superior moral and intellectual characteristics. Both sexes take part in the selection ; in the lower races chiefly the male, while in the higher races the choice rests ultimately with the female. It is probable that in future civilized mankind will exer- cise more care than now in the prevention of marriage of persons affected with serious physical and mental defects, such as chronic diseases, insanity, alcoholism, criminality, etc.; but beyond this, supervisory selec- tion cannot go. The supposition which is sometimes entertained by some persons, that mankind will ina state of higher civilization prefer physical to mental perfection, is certainly ill founded. And among men- tal qualities, a high value will always be attached to those which render social life easiest and most pleasant ; the standards of ease and pleasure being innumerable. Among the most conspicuous examples of the ac- tion of natural selection are to be found those resem- blances in color or form between animals and their environment which serve to conceal them from ene- | mies. Animals possessing such protective appear- ances naturally escape the observation of enemies which prey on them, while those which do not possess them are more readily captured and eaten. Much is to be found of interest on this attractive subject in the writings of Wallace, Poulton, Beddard, and others. The two authors first named ascribe these color and form characters to natural selection as a cause. This is, NATURAL SELECTION. 393 however, impossible; yet natural selection has un- doubtedly been the cause of their survival. Professor Poulton has demonstrated (p. 230) that the protective colors in lepidopterous pup are produced directly by the influences of light on the nerves of the ani- mal and its reflex action on the pigment depository process. The first objection to the belief that natural selec- tion is the primary cause of organic evolution has been already stated as follows: ‘‘A selection cannot be the cause of those alternatives from which it se- lects. The alternatives must be presented before the selection can commence.” But the supporters of the view that natural selection is the origin of variation allege that it produces this result by the continued survival of minute differences which are useful, thus accumulating variation. That minute advantageous differences will secure survival no one can doubt, but it must be remembered that the variations which con- stitute evolution have been in a vast number of cases too minute to be useful. But the general question is not affected by the supposition that advantageous va- riations may be sometimes minute. Minute or great, they have to be assumed in the argument for selec- tion; and whether minute or great, they have a def- inite cause. * ‘ x In conclusion of Part II. of this book, I trust that I have adduced evidence to show that the stimuli of chemical and physical forces, and also molar motion or use or its absence, are abundantly sufficient to pro- duce variations of all kinds in organic beings. The va- riations may be in color, proportions, or details of struc- ture, according to the conditions which are present. PART IIE THE INHERITANCE OF VARIATION. ) PRELIMINARY. N THE first section of this book I have endeavored to show that variation of character is not promis- cuous or multifarious, but that it is limited to certain definite directions. That this rule applies to all kinds of characters, whether they are of the less fundamen- tal kind which distinguish species, or of the more fun- damental kind which distinguish the higher divisions. In the second section I have endeavored to show that many characters, both those of the more super- ficial and those of the profounder kinds, are the direct result of chemical and physical stimuli, and of molar motion or use, or of the absence of the latter or dis- use. It now remains to ascertain whether the characters or variations so produced are inherited ; that is, whether characters so acquired are transmitted to succeeding generations. Unless this proposition is demonstrated, our knowledge of the method of evolution remains in- complete, and we must look for some new explana- tion of the progressive increments and decrements of structure which constitute the history of organic life. The present part of this book will be devoted to an examination of this question, and to the exposition of such laws as may be derived from such examination. CHAPTER VIII.—HEREDITY. 1. THE QUESTION STATED. T IS the popular belief that characteristics of par- ents are transmitted to their offspring through the medium of the reproductive cells. This opinion is founded on an infinitude of observations easily made on plants, animals, and men, and in fact it is not de- nied as a general statement by anybody. It is a fact of ordinary observation that many and apparently most of the structural characteristics of one generation are inherited by its offspring. Not only is this the case, but the functionings of organs which depend on minute histological peculiarities are inherited. Such are points of mental and muscular idiosyncrasy; of weakness and strength of all or any of the viscera, and consequent tendencies to disease or vigor of spe- cial organs. Darwin has collected in his work on the Descent of Man numerous instances of the inheritance of various tricks of muscular movement of the face, hands, and other parts of the body. But it is a fact of equally ordinary observation that some peculiarities of parents are not, or may not be inherited, and among these may be enumerated mutilations and injuries, as well as characters which are normal. It is then a question of essential importance to ascertain what HEREDITY. 