em ime ee ereree ENS fase a —-. a . ae — sang A BESS “4 ‘9 A: 4 Ar Vee eek : s v —- feat AS = buns ae e 1 toda ae iy z but i San ‘ ; iC a i bias ae a ia ANIMAL LIFE AN EVOLUTIONARY NATURAL HISTORY GENERAL Epitor: W. P. PYCRAFT ZOOLOGICAL DEPARTMENT, BRITISH MUSEUM REPTILES, AMPHIBIA, FISHES, AND LOWER CHORDATA EDITED BY J. T. CUNNINGHAM, M.A. Oxon. Tyontisplece PLATE XXIV NOILV Za tlh LdONC ‘SHHSIA AO SHITINVA TVOIdOUL ASHHL AO ANO GNV AVGIALNAYOVNOd AHL AO NHWIOUdS V (AVCINLNAOVNOd) ‘VONVS ‘WOdQV.L ANdAACNIV—V 100 INVITIIMG AHL ONILVALSOTII ‘AVGILNOGOLHYVHO AHL 40 WHE eae pect a a laa fice a a REPTILES AMPHIBIA FISHES AND LOWER CHORDATA RICHARD LYDEKKER J: T. CUNNINGHAM, M.A. Oxon. G. A. BOULENGER, D.Sc., F.R.S. AND J. ARTHUR THOMSON, M.A. ~~ > © As © WITH FOUR PLATES IN COLOUR, (HIRTY-THR#Y IN MONOTONE, A MAP, AND MANY TEXT FIGURES * ' METHUEN & CO. LTD. 36 ESSEX STREET W.C. LONDON First Published in 1912 PREP ACE HE kindly reception accorded to “A History of Birds "—the second in the order of final sequence of this series—encourages me to believe that these volumes on vertebrate animal life which I pro- jected really supplied a need felt by that increasing body of men and women who profess themselves “nature lovers”. Having conceived the plan of recording, at any rate in broad outline, the history of the vertebrates from the evolutionist’s point of view, I allotted to myself, as many already know, the task of writing the volume on Birds, and proposed to edit the remaining volumes rather than attempt to write them. And this because the day is now past when any single writer can hope to achieve such a task with even tolerable success, for this is the day of specialists. As a consequence, for the first time in the annals of natural history, the complete life-story of the reptiles, amphibia, and fishes, and those primitive creatures which lie at the foundations, so to speak, of the great house of the vertebrates, is told as only specialists can tell it. The very existence of these primitive ani- mals is unsuspected by most of us, but, as Professor Arthur Thomson shows, they present us with some most interesting and most important problems. The student of sociology will find in his chapters, no less than in those concerning more familiar creatures, much food for reflection bearing on the subjects of adaptation to environment, degeneration, and so on. \/ vi REPTILES, AMPHIBIANS AND FISHES Those who seek to discover the subtle and mysteri- ous factors which govern the transformation of animals will find much food for thought in Mr. Lydekker’s account of the reptiles, and in the chapters on the nursing habits of amphibia and fishes by Mr. G. A. Boulenger and Mr. J. T. Cunningham; and to these we would add the weird and fascinating chapter on the fish-life of the abysses of the ocean—a world wherein the light of day never penetrates, and where the pall of night is broken only by the pale phosphoresence emitted by the creatures doomed to dwell there! The names of the contributors to this volume alone vouch for the sterling merit of its contents: for they are men of established reputation. It is, therefore, unneces- sary to say anything of their attainments. But I should like to thank them here for the help they have so generously given me in my endeavour to make these volumes a landmark in the annals of zoological litera- ture. Some of our neighbours assure us that ‘‘ Darwin- ism is dead”. If these pages show anything they show that the contrary is emphatically the case! Though my task as Editor, with such contributors, could not but be an easy one as to the substance of the work, yet the burden imposed by purely mechanical details was really heavy, so much so that ill-health com- pelled me to hand over my labours before they had well begun to my friend, Mr. J. T. Cunningham. He has, however, displayed such scrupulous care and zeal that my practical retirement has in no way injured the scheme I had so much at heart. Wo PP ViCiCn Et? CONTENTS PREFACE List oF TEXT-FIGURES List oF PLATES 5 : ; SECLION. I REPTILES By RicHarp LYDEKKER CHAPTER I GENERAL CHARACTERS ‘ 3 6 : ; Reptiles in general. Definition of reptiles. Variety of forms. Ex- tremes of size. Some leading structural features of the class. CHAPTER II PEDIGREES AND RELATIONSHIPS : . . . Reptiles and their relationship to warm-blooded Vertebrates; the origin of reptiles. The Theromorpha and their relation to the Mammalia: the Ornithomorpha and their relation to birds. The tuatera: croco- diles and their ancestry: tortoises and turtles: flying reptiles: dinosaurs: snakes and lizards: geological history: plesiosaurs and ichthyosaurs: numerical strength. CHAPTER III Haunts anp Hagitats . : : ; : : The haunts of reptiles in past times and at the present day. The sea-iguana. Dearth of reptiles in polar and sub-polar regions. Haunts of tortoises and snakes. Tree-dwellers. Burrowers. Island forms. Geographical distribution, difficulties of interpre- tation. Relation to temperature, moisture, etc. Hibernation and zstivation. vii 15 32 mi REPTILES, AMPHIBIANS AND FISHES CHAPTER IV PAGE Foop AND GROWTH . : : : . . : . ° . ATi Food. Mode of killing prey. Snake-eating snakes; egg-eating snakes. Fascination. Rate of growth. Age. Vitality. Regeneration. Chronic disease. Sloughing of skin. Peculiar associations. CHAPTER: V SEX AND REPRODUCTION . 5 : : z , : d , 62! Some sexual features. Cries. Scent and scent-glands. Milky and other secretions. Significance of brilliant colouring. Habits in breeding season. Reproduction and care of eggs and young. Maternal instinct in pythons. Vipers swallowing young. Fertility of snakes. Incubation. CHAPTER VI COLORATION AND ITS INTERPRETATION . ; : ; ; : 75 Colour in relation to environment. Stripes and spots in lizards. Colour changes in relation to sex and age. Colours of young pit-vipers. Voluntary colour-changes. Green arboreal snakes. Desert-snakes. Sea-snakes. Bark-geckes. Lizards. Warning and _ protective colours. CHAPTER VII ADAPTATIONS . 4 ; : : : : ‘ - 3 4 + go ADAPTATIONS TO THE GENERAL CONDITIONS OF THE ENVIRONMENT: Terrestrial types. Arboreal types. Climbing types and effect on tail and feet. Running types. Bipedal types. Flying types. Swimming types. Sail-backed lizards. Limbless types. Burrow- ing types. Modification of the eye and ear. Dermal armour. CHAPTER VIII ADAPTATIONS (continued). ADAPTATIONS TO SPECIAL ENDS. ; 128 Devices for diminishing weight. Shoulder-girdle and pelvic girdle of tortoise tribe. Devices in regard to movement and breathing under water. Beaks and teeth in relation to food. Poison-fangs. The chameleon’s tongue. Blood-squirting from the eyes. Skin-secre- tions. Spitting. Fluid exudations. Hissing and intimidation. Death-feigning. Immunity to snake-venom. Lungs of chameleons and snakes. Voluntary fractures. Rattlesnake’s rattle. SECTION II AMPHIBIA By J. T. Cunnincuam, M.A., Oxon., and G. A. Bou tenGeER, D.Sc., F.R.S., Zoological Department, British Museum CHAPTER I GENERAL CHARACTERS : : ; ‘ - A C oP y7/ How distinguished from fishes, and from reptiles. Other general features of structure. Larval development. Classification. CONTENTS CHAPTER II EvoLuTION AND GEOLOGICAL History The earliest Amphibia, the extinct Labyrinthodonts. Absence of tran- sitional forms in the Secondary Formations. Number of Amphibia in comparison with other classes. CHAPTER III DISTRIBUTION AND HABITS Amphibia usually associated with stagnant fresh water. Geographical distribution. Notogaa and Arctogeea. Oceanic and other islands. Food and feeding. Relations to salt and moisture. Aistivation and hibernation. CHAPTER IV REPRODUCTION Tail-less Batrachia, general course of development. Various modes of protecting the eggs: nests in the water and on land. Eggs carried by the parents, round the legs by the male, on the skin or in a dorsal pouch by the female, in enlarged vocal sacs by the male. Pairing in Urodela. Protection of eggs by Urodeles. Salamanders that carry their eggs. Viviparous salamanders. CHAPTER V VARIATION AND ADAPTATION Metamorphosis and occasional ‘persistence of the larval condition (Neoteny). History of the axolotl and its explanation. Colour: protective and warning coloration. Adaptations for locomotion. Adhesive discs of tree-frogs. The spade of Pelobates. Respiratory adaptations: salamanders with neither lungs nor gills. Allantoic gills in embryos. Adaptations in the male for pairing. Convergent evolution in tree-frogs. SECTION III FISHES By J. T. CunnincHam, M.A., Oxon. CHAPTER I INTRODUCTORY Definition. General characters. Position in the Animal Kingdom and Classification. CHAPTER II EVOLUTION AND PAL#ONTOLOGY The extinct Ostracoderms. Fossil Elasmobranchs. Crossopterygians and Dipnoi evolved in fresh water. Pedigree of Teleostomi. Relative abundance of different orders in past and present times. 178 183 210 231 249 ox REPTILES, AMPHIBIANS AND FISHES CHAPTER III DISTRIBUTION AND LOCATION 3 ; : é : Fresh-water fishes and marine fishes. Physostomous fishes originally belonging to fresh water and Acanthopterygians which have returned thither from the sea. Fresh-water fish-faunz of the great contin- ents. Natural homes and habits in littoral zone. Tropical fishes. Arctic and Antarctic fishes. America and Europe. North Pacific. South Africa. India. New Zealand and Australia. Pelagic fishes. Abyssal fishes. CHAPTER IV CONDITIONS OF LIFE AND PHENOMENA OF SEX. : : 5 . Food and feeding. Relations to temperature, light, and salinity. Influences of seasons: migrations, hibernation, and zstivation. Association and commensalism, parasitism. Relations of sexes, sexual dimorphism. - CHAPTER V MOopES OF REPRODUCTION . : : : : c : : Oviparous and viviparous Elasmobranchs; eggs of Chimeroids. Nests of Protopterus and Lepidosiren. Nest-building of Ganoids and primitive fresh-water Teleosteans. Parental instincts in British shore-fishes. Nest of the Sargasso. Fishes which incubate eggs in the mouth. Pouch of the Pipe-fishes. Viviparous fishes. Curious breeding habits of the Bitterling. CHAPTER VI LIFE-HISTORIES . j ; ¢ . : c - ¢ : 5 Life-histories of Teleostomi. Eggs of Clupeide. Metamorphosis of Angler and Pleuronectide. Growth and Maturity. Diseases and Parasites. Life-history of Eels. CHAPTER VII VARIATION c 6 é : . : : : Continuous variations, their mathematical investigation and relation to locality. Discontinuous variations or mutations, e.g. in flat-fishes. Variation under domestication. CHAPTER VIII ADAPTATIONS < : : : 5 : 5 : < ¢ Shape, symmetry, colour. Flat-fishes. Synodontis. Locomotion. Fly- ing-fishes. Respiration, gills and accessory organs. Air-bladder, respiratory and hydrostatic function. Sense-organs and senses. Hearing, relation of auditory organ to air-bladder. Sense-organs of lateral line. Sight. Divided eyes of Anableps. Blind fishes. CHAPTER IX ADAPTATIONS (continued), PRropucTion or Sounp, LicHT AND ELEc- TRICITY Voice of fishes. Stridulating organs. Fishes which possess luminous organs. Microscopic structure of luminous organs. Electric fishes. Structure of electric organs. General remarks on evolution. PAGE 263 291 314 341 359 371 405>. CONTENTS SECTION IV CYCLOSTOMATA OR MARSIPOBRANCHS By J. ArtHuR THomson, M.A., Professor of Zoology in the University of Aberdeen CHAPTER I INTRODUCTORY General structural characters and definition. CHAPTER II MyxXINOIDS OR HAG-FISHES . > : : : : 2 ; Characters. Habits. Mode of feeding. Life-history. Variability. Adaptations in struggle for existence. Classification. CHAPTER Iil PETROMYZONTS OR LAMPREYS Characters, Habits. Mode of feeding. Life-history. Classification. Relationships of Cyclostomes. SECTION V By the Same THE LANCELETS. CEPHALOCHORDA . General position. Externalappearance. General characters. Functions and habits. Development and life-history. Relationships. Classi- fication. Distribution. Features of special evolutionary interest. SECTION VI ‘TUNICATES By the Same CHAPTER I STRUCTURE. General characters; a typical solitary Tunicate. CHAPTER II CLASSIFICATION AND DEVELOPMENT . : : ; : : Classification of Tunicates. Order I. Appendicularians. Order II. Ascidians. Order III. Salpians. Numbers, relationships. Life- history of a typical Ascidian. SECTION VII By the Same HEMICHORDA Animals included in the group; structure and habits of Enteropneusta, their development and distribution ; incipient species; structure of Cephalodiscus and Rhabdopleura. INDEX. xi PAGE 441 442 449 457 473 479 489 495 4 MIST OF TEXT-FPIGURES FIG. 1. A, Upper surface of skull of a Labyrinthodont. B, Upper surface of skull of Tuatera. C, Front view of lower end of tibia of a bipedal Dinosaur 2. Upper (A) and lower (B) views of the skull of a Belodont (Phyto- saurus). C, Upper view of skull of an extinct Crocodile (Metrio- thynchus) 3. African egg-eating snake and its mode of swallowing food . 4. Restoration of _Iguanodon 5. A, Skeleton of a short-tailed Pterodactyle (Pterodactylus spectabilis). B, Restoration of a long-tailed Pterodactyle (Rhamphorhynchus phyllurus). (By permission of the Trustees of the British Museum) : é ; : c : : ; : ‘ 6. A, Skeleten with outline of body of an Ichthyosaur. B, Skeleton with outline of body of a Plesiosaur. After Dr. C. W. Andrews . 7. The Luth or Leathery Turtle, to show the paddle-like limbs and shell. B, Portion of bony shell showing mosaic-like structure : 8. Restoration of skeleton of an Armoured Dinosaur (Polacanthus foxt). After Baron Nopcsa : g. A, lateral, B, dorsal views of the skull of a Horned Dinosaur (Tricera- tops flabellatus) showing the horn-cores and neck-shield to. Ventral view of the skeleton of the existing Logger-head Turtle (Thalassochelys) with the plastron removed to show the position of the shoulder-girdle and pelvis within the ribs. (By permission of the Trustees of the British Museum) : 5 : : c tz. A, skull of Iguanodon, to show predentary bone and edentulous fore part of jaw. B, skull of Dicynodon, showing single pair of upper tusks and horny beak. CC, skull of carnivorous Mammal-like Reptile (Galesaurus) to show differentiated dentition. (A and B, by permission of the Trustees of the British Museum) 12. Palatal view of dentition of existing Tuatera (A), and extinct Pavement- toothed Tuatera (B). C, Palatal view of dentition of Bean- toothed Reptile (Cyamodus). B, after Boulenger. (C, by permis- sion of the Trustees of the British Museum) : - : . Xiil PAGE It 21 50 97 Io0o 104 108 122 124 131 138 xiv REPTILES, AMPHIBIANS AND FISHES FIG. 13) I4. 15. 16. LZ 18. IQ. 20, 21. 22. 23" 24. 25. 26. Pa fe 23s 29. 30. 3. 32. A, Development of Hylodes martinicensis. 1, embryo seven or eight days old; 2, twelve days old; 3, young just hatched. B, Stages in metamorphosis of Common Frog. 1, tadpoles soon after hatching ; 2, tadpole with external gills, from above; 7, young frog . : : Midwife Toad (Alytes obstetricans) ; a, male with eggs; 6, female Hyla goeldii, a tree-frog carrying eggs on its back; a, from above; 0, from side; c, young when hatched : : : A, Rhinoderma darwini, external appearance. B, A specimen of the same with the gular sac cut open showing contained embryos Development of Crested Newt; a, b, stages within the egg; ¢, d, e, f, stages of the larva magnified : 5 : : 0 : : Skull and gill-arches of Dog-fish (Scyllium canicula) with anterior part of vertebral column ¢ ¢ Dermal denticles of the Dog-fish (Scyllium canicula). A, from above; B, from below; C, in section. A, cycloid scale of Plaice, showing four zones of annual growth; B, ctenoid scale of Sole Development of the homocercal tail as seen in a young flat-fish. After Alex. Agassiz. a, youngest stage, almost symmetrical without fin-rays; b, end of notochord slightly up-turned, dermal fin-rays appearing on the ventral side; c, d, end of tail more bent up and reduced ; ¢, f, the ventral rays become terminal Skull of Salmon ‘ : . : é : : : : : A, Holoptychius femingi from Upper Old Red Sandstone; B, Dipterus valenciennesi from Middle Old Red Sandstone Fierasfer and Holothurians. After Emery South American Lung-fish, Lepidosiven pavadoxa. A, female, B, male ; : 3 Nest of American Bow-fin (Amia calva). After Bashford Dean Three stages in the metamorphosis of the Angler, Lophius piscatorius. A, newly hatched; B, some days old; C, with fin-rays developed Four stages in the metamorphosis of the Flounder, Pleuronectes flesus. Variation in number of dorsal fin-rays in a sample of North Sea plaice . Microscopic structure of Luminous Organs of Fishes. A, Section of pearl-like organs of Myctophum. After von Lendenfeld. B, Section of head of Ipnops. After Moseley. C, Portion of same more highly magnified. D, Longitudinal section of lateral line of Halosauropsis. After von Lendenfeld Microscopic Structure of Luminous Organs of Fishes. A, Surface view of the three partially united postero-lateral organs of Sternoptyx diaphana. B, Longitudinal section of same. After von Lendenfeld. C, Section of phosphorescent organ of Porichthys notatus. After Greene . A, the Hag-fish, Myxine glutinosa. B, Californian Hag-fish, Bdello- stoma stouti. C, River Lamprey, Petvomyzon fluviatilis PAGE 185 192 194 197 200 234 237 238 239 241 255 303 320 322 344 346 360 424 427 447 Pisivor PLATES A specimen of the Pomacentride and one of the Chetodontida, illustrat- ing the brilliant coloration of these tropical families of fishes. A, Abudefduf taupou, ‘Samoa(Pomacentride). B, Tetragonopterus ephippium, East Indies (Chztodontide) . ° : Frontispiece From a drawing in colour by James Green. PLATE FACING PAGE I. Skeleton of Pariasaurus, the Ancestral Type of the Mammal-like Reptiles. (By permission of the Trustees of the British Museum) 16 II. A, The New Zealand Tuatera (Sphenodon punctatum), the sole survivor of the most primitive and most ancient group of the bird-like Reptiles. B,a girdle-tailed Lizard (Zonurus giganteus) of South Africa, to show apparent similarity of Lizard and Muatera, . : ; é : . é 20 Photos, fer Medland. III. A Giant Land-tortoise (Testudo an of South Aldabra Island ; 5 : é A E 2 ‘ 30 Photo from life. IV. A, The Mata-mata Terrapin (Chelys fimbriata). (By permission of the Trustees of the British Museum). B, The Nile Soft-tortoise (Trionyx triunguis) . : : . : . : : 54 V. Protective Resemblance in Reptiles: Bark-geckos of Madagascar. A, The Lichen Bark-gecko (Uvoplates fimbriatus lichenium), B, The Common Bark-gecko (Uroplates fimbriatus) . : oy From a drawing in colour by James Green. VI. A, Moloch Lizard (Moloch horridus). B, “ Horned Toad” (Phrynosoma cornutum). To show depressed form of body characteristic of terrestrial types of lizards. C, Bearded Lizard (Amphibolurus barbatus) of Australia : ° ° : =) Or VII. A, Chameleon Lizard. B, Common Chameleon. To show similarity of arboreal forms in widely different types . 5 - 94 B, Photo, Lewis Medland. VIII. A, Australian Frilled Lizard (Chlamydosaurus kingi), from a stuffed specimen. B, C, same in running positions (from life) . 96 B and C, Photographed by the late Saville Kent, Esq. IX. A, Part of the skin of an African Python (Python seb@), showing external vestiges of the hind limbs. B, Complete bones of the hind limb-girdle in same species. (By permission of the Trus- tees of the British Museum) : c ° : B . * BEI xv b XI. XII. XIII. XVI. XVII. XVIII. XIX. XX. XXI. XXII. XXIII. XXIV. REPTILES, AMPHIBIANS AND FISHES FACING PAGE A, Glass-snake to illustrate the assumption by lizards of the glid- ing type of conformation characteristic of snakes. B, Ring- snakes hatching . : : : ; ¢ < c : A. Photo, Lewis Medland. B, Photo, W. S. Berridge. S-necked (Cryptodiran) Tortoise (Hornopus arcolatus), to show how the head is withdrawn by a vertical S-like curve. Side- necked (Pleurodiran) Tortoise (Sternothoerus niger), to show how the head is withdrawn by a lateral flexure of the neck. (By permission of the Trustees of the British Museum). . A, Head of African Puff-adder (Bitis arietans), showing poison fangs. B, Poisonous lizard (Heloderma suspectum) of Arizona . A, Photo, W.S. Berridge. B, Photo, Lewis Medland. A, Crested Newt (Molge cristatus), male. B, Ditto, female. C, Bull-frog (Rana catesbiana) C, Photo, Lewis Medland. . A, Proteus anguinus. B, Amphiuma means. C, Siren lacertina A, Photo, W. S. Berridge. B and C, Photos, Lewis Medland. . A, Smooth-clawed Toad (Xenopus levis), dorsal aspect. B, Same, ventral aspect. C, Common Toad (Bufo vulgaris) Photos, Lewis Medland. A, Surinam Toad (Pipa americana), female with young escaping from cells in skin of back. B, Nototrema marsupiatum, with dorsal pouch half developed. C, Same with pouch full of eggs . A, a worm-like Amphibian (Ichthyophis glutinosus) with eggs. B, Embryo of same taken from egg, ets C, Embryo just before hatching, magnified . : : : é A, Axolotl, albino specimen. B, Intermediate stage in metamor- phosis of Axolotl. C, Fully developed Amblystoma tigrinum A, Photo, W. S. Berridge. A, Hyla baudini, a Tree-frog of Nicaragua; an example of pro- tective coloration. B, Bombinator igneus, the fire-bellied Toad, showing warning coloration and warning attitude. ¢ . From a drawing in colour by James Green. A, The African fringe-finned Ganoid, Polypterus bichir. B, Com- mon Sturgeon, eee sturio. C, American Bony- nee Lepidosteus osseus. : : Photos from stuffed specimens in fe) British Maseunt A, Hammer-headed Shark. B, Head of Chlamydoselachus anguineus, after Garman, showing the six gill-slits. C, Chimera colliet. Atter Bashford Dean Map of the World on Mercator’s projection, showing Distribution 113 127 144 162 164 167 195 206 212 220 240 253 OMHIshess.;) cs =, wees. o RUe—enoeen A, Arapaima gigas. B, Galeichthys assimilis, a Central Ameri- can Siluroid. C, Hydrocyon goliath, an African Characinid A, Photo from specimen in Natural History Museum. Frontispiece. 