399 kinds of characters are inheritable, and what are not inheritable. It has been insisted by Weismann and others that characters which are newly acquired by an organism are not inherited, whether they be products of normal or abnormal conditions. In support of this view, he points to the early isolation in embryonic life of the reproductive cells from the remainder of the organism, and their continued isolation during later life, so that they are protected from the stimulating influences which affect the remainder of the body. He also points to the permanence of this isolation of the germ plasma from generation to generation, which insures only the transmission of those characters which it contains, as distinct from those which are found in the remaining cells of the organism, which constitute the body orsoma. The characters which are inher- ited, and which are present at birth are termed con- genital, while those which appear in the body under the influence of external stimuli are termed acquired. The theory of Weismann then is, that the acquired characters are not inherited. Besides the fact that sporadic injuries and mutila- tions of the soma are not inherited, there have been cited various cases of the non-inheritance of mutila- tions which have been often repeated and for long periods of time. Thus the rupture of the hymen in human females has not been followed by its abolition. The practice of circumcision by the Jews has not re- sulted in the disappearance from that race of the por- tion of the body thus artificially removed. The con- tinued cutting of the hair of men of many races has not made it less abundant. The practice of distorting the feet of a class of their women by the Chinese has 400 PRIMARY FACTORS OF ORGANIC EVOLUTION, not modified the shape of that part of the structure in the race. I have myself cut off the tails of nine suc- cessive generations of mice without producing the slightest effect on the length of the tails of the tenth. Nevertheless such negative evidence only demon- strates that such modifications of the structure may not be inherited. A single undoubted example of the inheritance: of a mutilation would prove that no in- surmountable barrier to such inheritance exists. And well authenticated examples of such cases are known and will be mentioned later on. But it is not with mutilations that the paleontologist has to do. The rupture of the hymen and circumcision, and most muti- lations, can only occur once in the life of the individual, and generally they produce no appreciable direct effect on his or her metabolic physiology. Moreover, the mutilations above cited as not inherited are experi- enced by but one sex, except in the case of the tails of the mice. The question is widely different with re- gard to the parts of the structure in which we observe the real differences between organic types. The defi- nitions of natural divisions rest to a great extent on the diversities displayed by the organs of motion and nutrition. Now these are in use in animals during most of the hours not spent in sleep. Their move- ments are perpetual, and their activities only cease with death. It is then quite unreasonable to cite the history of mutilations as evidence against the inheri- tance of natural characters produced by oft-repeated and long continued natural causes. It has been shown in Part Second of this book that structural characters are produced by use and other stimuli to growth. It has also been shown in Part First that the characters so produced show a progres- HEREDITY. 401 sive increment or evolution from earlier or later geo- . logic periods. There are two possible explanations of this phenomenon. The one is that the characters of one generation are inherited by the next, which adds to them by the activity of the same stimuli which gave them origin, thus producing progressive increase of growth. The alternative is, that these structural characters are produced by each generation for itself. It is obvious that the latter hypothesis provides for no additional development of a character in one gen- eration above another. There are other objections to the latter view, but letting these pass for the present, it is only necessary to examine the embryonic history of animals to show that it is entirely untenable. For if some or all of these acquired characters can be found present in the early stages of growth, as in the egg, the pupa, the fcetus, etc., it becomes clear that such acquired characters have’been inherited. That such is the fact is abundantly demonstrated by embryo- logical researches. This fact alone is sufficient to set at rest by an affirmative answer the question as to the inheritance of acquired characters. And that this answer applies to all time and to all evolution is made evident by the fact, which is disclosed by paleontology, that all characters now congenital have been at some per- tod or another acquired. 