268 EISSOr PEATES PLATE FACING XXV. Male and Female Dragonet (Callionymus lyra) in pairing atti- tude. The male is much larger and more brightly coloured than the female . : . é : : ; : From a drawing in colour by Tames Gren -XXVI. A, Egg-capsule of Chimera colliei, after hatching of embryo; lateral aspect. B,Same; dorsal aspect. C, Two capsules pro- truding from oviducts of female Chimera . : : XXVII. A, Spawn of Lumpsucker (Cyclopterus lumpus). B, Eggs of Gar-fish (Rhamphistoma belone) attached to sea-weed. C, Butter-fish guarding its clump of spawn . ; - : A and B, After Photos from nature by Hbrenbatie C, Photo, R. Vallentin, Esq. XXVIII. Nest of Sea-stickleback (Gastervosteus spinachia) with clump of spawn in the centre. The fronds of weed are bound together by a thread spun by the fish . : : . XXIX, Stages in the metamorphosis of ee brevirostris, the larva of the Common Eel c é - : : XXX. A, Upper, right, side of abnormal Turbot, 4°4 cm. long, captured on the coast of Cornwall in 1906. B, Lower, left, ide of same. C, Lower side of normal Flounder after exposure to ae for 14 months : ° : XXXI. A, Synodontis batensoda, a fish of the Nile which Panay Swims in an inverted position. B, The Four-eyed Fish, Ana- bleps tetrophthalmus. C, Myctophum remiger, an oceanic fish with luminous organs. D, Ipnops agassizii XXXII. A, Ceratodus australis, the Australian Lung-fish. B, Sacco- branchus fossilis. C, The Electric Eel, Gymnotus electricus B, Photo, Lewis Medland. XXXIII. A, Hag-fish (Myxine glutinosa), anterior end dissected to show the six rounded gill-pouches, etc. B, Hag-fish opened to show hermaphrodite generative organ in the male condition. C, Eggs of Californian Hag-fish connected by their polar filaments. D, Polar filaments of same enlarged. E, F, Ends of single filaments more enlarged. : : : : . . XXXIV. A, Amphioxus external surface, left side. B, Amphioxus dissected. C, Larva of Amphioxus enlarged, lying on left side. D, A soli- tary Ascidian (Ascidia mentula), left side, external surface. E, The same, internal structure XXXV, An Appendicularian (Folia ee from the tee vata Sea. B, Portion of a specimen of Botryllus, showing three systems oF zooids. C, Acompound individual (gemmarium) of Pyvosoma. D, Open end of same. E, Three zooids of same Ss acee 1p Salpa democratica, a Sentai individual . . : ‘ XXXVI. A, B, C, Three stages in the metamorphosis of a simple nection: D, Balanoglossus kowalevskit. E, Cephalodiscus dodecalophus, ventral view P : ; F A C F . “ é XVil PAGE 310 316 324 327 355 366 376 432 458 482 486 4 SECTION 1 REPTILES CHAPTER: I GENERAL CHARACTERS Reptiles in general. Definition of reptiles. Variety of forms. Extremes of size. Some leading structural features of the class. UR forefathers of classic days evidently had no acquaint- tf) ance with either the flying pterodactyles, the bipedal iguanodons, or the whale-like ichthyosaurs of a long past epoch, or, for the matter of that, with the so-called fly- ing dragons of the Malay countries of to-day. Otherwise they would scarcely have applied the name reptiles (Latin repo, Greek é&pmw) to the creatures forming the subject of the present section, together with frogs, salamanders, and their kin, which constitute the class Amphibia, or Batra- chia. As a matter of fact, however, having eliminated the Amphibia, and regarding the marine turtles as forming the exception which proves the rule, we find that, etymologically, the term reptiles adequately expresses one of the leading ex- ternal features of the existing members of the class. For modern reptiles, whether provided with limbs or no, are essenti- ally creatures whose “belly cleaveth to the ground,” and which consequently creep (or “reep”) in the true sense of the word. Indeed, the one great external characteristic of nearly all limbed reptiles of to-day is that the body is carried close to or even touching the ground, and is never raised high above it in the manner characteristic of mammals and the majority of birds. In some of the arboreal forms this trait is less marked, while I 2 REPTILES species like the Australian frilled lizard, which at times run on their hind legs, form in some degree an exception; but in the main the diagnosis is true. Seeking for something more nearly approaching a scientific definition, and confining our attention to the living members of the class, it may be affirmed that reptiles are cold-blooded vertebrates, unprovided with hair or feathers, which breathe atmospheric air by means of lungs, and do not undergoa marked transformation, or metamorphosis, after leaving the egg, whereby they pass from gill-breathing to lung-breathing creatures. By the first two characters they are distinguished from mammals and birds; by the third, coupled with the circumstance that their limbs do not partake of the character of fins, they are differentiated from fishes; while in the fourth feature they differ from amphibians in general, despite the fact that certain more specialised members of the latter have “ slipped” some or all of the early stages of development. Another important feature of modern reptiles, and one also available in the case of their fossil relatives, is that the skull is articulated to the first vertebra by means of a single, although tripartite, knob, known as the occipital condyle; this character distinguishing them from mammals and amphibians, although not from birds. When we endeavour to go much further in regard to differ- entiating both recent and fossil reptiles from the other three classes of terrestrial vertebrates, we shall be met with increas- ing difficulties. For it is manifest that while reptiles are de- scended from amphibians, it is equally clear that reptiles themselves have given origin on the one hand to birds and on the other to mammals. Consequently, were the whole scheme of animated nature displayed before our eyes, we should expect to find amphibians with a single occipital condyle formed by the development of a median element between the two lateral ones, and without a metamorphosis or, what is the same thing, rep- tiles with a metamorphosis. On the other hand, we should look for reptile-like creatures approximating more or less closely to birds, and acquiring at some stage of their develop- ment wings of a bird-like type, feathers, and warm blood. — In- deed, for all we know, pterodactyles, although off the bird-line, may have been (and very probably were) warm-blooded, GENERAL CHARACTERS 3 Finally, there must have been mammal-like reptiles which ac- quired two occipital condyles by the suppression of the median element of the reptilian one, and at some stage of their progres- sive evolution exchanged their scales (if they possessed them) for hair, and likewise raised the temperature of their blood markedly above that of the surrounding medium. Obviously, then, as in all analogous instances, it is not of much use attempting to define distinctions which may have never existed between extinct reptiles and the other three classes of terrestrial vertebrates. It must therefore suffice to give the foregoing definition of existing reptiles, and to consider that such extinct vertebrates as come within its limits, or which approximate more or less closely thereto, are likewise to be in- cluded within the limits of the class. Existing reptiles include only a small number of leading groups, or orders; but such groups do not altogether coincide with the popular classification of these creatures. In popular language reptiles are roughly classed as crocodiles (inclusive of alligators and gharials), tortoises and turtles, lizards, and snakes. Crocodiles, in this wider sense, constitute an ordinal group (Crocodilia) by themselves, as do tortoises and turtles a second (Chelonia). Here, however, the agreement between popular and scientific classification ceases, for whereas in the former the lizard-like tuatera (Spenodon) of New Zealand is reckoned as a lizard, in the latter it is regarded as the sole surviving representative of an extremely generalised order (Rhynchocephalia), with but little in common with the group in which the true lizards are included. Again, lizards and snakes are popularly classed as widely sundered groups, whereas in the scientific scheme the two (exclusive of the aforesaid tuatera) are brigaded in a single group (Squamata) ranking in value with the Crocodilia, Chelonia, and Rhyn- chocephalia. Nor is this all, for chameleons, which are commonly regarded as lizards, are considered by some natural- ists to form a third subgroup (Rhiptoglossa) of the Squamata, with the same value as the two respectively containing the lizards (Lacertilia) and the snakes (Ophidia). Recent reptiles and their immediate extinct kindred are consequently classed as follows :— 4 REPTILES Order I. Rhynchocephalia—Tuatera. II. Crocodilia—Crocodiles, alligators, ete. III. Chelonia—Tortoises and turtles. IV. Squamata. Suborder 1. Rhiptoglossa—Chameleons. 2. Lacertilia—Lizards. 3. Ophidia—-Snakes. When, however, extinct types are also taken into considera- tion, the list becomes greatly extended, and we have the follow- ing orders of leading groups, viz. :— Subclass I. Theromorpha—Mammal-like brigade. Order I. Pariasauria. IT. Cotylosauria. III. Anomodontia. Subclass If. Ornithomorpha—Bird-like brigade. Order IV. Rhynchocephalia—Tuateras. V. Pelycosauria. VI. Parasuchia—Belodonts. VII. Acrosauria. VIII. Squamata—Lizards and snakes. Suborder 1. Rhiptoglossa—Chameleons. 2. Lacertilia—Lizards. 3. Pythonomorpha — Sea-ser- pents. 4. Dolichosauria — Snake-liz- ards. 5. Ophidia—Snakes. IX. Chelonia—Tortoises and turtles, X. Placodontia—Bean-toothed reptiles. XI. Sauropterygia—Plesiosaurs. XII. Ichthyopterygia—Ichthyosaurs. XIII. Crocodilia—Crocodiles. XIV. Dinosauria—Giant reptiles. XV. Ornithosauria— Pterodactyles. It should be mentioned that a somewhat different arrange- ment of the two main divisions of reptiles has been proposed ; some authorities including within the first brigade (under the name of Synapsida) the Chelonia and Sauropterygia, on the ground that they agree with the Theromorpha in possessing single or united temporal arches, and including in the second GENERAL CHARACTERS 5 brigade, under the name of Diapsida, all the other orders, which are characterised primarily by the possession of double or separated temporal arches.' It has been pointed out, however, that it is illogical to separate widely from one another groups like the Chelonia, Sauropterygia, Rhynchocephalia, and Crocodilia, all of which agree in the possession of abdominal ribs (modified in the first into a true plastron); and, moreover, that the Theromorpha differ much more widely from all other groups of reptiles collectively than do any of the latter from one an- other. An alternative classification, in which some of the groups ranked above as orders are reduced to the grade of suborders, is the following :— Extinct Groups are marked with a ¢ Plesiosaurs VIII. + lcHTHYOPTERYGIA Ichthyosaurs ORDER SUBORDER 4 se 1 Dicynodontia & 00 I, +ANOMODONTIA : ( 2 Theriodontia S09 Anomodonts ( 3. Cotylosauria So , Pariasauria eas i ( -- ~ t + Protorosauria _ Il, Ray NCHOCEPHALIA (| 2 Rhynchocephalia Vera | TENSES. rt 3. Acrosauria | III. + PELycosauria | 1 Lacertilia | IV. SQUAMATA . : , | 2 Gerais | Snakes and Lizards | 3 Seka | 4 + Dolichosauria 5 + Pythonomorpha o -I Cryptodira S : V. CHELONIA . : : | 2 Pleurodira cleo Tortoises and Turtles | 3 + Amphichelydia a | 4 Trionychoidia g | VI. PLacoponTia = | Placodus ae) a | VII. + SAUROPTERYGIA | | eee ‘ 7 t Eusuchia [1 ix" Crocopmia. : cn Is i : Crocodiles \ ; lesa X. + DINosaurRIA . | 3 Sree Disasaurs l 3. Ornithopoda XI. + ORNITHOSAURIA L Pterodactyles The length of the foregoing lists, as compared with the one dealing solely with existing forms, must very materially modify ‘In the Squamata the lower temporal arch has disappeared. 6 REPTILES our conception of the relative importance of the class Reptilia in the animal kingdom; showing, as it does, that in place of com- paratively few, the class really contains a very large number of widely different structural types. A mere inspection of the lists does not, however, by any means reveal all that is implied as regards the differences between recent and extinct reptiles. For example, with the exception of crocodiles and their kin, giant land-tortoises and turtles, and pythons and anacondas, no modern reptiles can be regarded as really large animals, while the great majority of the class are comparatively small creatures. Even in the case of crocodiles and alligators a length of twenty-five feet is but very rarely, if ever, attained ; while thirty feet is an unusually large size for a python or an anaconda, although it is possible that individuals considerably exceeding these dimensions may now and then be met with. Again, as already stated, practically all the recent members of the class are characterised by the relative shortness of their limbs (when these appendages are present at all), and the consequent approximation of the lower surface of the body to the ground. Contrast the comparatively small dimensions and creeping gait of the majority of living types of reptiles with the huge bulk and the elevated or even upright position of the body in many of their extinct relatives of the Jurassic and Cretaceous epochs. The American dinosaur Dzplodocus, for instance, had an ap- proximate length of between sixty and seventy feet, and pro- bably stood not far short of twelve or thirteen feet in height; while its relative the iguanodon, when in its habitual upright position, towered to something like twenty feet. The body, too, of Diplodocus and its four-footed relatives was raised so much above the ground that a man:somewhat below the ordinary stature could have probably walked under its belly without much stooping.! During the same epochs the seas were inhabited by gigantic ichthyosaurs and plesiosaurs, which filled the place in nature now occupied by whales, and some of the larger and more typical representatives of which attained dimensions of between thirty and forty feet, or possibly even more. Larger still were those big-headed members of the plesiosaurian group known as pliosaurs, in which the length of ‘Some naturalists are of opinion that Diplodocus had a pose like that of a crocodile, with the belly close to the ground. GENERAL CHARACTERS 7 the lower jaw was close on six feet and that of the thigh-bone, or femur, fully a yard, from which some estimate may be formed as to the total dimensions of these monsters. As the plesio- saurs and ichthyosaurs began to wane, their place was filled by the sea-serpents, or Pythonomorpha, of the Upper Cretaceous, the length of some of which is estimated at not less than forty feet. Neither were these huge bodily dimensions by any means con- fined to the extinct Mesozoic ordinal groups, for we find many of the Lower Eocene, and even in some cases the later Tertiary, representatives of modern groups far exceeding in point of size their living relatives. The giant Hoche/one of the London Clay had, for instance, a skull fully three times the size of that of the modern leathery turtle, or luth (Dermochelys) which is itself one of the largest of living reptiles; while a snake (Gzgandéophis) from the Lower Eocene of Egypt probably fell but little short of fifty feet in length. Again, in the Lower Pliocene the gharial- like Rhamphosuchus of India has been estimated to have reached a length of between fifty and sixty feet ; while in the Pleistocene the giant monitor (Varanus priscus) of Queensland probably grew to something like fifty feet. Finally, as the dinosaurs or giant land-reptiles, and the earlier mammal-like theromorphs took the place during the Mesozoic epoch now held by terrestrial mammals, while the ichthyosaurs, plesiosaurs, and sea-serpents played the part of the whales of our modern seas, so the flying pterodactyles ful- filled the 7é/e of birds, which during the earlier part of that era of the world’s evolution were probably non-existent, while in the later stages of the same they apparently occupied a subordinate position and had not yet gained the dominion of the air. As the crocodiles, tuateras, and tortoises and turtles of the Mesozoic represented the reptiles of the present day, it is thus evident that the reptilian class, at that distant epoch, occupied the positions in nature now filled collectively by mammals, birds, and reptiles, and that the title “age of reptiles” which has been bestowed on the era in question is fully justified. “It is noteworthy,” writes Dr. A. Smith Woodward, “ that nearly all reptiles with well-formed limbs—whether adapted for habitual support of the body on land, for flight, or for con- stant swimming—flourished only before mammals and _ birds became dominant; the vast majority of the survivors during 8 REPTILES the Tertiary period and in the existing world being compara- tively degenerate types.” A few of the distinctive features of the reptilian class have been already mentioned ; these and others may now be some- what more fully noticed, especially when they display marked adaptive or evolutionary modifications. First, with regard to the single occipital condyle by means of which the skull is articulated to the atlas, or first, vertebra in all existing reptiles and probably also in all extinct ones with the possible exception of some of the anomodonts. In its most typical form, as exemplified by crocodiles and alligators, this condyle consists of a single knob formed exclusively by the basioccipital bone of the skull; it is then known as the typical monocondylic form. In many reptiles, such as iguanas, pythons, and turtles, lateral elements are developed to a larger or smaller extent from the exoccipital bones, and the structure then becomes of the tri- partite monocondylic type. In some tortoises (T7estado), and more especially in the anomodonts (such as Decyxodon and Cynognathus) the lateral elements tend to develop at the ex- pense of the median basioccipital element, which becomes creatly reduced, and we thus have the transitional dicondylic type, which, it is noteworthy, also exists in certain mammals, Finally, the total disappearance of the basioccipital element would result in the typical dicondylic form, or, in other words, in the double condyles of ordinary mammals. Such (to some- what anticipate matters) appears to have been the mode in which the double occipital condyles of mammals have been evolved from the single reptilian condyle. Here it may be mentioned that amphibians also possess double condyles, and it was accordingly at one time supposed that mammals took their origin from amphibians rather than from reptiles, but there are structural features which militate strongly against such a view. The single monocondylic type has, on the other hand, persisted in the evolution of reptiles to give origin to birds. The next most important feature in the skeleton of reptiles is the fact that, as in birds, the mandible, or lower jaw, consists of a number of separate bony elements, and articulates with the squamosal region of the cranium, or skull proper, by means of a quadrate bone. Much discussion has taken place as regards the fate of the quadrate bone when the theromorph reptiles GENERAL CHARACTERS 9 developed into mammals, in which the mandible articulates directly with the squamosal, and consists of only a single element on each side, corresponding to the dentary bone in the reptilian lower jaw. According to one of the latest author- ities! on this intricate subject, it seems probable that the reptilian quadrate has been taken up into the internal ear of mammals to form the incus, while the articular, or hindmost bone, of the lower jaw of the former has likewise been incor- porated in the same region of the ear, where it forms the malleus. It is added that a meniscus of cartilage found between ‘the condyle of the lower jaw and the squamosal in most mam- mals is a new element, unrepresented in reptiles. Another interpretation of the fate of the quadrate is, how- ever, suggested by Dr. R. Broom,’ who has paid particular at- tention to the relationship existing between the theriodont anomodonts and mammals. After pointing out that, according to his interpretation, the quadrate in the theriodonts is reduced to an exceedingly small bone affixed to the extremity of the squamosal, the author suggests that the reptilian quadrate is represented by the aforesaid meniscus, or interarticular cartilage, in mammals. “On the other hand,” he adds, “it is quite pos- sible that the quadrate is entirely absent in all mammals; yet the presence of a cartilaginous element in a situation exactly corresponding to that of the quadrate in theriodonts seems strongly to favour the view that in the meniscus we have the modified equivalent of the reptilian quadrate.” Whatever may be the ultimate verdict on this difficult question, it is quite evident that among the theriodont anomo- donts there must have existed forms in which the quadrate bone had more or less. completely lost its articular function, and in which the two branches of the lower jaw had well-nigh discarded all their separate elements situated posteriorly to the dentary. The mode in which this transition from the reptilian to the mammalian type has been effected is practically demonstrated by specimens figured in Dr. Broom’s paper. The quadrate, ac- cording to the author’s interpretation, forms a small knob to the descending process of the squamosal in the theriodonts, and has 1 Dr. Kjellburg, in Gegenbaur’s Morphologisches Fahrbuch for 1904. * Proceedings Zool. Soc. London, 1904, vol. i., p. 490. 