2. EVIDENCE FROM EMBRYOLOGY. a. Vertebrata. I have already (p. 292) pointed out the gradual evolution through mechanical causes of the tongue and groove-joints in the Mammalia as exhibited by the distal ends of the metapodial bones of the feet where 402. PRIMARY FACTORS OF ORGANIC EVOLUTION. they articulate with the phalanges. Mr. Carey having agreed with me that those have been produced by me- chanical causes, he alleges that they are not inherited, but are produced by each generation for itself. To this Dr. J. L. Wortman remarks as follows: ‘‘With reference to Mr. Carey’s first proposition that the metapodial crests are produced during the life of each individual by the necessary interaction of parts, it appears to me to bea very simple one indeed. If they are produced, by pressure during the lifetime of each individual, and are not inherited, then clearly we should find the crests absent in new born animals that had never walked, and in which the metapodials had not been subjected to any impact or pressure what- ever. Ihave taken the trouble to examine a number of such examples in which the distal ends of the bones were entirely cartilaginous, and I find that the keels and grooves are as well developed as they are in the adult animal. I will cite one case in particular in which I happen to know the history completely. Dur- ing the past winter, a young hippopotamus was born in the Zodlogical Gardens in Central Park, New York, and it was stated to have been a premature birth ; the animal lived but twenty-four hours, and I was informed by the keeper that it never stood upon its feet. An examination of the feet shows that the distal ends of the metapodials are entirely cartilaginous, and in them the keels are as well prefigured in cartilage as they are formed in bone in the adult animal. I have also found the same to be true of new-born rabbits and guinea- pigs. In another case of a young buffalo calf preserved in the American Museum Collection, the distal keels of the metapodials are complete notwithstanding the fact that the epiphyses of all the bones are very im- HEREDITY. 403 perfectly ossified. This evidence, it appears to me, effectually disposes of the question of the production of these structures during the lifetime of the individ- ual. They are as truly inherited as is the number of digits or any other important organ in the animal econ- omy.” Such observations may be repeated indefinitely. Thus the astragali of the higher Mammalia are already grooved before birth, and are not flat up to that time as in their Puerco ancestry. The reduction of digits appears very early in foetal life, and the ball and socket articulations of the cervical vertebre of the Diplarthra are by no means introduced after birth. The teeth possess the normal structure of their crowns while yet in the alveoli before eruption. In some cases the transition from a primitive to a modern type of tooth has been observed to take place in the embryo. Dr. A. von Brunn! has shown that in the embryos of the rat, the enamel-producing epithelial layer of the molar teeth undergoes a remarkable change at the places where the transverse crests of the crowns are to appear. Before the enamel layer is deposited, the portion of the epithelial layer corresponding to the cross-crests undergoes degeneration, as a result of which it acquires the character of a stratified squamous epithelium. Thus no enamel is laid down on the sum- mits of the cross-crests, which present the exposed dentine when erupted. Now it is a fact that the crowns of the molar teeth of the ancestors of the genus Mus, were covered with enamel at maturity, like all other 1“ Notiz tiber unvollkommene Schmelzentwickelung auf den Mahlzahnen der Ratte, Mus decumanus’’: Archiv fir Mikroskopische Anatomie, 1880, XVII, Pp. 241-243, Pl. XXVII. Ryder, American Naturalist, 1888, p. 547. 404 PRIMARY FACTORS OF ORGANIC EVOLUTION. primitive Glires. The removal of the enamel from the apices of the tubercles and crests of their descendents is due to the abrasion consequent on ordinary use. On this Ryder (/. c.) remarks: ‘‘ The great value which is to be attached to the fact that abrasions of the enamel of the adult, which have reacted upon the functional activity of the enamel organ of the embryo rat, so that such mechanically induced alterations could be inherited, does not consist so much in the proof it affords that mutilations can be inherited, as it does that mutilations incurred in the ordinary struggle for existence, may, under Certain conditions in certain practically feral species, be transmitted.” Having shown by these examples that acquired characters can be inherited, I offer some other illus- trations which are at hand. b. Arthropoda. It has been already rendered probable if not certain (p. 268) that the segments of the body and limbs of the Arthropoda were originally produced by the movements of definite tracts on each other, during the period that the external surfaces were becoming hardened by chi- tinous or calcareous deposits. It is well known that this segmentation is no longer produced by this mechanical cause during the adolescent or any other post-embry- onic stage of the life of the individual, but that it ap- pears during the various stages of embryonic life, and is therefore inherited. Thus segmentation of the body appears in the embryo while still attached to the yolk. During the larval life of many insects the process of segmentation is suspended, but during the repose of pupal life, it goes on with great rapidity. In this stage while protected from external mechanical stimuli, the HEREDITY. 