10 REPTILES disappeared as a functional element in the mammal. In the true theriodonts the dentary element is much larger proportion- ally than in ordinary reptiles ; and while in Lycosuchus the sur- angular, angular, and articular elements of the hinder part of the lower jaw are fairly well represented, in Cynognathus the former has vanished, and the two latter are evidently on the point of following its example. In the words of the author :— “The examination of the theriodont lower jaw and of its mode of articulation show that the condition is already so nearly mammalian that only a very slight modification, and that very easily understood, is required to convert the therio- dont lower jaw into that of the mammal.” In certain of the earlier and most primitive types of reptiles, such as Pariasaurus and Procolophon among the mam- mal-like group, the upper surface of the skull is more or less completely roofed in by superficial, or membrane, bones, so that, as in the former of these, the only apertures in this roof are those for the eyes, the nostrils, and the median parietal (or interparietal) foramen. In this respect the skulls of these primitive types resemble those of the primeval salaman- ders, or labyrinthodonts (Stegocephalia). On the other hand, more advanced types show a gradual opening-out of this roof, as if portions had been cut away with a knife, till finally the whole of the upper surface of the cranium proper is exposed. In some cases a single temporal arch, as in the Anomodontia, Chelonia, and Sauropterygia, remains to protect and strengthen the lateral region of the skull, but more generally, as in Crocodilia and Rhynchocephalia, there are two such temporal arches, with a space between. The ac- companying illustrations show the closed and open types of skull. Among the more specialised orders, the Ichthyo- pterygia retain to a large extent the original in-roofing of the skull. Itshould be added that in the labyrinthodonts the upper surface of the skull displays a characteristic sculpture, which is retained in Partasaurus and possibly also in croco- diles. With regard to the above-mentioned parietal foramen, which occurs in many reptiles, and attains unusually large dimensions in the tuatera and the ichthyosaurs, it should be explained that this is a perforation in the forehead between GENERAL CHARACTERS i the two parietal bones which communicates with that region of the brain known as the epiphysis. It is also present in the primeval salamanders; and in the tuatera overlies the remnants of an aborted and degenerate eye. From a certain want of symmetry in this structure in the tuatera it has been inferred Fic. 1.—A, Upper surface of skull of a Labyrinthodont Amphibian to show the roofed type. B, Upper surface of skull of Tuatera (Sphenodon) to show open type. C, Front view of the lower end of the tibia of a bipedal Dinosaur with the closely- applied astragalus. The hole in Pf. is the parietal foramen. that the parietal eye was originally double; and that ancestral vertebrates were furnished with a pair of such eyes, which may have been serially homologous with the single pair of their descendants. It should, however, be added that no creature has hitherto been discovered whose skull exhibits apertures for this hypothetical second pair of eyes. In regard to limbs, it would appear that reptiles were originally four-footed creatures, with five toes to each foct. In 12 REPTILES the specialised and comparatively modern group of Squamata there is, however, a great tendency to the reduction of one or both pairs of limbs, this tendency culminating in the snakes, in all of which the limbs are altogether wanting, although re- presented by slight internal and external vestiges in certain families. Except in the more primitive types, the limbs of reptiles exhibit great plasticity, so that in many cases they be- come profoundly modified in accordance with the needs of adaptation to special modes of life. A feature in the tarsus of all the members of the bird-like brigade (Ornithomorpha) is that the ankle-joint occurs between the upper and lower rows of small bones forming that segment of the skeleton, and not, as in mammals, between the upper row of the tarsus and the lower ends of the tibia and fibula. But this is by no means all, for in certain dinosaurs which habitually assumed the upright posture the astragalus and calcaneum, forming the upper part of the tarsus, become closely applied to the tibia and fibula so as to form practically a single bone with each. Moreover, the tibia, with the closely applied astra- galus, may become the sole functional bone of this segment of the limb. In the figured specimen the astragalus sends up only a short process in front of the tibia, but in other instances, as in Dryptosaurus, this process becomes much larger and is in fact practically similar to that of the astragalus of birds, which in the adult condition becomes indissolubly fused with the tibia. Furthermore, in certain dinosaurs with three toes the three sup- porting metatarsal bones become closely applied to one another and likewise to the lower row of tarsal bones (reduced to two in number), thus combining, at all events occasionally (? patho- logically) to form a cannon-bone corresponding to the tarso- metatarsus of birds, And yet even this extreme instance is but a case of parallel adaptive modification for the same end (namely, the assumption of the upright posture), for it seems quite evident that these dinosaurs were not the ancestors of birds. Nevertheless other reptiles with a tibio-tarsus and a tarso-metatarsus, or, what is the same thing, birds with a separate tarsus and disunited metatarsals, must once have existed. Whether in the tarsus of the mammal-like brigade the ankle- joint normally occurred between the upper and the lower rows GENERAL CHARACTERS 13 of that segment, does not seem to have been ascertained. But here, again, it is clear that there must either have been some theromorphs with a mammal-like tarsus, or mammals with a tarsus of the reptilian type. As regards other features of the skeleton, it may be noted that a true sternum, or breast-bone, exists, and that in many instances the ribs are furnished with uncinate processes, that is to say from the posterior side of each rib proceeds an oblique process which overlaps the succeeding rib. These same uncinate processes occur in birds, and may be inherited from reptiles. The few features of the soft parts of reptiles to which space permits of allusion may be very briefly dismissed, since it is generally impossible to say whether they were common to the extinct types. Firstly, it is important to mention that epidermic scales—or their equivalents, large horny plates—form the char- acteristic covering of reptiles, and that the feathers of birds and the hairs of mammals are alike conspicuous by their absence. Here, in connection with the evolution of birds from reptiles, a great difficulty presents itself, for it seems almost impossible to imagine a scaly bird, while if we are driven to postulate a feathered reptile we are confronted with the difficulty of as- signing an adequate reason for the development of such a type of covering. Less difficulty is connected with the idea of a scaly mammal, or even of a hairy reptile. Cold blood is characteristic of all living reptiles, as is also the fact that the corpuscles of this fluid (like those of birds, but not of mammals) are furnished with a nucleus. The heart is in the main tripartite, but of such a type that the four- chambered organ of both birds and mammals might be easily evolved therefrom. Similarly, the presence of aright and a left aortic arch, permits us to see how, by the suppression of one or the other, the mammalian and the avian type of circulatory system have been respectively evolved. The gills of the young of the ancestral amphibians have been completely lost in reptiles so that, as in the two higher vertebrate classes, respiration, takes place by means of lungs alone. The excretory organs, as in birds and the lower mammals, discharge into a common outlet—the cloaca. During development the embryo is en- veloped in the two membranes respectively known as the amnion and the allantois; and the egg is of what is termed the 14 REPTILES meroblastic type, that is to say, only a part of its contents segments and enters directly into the formation of the embryo, the remainder serving as nutriment to the latter in later stages of its development. In both these respects reptiles differ from amphibians and resemble birds ; and it can scarcely fail to be noticed that throughout the organisation of the existing members of the class under consideration there is not a single feature which militates against their having been derived from amphibians, or it may be added, against their having given rise to birds on the one hand and to mammals on the other. CHAPTER Ni PEDIGREES AND RELATIONSHIPS Reptiles and their relationship to warm-blooded Vertebrates: the origin of reptiles. ‘The Theromorpha and their relation to the Mammalia: the Ornitho- morpha and their relation to birds. The tuatera: crocodiles and their ancestry : tortoises and turtles: flying reptiles: dinosaurs: snakes and lizards: geological history: plesiosaurs and ichthyosaurs: numerical