405 limbs with their specialized segments are fully devel- oped, so that the individual is mature as it issues from its prison. This illustration of inheritance derives its point in the present connection from the fact that it presents an example of the inheritance of characters which were plainly acquired by mechanical stimuli during post-embryonic life of the primitive ancestors of the Arthropoda. 3. EVIDENCE FROM PALEONTOLOGY. a. The Impressed Zone of the Nautiloids. I have already quoted Professor Hyatt on the par- allelism which is characteristic of the various series of nautiloid Cephalopoda, as discovered by paleontologic research. (P.182.) The impressed zone is a character which has been produced by mechanical causes (pres- sure), and Prof. Hyatt has observed cases where this acquired peculiarity has been inherited in instances where the mechanical cause which produced it no longer existed. He describes these instances as follows :1 ‘¢The characteristic dealt with in the paper of which this is an abstract, is of essential importance among nautiloids and ammonoids or all of the Cepha- lopoda having chambered shells and living within their shells. It consists mainly of an impression made on the inner side or dorsum of each outer whorl during the coiling up, as the whorl grows and is moulded over the venter or outer side of the next inner whorl. “‘This matter will be better understood, if a short description is given of the following figures. Fig. 115 shows an almost complete fossil cast of a full grown lAmerican Naturalist, 1893. October, p. 865. Professor Hyatt has person- ally looked over and corrected these quotations. ‘ 406 PRIMARY FACTORS OF ORGANIC EVOLUTION. Metatoceras cavatiformis Hyatt, and some of the lines or sutures made in the external surface of the cast by the intersections of the partitions or septa that cut up the coiled tube of the living shell into air cham- bers. Fig. 116 shows a broken specimen of the same species, but with the outer and older whorls in large part removed. The innermost septum near the center of the coil was built across the interior after the animal had constructed the hollow apex or point. It then moved along, adding to the external wall of the tube, Fig. 115. which has been destroyed and removed from this cast, and built the second septum, and so on until it reached the tenth septum. By some freak of fossilization a number of the septa beyond this have been destroyed, so that if we were to remove the fragment of the ex- ternal whorl and take out the center which has just been described, this would have the exact aspect of a cast of a young shell with ten air chambers.! The 1The shaded area in the center, shaped like a large inverted comma, was an open space in the living shell. This is almost invariably filled by the HEREDITY, 407 eleventh air space or chamber being open and without divisions would then appear to be the living chamber which the animal occupied when it built the tenth sep- Fig. 116. tum. Normally the shell really continued to progress from the tenth septum by additions to the outer wall and put in new septa behind it, together with the con- Fig. 117. necting tube until it reached s’, and finally the last septum, /s. This one, /s, was really the last one built rocky matrix in which the shells occur and is often, as in this specimen, al- lowed toremain. Seealso 4, 5, 6, of Fig. 119, which show the comma shaped umbilical perforations or openings left at the center through the crytoceran form of the young. 408 PRIMARY FACTORS OF ORGANIC EVOLUTION. and it formed the floor of a true living chamber, Zc, formerly occupied by the animal at the time of its death and burial in the sediment of the carboniferous period. Fig. 115 shows a similar fossil but with a longer, although still incomplete living chamber. If the external wall of shell had been preserved none of these structures could be seen. Fig. 117 shows a fossil Temnochilus crassus, a shell of the same family, with this external wall preserved, and all these internal structures covered up. The impressed zone is the reéntrant curve shown in all these figures me and especially marked in Mee the lower outline of an outer whorl of another carboniferous species, Metacoceras adubium Hyatt, Fig. 118, zm. 2. ‘«It is not necessary to go into a discussion of the details of internal structures and their relations to the impressed zone in this abstract, but it is essential to give a general description of the morphogeny of the order of nautiloids. ‘« This group of chambered Cephalopods contains the following classes of forms: first, straight, conical shells, type Orthoceras, Fig. 119, No. 1; second, curved cones, Cyrtoceras, Fig. 119, No. 2; third loosely coiled, open whorled cones, do., No. 3; fourth, coiled cones with the whorls more or less enveloping, do., No. 5. The fourth and fifth forms are usually included in the old genus, Nautilus. Practically, it is better to designate the first class as orthoceran, the second as cyrtoceran, the third as gyroceran, and the fourth and fifth as nautilian. In tracing genetic series through HEREDITY. 409 time they are found to diverge in their evolution, start- ing with the orthoceran and passing through parallel lines of forms, many of the genetic series having in succession cyrtoceran, gyroceran, and even nautilian forms of the fourth and fifth classes. Others are not so perfectly parallel, stopping short with the cyrto- ceran class of forms or the gyroceran. Many also begin with cyrtoceran shells, while others diverge from the gyroceran, and still other series have only nautilian shells of different grades of close coiling and involu- tion. «