strength. N the preceding chapter it has been mentioned that reptiles appear to have given rise on the one hand to mammals and on the other to birds; they are thus the parental stock of all warm-blooded vertebrates. Modern authorities are generally agreed that reptiles themselves are derived from the primeval salamanders, or stegocephalian amphibians, and that the evolu- tion of the class took place during or about the Lower Permian period, that is to say towards the close of the Paleozoic era. There is, however, some difference of opinion as to whether the origin of reptiles from amphibians was single or dual (mono- phyletic or diphyletic), and also as to the exact limitations of the two classes. Dr. H. Gadow, for instance, in the Cambridge Natural Fiistory includes in the Reptilia certain Permian forms, such as Hryops and Crécotus, which are usually regarded as stego- cephalian amphibians, to form a group termed the Proreptilia, which is regarded as the starting-point from which all true reptiles have sprung. He also classes as reptiles the so-called Microsauria, which are likewise generally considered to be stegocephalians, brigading them with the Rhynchocephalia (Protorosaurus, Sphenodon, etc.) to form a group—Prosauria— connecting the Proreptilia with other reptiles. He thus _pos- tulates a monophyletic origin for the Reptilia in general. On the other hand, another authority, Mr. G. A. Boulenger, in an article published in the Proceedings of the Zoological Society of London for the year 1904 refused to admit Dr. Gadow’s Proreptilia and the Microsauria into the class of md T5 EO!» REPOILES reptiles, for which he suggests a dual origin from two distinct groups of the Stegocephalia, namely the Labyrinthodontia on the one hand, and the Microsauria on the other. Although in one sense such diversities of opinion are a matter for regret, yet like political parties, they have their ad- vantage, for in this particular instance they tend to show, even in the present crude and imperfect condition of our knowledge, the existence of a kind of border-land, or “ buffer-state,” between stegocephalian amphibians and reptiles, some of the inhabit- ants of which may be referred to the one or the other of these groups according to the idiosyncracy of the particular writer who may be attempting their classification. Assuming, then, that we are right (following Mr. Boulenger) in dividing reptiles into two main divisions, or brigades, it may be considered not improbable that while the mammal-like divi- sion (Theromorpha) may trace its origin to the Labyrintho- dontia, the bird-like brigade (Ornithomorpha) may have sprung from the Microsauria.! The Theromorpha, which appear to be confined to the Per- mian and Triassic periods, are connected with the labyrintho- donts by means of the Pariasauria, as represented by the huge amphibian-like reptiles from the Trias of Africa and Russia constituting the genus Pariasaurus. In this group the skull was as completely closed in by roofing bones as in the labyrin- thodonts (Fig. 1), but its under surface had lost the para- sphenoid bone characteristic of amphibians in general. From Pariasaurus there is a transition in one direction to the Cotylosauria, as represented by Pariotichus, Empedias, Diadectes, Pelerpetum, and Procolophon which still have the temporal region of the skull more or less completely roofed over with bone, but which differed from the Pariasauria by the greater number of joints in the toes. Some of these reptiles, it may be remarked, had the cheek-teeth expanded transversely, which suggests kinship with mammals, despite the fact that the group appears to have died out without descendants, being merely a side-branch from the Pariasauria. On the other hand, the latter group appears to have also given rise to the Anomodontia, which apparently included, in ‘Since this was written fuller details as to the relationships of the Micro- sauria have been published by Mr. R. S. Moodie. PLATE I SHTUILdAY AMITIVNNVN AHL JO AdAL TWULSHONY HHL SOYNVSVINVd JO NOLATAMS 453+ eee ae PEDIGREES AND RELATIONSHIPS 7. the form of the Theriodontia, the ancestors of mammals, and are characterised, among other features, by the opening-up of the temporal region of the skull, with, as already mentioned, the formation of a single temporal arcade, which, however, at least in some cases, as Cynognathus, for example, consists of two elements. To describe in detail the mammalian features presented by the anomodonts and other Theromorpha, or indeed the general characteristics of that subclass, would be cut of place on the present occasion, It may be mentioned, however, that the skull, with the exception of the retention of a quadrate, a com- pound lower jaw, and a prefronta] bone (which is a character- istic reptilian feature) in many cases might almost be described as that of amammal. The reduction in the size of the quadrate and of the posterior elements of the lower jaw in this group has been already noticed. In the theriodont, or carnivorous, section of the anomodonts, as exemplified by Cyzognvathus and Galecynus, the teeth recall those of carnivorous mammals, The shoulder and pelvic girdles are also essentially of the same type as those of the monotreme, or egg-laying, mammals, and present, moreover, a remarkable serial homology; that is to say, the three elements of the one (scapula, epicoracoid, and coracoid) correspond almost exactly with those of the other (ilium, pubis, and ischium) while the obturator foramen, or perforation between the pubis and ischium is an essentially mammalian feature. Nor is this all, for the theromorph shoulder and pelvic girdles are essentially unlike those of other reptiles, this being especially the case with the pelvis, in which among many reptiles, such as dinosaurs, the pubis and ischium, are long divergent rods. The humerus, or upper arm-bone, is also essentially like that of the lower mammals, having a strongly developed deltoid or radial crest in its upper half, and a perforation (entepi- condylar) on the inner, or ulnar, side of its lower extremity. Not less important is the fact that the number of joints in the toes, at least in several cases, is the same as in mammals; while it would seem that, as in that group, the ankle-joint was situ- ated between the tibia and fibula and the first row of the tarsus. In concluding these observations on the relations between reptiles and mammals, the following passage from a paper 2 18 REPTILES published by Professor H. F. Osborn in the American Natural- ¢st for 1898 may be quoted :— “Important, also, among the resemblances between the Theriodontia and Mammalia is the general bodily form, so far as it is known in the former, the proportions of the limbs to the back, and the apparent elevation of the body considerably above the ground. This, taken together with the peculiar special- isation of the teeth into carnivorous and herbivorous types, indicates that the Theriodontia filled somewhat the same 7é/e in the economy of nature as is filled by the Mammalia at the present time. The most striking general difference is the very large size of several of these animals, such as Cynoguathus. We had rather anticipated from our knowledge of the earliest Stones- field (Lower Jurassic) mammals that their reptilian ancestors would have been very small. ‘The large size of these Permian (or Triassic) theriodonts is, however, not incompatible with the hypothesis that smaller and less specialised members of the sroup may have constituted a persistent phylum (or branch).” Passing on to a proportionately brief survey of the probable evolutionary history of the numerous ordinal groups included in the subclass Ornithomorpha, it may be repeated that the most primitive and at the same time the most ancient of these orders is undoubtedly the Rhynchocephalia, which includes the Permian Protorosaurus, and may have sprung directly from the microsaurian amphibians. The Rhynchocephalia, now surviv- ing only in the form of the New Zealand tuatera (Sphenodon punctatus, Plate I1.), are characterised, among other features, by the presence of complete upper and lower temporal arcades to the skull, which also shows a large parietal foramen, by the bodies of the vertebrae being cupped at both ends, by the pres- ence of intercentra between many of the vertebra, and of chevron-bones attached to the lower surface of those of the tail, by the fully developed shoulder-girdle and five-toed limbs, the firmly fixed quadrate bone, the acrodont teeth, and the presence of a foramen (entepicondylar) on the inner side of the lower end of the humerus, and of a system of abdominal ribs, forming an incipient plastron on the lower surface of the body. Of the numerous divergent branches which have been given off from the rhynchocephalian stock, one has apparently culminated in the birds. Among the fossil representatives of the group may be PEDIGREES AND RELATIONSHIPS 19 mentioned Palgohatterta of the Lower Permian of Saxony, a long-tailed reptile of about a foot and a half in length, and the aforesaid Proforosaurus of the Upper Permian of Thuringia and Durham which attained a length of between four and five feet. These are the earliest and most primitive types. More nearly allied to Sphenodon is Rhynchosaurus of the Upper Trias, or Keuper, of Shropshire and Warwickshire, and /Hyferodapedon, with its pavement-like series of upper teeth, from the Upper Trias of both England and India. Much nearer to the living type is Homeosaurus of the Upper Jurassic of the Continent; while in the Upper Cretaceous the group is represented by the long-snouted Champsosaurus, Very interesting is Proterosuchus of the South African Trias, which appears to be a rhyncho- cephalian showing a considerable degree of specialisation along the line which gave rise to the crocodiles. Very few words must suffice for the Pelycosauria, a group of chiefly North America Permian reptiles, whose affinity seems to be with the Rhynchocephalia, from which they apparently form a side-branch. They have frequently been confounded with the theriodont anomodonts, to which they assimilate in their dentition. The quadrate-bone is remarkably small and surrounded by the adjacent elements ; and in the typical forms, such as Vaosaurus, the spines of the dorsal vertebrz, as will be more fully noted in the sequel, are enormously elongated. The dentition is thecodont. , Next in the direct line of descent from the rhynchocepha- lians may be placed the Upper Triassic belodonts, or Para- suchia (sometimes, although incorrectly, termed Thecodontia). These reptiles, as typified by Phytosaurus (or Belodon), are fre- quently classed with the crocodiles, with which they agree in having the teeth set in distinct sockets, and in possessing a pitted dermal bony armour ; the teeth, as in crocodiles, having hollow roots in which the germs of their successors are deve- loped, like a “nest” of thimbles. As a matter of fact, the belodonts appear to be equally nearly related to a number of others. They resemble, for instance, pelycosaurians, typical dinosaurs, and crocodiles in their socketed teeth; and thereby differ markedly from rhynchocephalians. On the other hand, they agree with the latter group and with pelycosaurians, and differ from crocodiles and dinosaurs in the retention of distinct 20 REE TIEES clavicles, or collar-bones ; and also approximate to the rhyn- chocephalian type in their system of abdominal ribs. A marked difference from crocodiles is to be found in the circum- stance that three elements of the pelvis (ilium, pubis, and ischium) enter into the composition of the acetabular cavity for the reception of the head of the thigh-bone, or femur—a generalised feature they possess in common with dinosaurs, Phytosaurus itself was a reptile of the dimensions of a good- sized crocodile, with an enormously long snout, at the base of which opened the nostrils. The belodonts are remarkably interesting reptiles, for in a provisional genealogical tree published in the Zoological Society’s Proceedings. for 1904 they occupy the intermediate position between rhynchocephalians and birds, being placed, however, much closer to the former than to the latter group. As a matter of fact, we know at present nothing in any way inter- mediate between belodonts and birds, between which there is evidently a very long gap; and we are consequently totally ignorant of the mode of evolution of the many peculiarities in a bird’s skeleton. Not that this affords any argument against the evolution of birds from reptiles, any more than does the fact that we are ignorant of the direct ancestors of the ptero- dactyles, and consequently the mode of evolution of the skeleton of their wings. There are several notable features in the skull of the belo- donts, in addition to the backward position of the opening of the posterior nostrils. In the first place, the comparatively small size of the apertures in the temporal region is a primitive feature, very different from what obtains in the crocodiles as shown in the figure of the skull of Wetriorhynchus. A second feature is the creat length of the premaxilla, which is also in marked contrast to what obtains in crocodiles. Reverting to the Parasuchia as a starting-point, we may probably regard the ichthyosaurs (Ichthyopterygia) as a side branch from this stock, in which specialisation has been con- centrated on the needs of an aquatic existence. Ichthyosaurs agree with belodonts in their complete clavicles, but their teeth are implanted in grooves, and evidence of kinship with stego- cephalian amphibians is manifested in the retention of a partial roof to the temporal region of the skull. Another early side PEDIGREES AND RELATIONSHIPS 21 branch from the belodonts is that of the crocodiles which retain the dermal armour and socketed teeth, but have lost their clavicles and the pubis has become excluded from the aceta- bular cavity of the pelvis. Specialisation early displays itself nO 0° OC 46 e = 5605: Me) = 00 nt qo00Geed9 O00 | — 0” 9850006000 °°° a See o © R, 00000 3 of SS200090009/000005, sXe) ie] Fic. 2.—Upper (A) and lower (B) views of the skull of a Belodont (Phyto- saurus) a primitive reptile from the Upper Trias: nar. nostrils; p. na. posterior nostrils ; pmx. premaxilla. C. Upper view of skull of an extinct Crocodile (Metrio- rhynchus) to show the open temporal region and the small premaxille as con- trasted with the corresponding parts in Phytosaurus. in the backward transference of the posterior openings of the nostrils into the mouth, by means of the development of a false floor to the palate of the skull, but it was not until the Tertiary period that this feature obtained its full development. The culmination of this adaptive modification was attended by the 22 REPTILES development of ball-and-socket (instead of nearly flat or slightly hollowed) articulations to the vertebrz ; the socket being in front and the ball behind, thus constituting the proccelous type of vertebral structure. From a branch arising directly from the rhynchocephalian stock, without the intervention of the belodonts, may probably be derived plesiosaurs (Sauropterygia) and tortoises and turtles (Chelonia); these two orders having many points in common, such as a single (upper) temporal arcade, a fixed quadrate-bone, the mode of articulation of the ribs to the vertebral column, and the presence of clavicles, and of either a true plastron or a system of abdominal ribs, Specialisation has, however, taken different lines in the two groups: in the one case the protection of the body in a shell, together with the loss of the teeth; while in the other socketed teeth are retained, and the limbs in the more specialised forms become modified into paddles. Although the plesiosaurs can be traced downwards into terrestrial or fresh- water forms, we are at present quite ignorant of the ancestry of the chelonians. The placodonts, or bean-toothed reptiles (Placodontia) of the Trias, were long regarded as members of the theromorphous, or mammal-like brigade, but a later view is to consider them as members of the branch which gave rise to the chelonians and plesiosaurs, with both of which they agree in the general characters of the skull, while they resemble the latter in their socketed teeth. A peculiar feature, as will be noticed later, is the extension of the teeth on to the palate, and their crushing type of crown. A totally independent branch from the main stem is formed by the pterodactyles, or flying reptiles (Ornithosauria), the whole of whose organisation is profoundly modified for the purpose of aerial flight. Pterodactyles retain evidence of affinity with the ancient parasuchian (and thence with the rhynchocephalian) stock in the retention (except in the more specialised forms) of socketed teeth, fixed quadrates, double temporal arcades, and distinct clavicles. Although the group cannot at present be traced into direct connection with terrestrial, non-volant reptiles, it is quite clear that it has nothing to do with birds. The branch which has given rise to the largest of all reptiles is that of the Dinosauria, which, in the opinion of some author- : ' | ‘ | ) 1 [ PEDIGREES AND RELATIONSHIPS 23 ities, ought to be split into two separate orders, namely the Dinosauria proper, as typified by the carnivorous and thecodont Megalosaurus and its allies, and the Orthopoda, to include the herbivorous types, in which the teeth have more or less com- pletely lost the original socketed character. Whichever view be adopted, it is evident that all these reptiles are very closely related, and that it is the carnivorous group which connects them with the belodonts (Parasuchia). With that group dinosaurs agree in their double temporal arcades, fixed quadrates, and the inclusion of the pubis in the acetabular cavity of the pelvis, although they have lost the clavicles. The more specia- lised dinosaurs, like crocodiles, have developed ball-and-socket articulations to the vertebrz, but it is remarkable that the posi- tions of the balls and sockets are just the reverse of those in the latter group, the socket being at the hind and the ball at the front end of the vertebral column, thus forming the opistho- ccelous (in contradistinction to the proccelous) type of verte- bral structure. A few words may be devoted in this place to the supposed direct phylogenetic relationship between birds and dinosaurs. When the many curious resemblances undoubtedly existing between the bones of the pelvis and hind-limbs of those dino- saurs which habitually assume the upright posture, such as the incipient union between the lower end of the tibia and the as- tragalus, and the consequent tendency towards the formation of a tibio-tarsus and a tarso-metatarsus, were first brought to light, it was confidently assumed that dinosaurs represented the ancestral stock from which birds had originated. This view was strengthened by the fact that primitive birds were furnished with teeth implanted in sockets after the megalo- saurian type. Neither was it weakened by the circumstance that birds have a free quadrate-bone and only a single (lower) temporal arcade; the loosening of the attachment of the quadrate being doubtless correlated with the loss of the upper temporal arcade. A more serious, although not insuperable objection, is the retention of clavicles by birds. When, however, the doctrine of parallelism in development, owing to structural adaptations for more or less nearly similar modes of life, became generally recognised as a powerful factor in the evolution of animals, opinion veered in the opposite 24 REPOTEES direction, and it was considered by many high authorities that the resemblances existing between the pelvic and limb skeletons of dinosaurs and birds were solely the result of adaptation, and were not indicative of genetic relationship between the two groups. It is true, that a guarded protest was raised by the celebrated American palaeontologist Professor H. F. Osborn in a paper published in the Ammerzcan Naturalist for 1900, In this communication it was remarked that the passage from a quadrupedal to a bipedal mode of progression would mark the transition from the protorosaurian rhynchocephalians (the ad- mitted ancestors of the ornithomorphous brigade of reptiles) to dinosaurs, and consequently “that our present knowledge and evidence justify us in saying that in this bipedal transition, with its tendency to form the tibio-tarsus, the avian phylum may have been given off from the dinosaurian ”’. Later opinion has, however, on the whole been markedly against this partial reversion to the older theory of the descent of birds from dinosaurs. Another well-known authority, Dr. H. Gadow, points out, for instance, in the volume on reptiles in the Cambridge Natural History that those dinosaurs which exhibit the most marked avian resemblances did not come into existence until about or after the time that birds had been fully differentiated. After putting these later types out of court, the author proceeds as follows :— “There remains only Axchisaurus of the Upper Trias, more or less contemporary with Lrontozoum, which left its three-toed foot-prints (Arch@opteryx' has four well-developed toes) with Zanclodon. Moreover, the most bird-like foot is either that of the Theropoda [MWegalosaurus, etc.] which, like Anchisaurus and Zanclodon, differ from birds by the formation of the pelvis, or of some of the latest Ornithopoda [| /guwanodon, etc.]. What, then, is the good of selecting a number of bird-like features from members of dinosaurs which we are bound to class in different groups, and which existed, some in the lower, others in the middle, or even in the latest Mesozoic periods?” Dr. Gadow’s contention appears to receive support from the British Museum specialist on reptiles, Mr. G. A. Boulenger, who, in a paper in the Proceedings of the Zoological Society for 1904, to which reference has been already made, derives ' The toothed and long-tailed Upper Jurassic bird, PEDIGREES AND RELATIONSHIPS 25 dinosaurs from a side branch of the stem connecting belodonts with birds. According to this view the proximate reptilian ancestors of birds still remain to be discovered. Possibly it was in Africa that the annectant forms were developed. The last branch of the Reptilia is the one culminating in the Squamata (snakes and lizards), which, like plesiosaurs and chelonians, appears to be directly derived from the primitive rhynchocephalian stock, with which it is connected by means of the Acrosauria, a group provisionally admitted to ordinal rank, and typified by the lizard-like Acvosaurus and Pleurosaurus of the Upper Jurassic of the Continent. The Squamata have either retained the ancestral acrodont dentition of the Rhyncho- cephalia, or have modified this into the pleurodont type. They have also inherited the ancestral clavicles; but, on the other hand, they have completely discarded the rhynchocephalian plastron, or abdominal ribs and also the uncinate processes of the true ribs. They have likewise lost the lower temporal arcade, that is to say, the one connecting the quadrate by means of the quadrato-jugal and jugal with the maxilla ; and probably cor- related with the loss of this bar is the loosening of the attachments of the quadrate itself, which is movably attached to the skull, and thus makes a loose hinge for the lower jaw, which is of special value to the pythons, boas, and other snakes which gorge their prey in bulk. The complete scaling of the body is, as its name implies, a very characteristic feature of the order, although in some cases this has been lost. Next to fishes and amphibians (as represented by the Ste- gocephalia), reptiles are the oldest vertebrates ; in other words, they are the oldest of the amniotic vertebrates. Bearing this in mind, and also remembering the fact that (unlike fishes, which are restricted to the water) during the Mesozoic epoch they alone filled, in the main, the places now occupied in nature conjointly by terrestrial and aquatic mammals, birds, and reptiles, it is not surprising to find that, with the progress of time, they have lost more heavily in ordinal groups than any other vertebrate class. They have, in fact, been much more than decimated, for out of the fifteen orders recognised in the table on page 4 only four, as already said, now exist, and of these one is represented only by a single genus and species, or, at most, by two or three closely allied species of that genus, 26 REPOIVES Of the various groups of reptiles, the entire subclass Theromorpha appears to have been confined to the Permian and Triassic periods ; its oldest known representatives being probably those from the Permian of Texas. Whether the latter strata are somewhat older or newer than (it is unlikely that they are the exact equivalent in age) the Lower Permian of Europe, cannot be determined ; and it is therefore impos- sible to say whether the Theromorpha are older than Ornitho- morpha, or vce versd. Be this as it may, the Theromorpha, as represented by the Pariasauria, probably originated from the labyrinthodont stegocephalians in the Permian, or, at all events, in the later Carboniferous; while the Theriodontia probably gave rise to mammals in the Trias. Having achieved this triumph, their task appears to have been accomplished, and they gradually waned and finally disappeared from the scene. Previous to, or about, the time that they gave rise to the Mammalia, the Theromorpha appear to have been the dominant forms of terrestrial vertebrate life, the contemporary rhyncho- cephalians, like most of their successors, being comparatively small creatures. For a long period their descendants the mammals were, however, at least over a large part of the world, in a subsidiary position ; and did not in fact regain the in- -heritance of those reptilian ancestors as the lords of creation till the Tertiary period. Turning to the second, or bird-like brigade (Ornithomorpha), we find that the most primitive order, namely the Rhyncho- cephalia, is the most ancient, or, at all events, as ancient as the nearly related Pelycosauria (on the supposition that it is the ancestor of all the class, it must of course be the oldest). As already said, its oldest known representatives are Pal@ohatteria of the Lower Permian and Protorosaurus from a somewhat higher stage of the system; the presumed evolution from the microsaurian stegocephelians having probably taken place in either the Lower Permian or the Upper Carboniferous. In the Trias the rhynchocephalians were well represented, and they are known by several forms from the Jurassic, while there is evi- dence of their existence in the Cretaceous. The links between the Jurassic forms and the existing New Zealand species are, however, still unknown, and may not improbably have flourished in southern lands which have long since been sub- PATE LL THE NEW ZEALAND TUATERA (SPHENODON PUNCTATUM) THE SOLE SURVIVOR OF THE MOST PRIMITIVE AND MOST ANCIENT GROUP OF THE BIRD-LIKE REPTILES TOTAL LENGTH ABOUT I5 INCHES A GIRDLE-TAILED LIZARD (ZONURUS G/JGANTEUS) OF SOUTH AFRICA, TO SHOW APPARENT SIMILARITY OF LIZARD AND TUATERA PEDIGREES AND RELATIONSHIPS 27, merged. The Pelycosauria, which, as already said, formed a specialised side branch from the rhynchocephalian stock, appear to have waxed and waned within the duration of the Permian period, when they were represented on both sides of what is now the Atlantic. The mantle of the Rhynchocephalia may be said, however, to have fallen on the Squamata (lizards and snakes), which is at the present day the most numerously represented of all rep- tilian orders, holding a position in the class analogous to that occupied by the perching birds in the class Aves. Here, then, we may trace a parallel as regards evolutionary history between the anomodonts on the one hand and the rhynchocephalians on the other. As mammals—the descendants of the anomodonts —after passing through a long epoch when they were, so to speak, under a cloud, eventually rose to pre-eminence, so the Squamata, after a period of subordination, eventually attained the dominant position among reptiles of to-day. If we regard birds as likewise the direct descendants of the rhynchocephalians, the parallelism between that group and the anomodont-mammal line is still more remarkable. As to the epoch when the rhynchocephalian type blossomed out into that of the Squamata we are still in the dark. If the Acrosauria formed the connecting link, the evolution must have been during or later than the Upper Jurassic. Squamata are known from the Cretaceous, but as these (Dolechosaurus) are marine types they almost certainly imply the existence of earlier terrestrial forms, unless indeed they be independently derived from the aquatic rhynchocephalian Champsosaurus, From the early Tertiary remains of both lizards and snakes are known: and since the Egyptian Lower Eocene Gigantophis cromeri was of very large size, it is practically certain that there were earlier types of inferior dimensions. The belodonts, or Parasuchia, appear to have been solely of Triassic age. The ichthyosaurs, on the other hand, ranged from the Trias to the Upper Cretaceous, and may possibly have survived in some parts of the globe into the earliest Tertiary. Even the Triassic forms appear to have been marine, and we have consequently yet to find the connecting links between this group and earlier reptiles. The members of this order had attained their maximum development in point of size as early 28 REPRILES as the Lower Lias; but as some of the Cretaceous more special- ised species were equally large, the dying-out of the group seems to have been sudden, as there is no sign of decadence or degeneration. The cause of this apparently sudden exter- mination is altogether beyond our ken! The geological history of the plesiosaurs (Sauropterygia) is very similar to that of the ichthyosaurs, except that the Triassic types were smaller and more nearly akin to terrestrial forms, and that the largest species occurred in the Upper Jurassic and Cretaceous periods, although some of those from the Lias were gigantic. The small group of placodonts (Placodontia) appears to have been solely Triassic. Chelonians, as represented by Proganochelys and Chelyther- zum, date from the Upper Trias, but these forms are as typically chelonian as is any modern tortoise, so that we have evidently to go further back before the order can be traced to the ancestral stock. From the Trias chelonians appear to have steadily in- creased in numbers to the present day, when they form the second largest reptilian order. The Chelonia of the Jurassic strata belong to a generalised group known as the Amphichel- ydia, from which appear to have diverged the modern Crypto- dira and Pleurodira, of which the latter is now on the wane. The origin of the soft-tortoises, or Trionychoidea, is unknown, but they were abundant in the early Tertiary, and also occur in the Upper Cretaceous. True marine turtles (Chelonide) also date from the Upper Cretaceous, and were preceded by ancestral forms in the Upper Jurassic. Upon the question whether the Triassic Psephoderma is really chelonian or not, largely depends the problem as to the nature of the relationship between the true turtles (Chelonzde) and the leathery turtles (Dermochelyida) of our modern seas. The pterodactyles, or Ornithosauria, according to our present information, date from the Lower Lias and continued right through the overlying Mesozoic formations. Since, however, their earliest known representative—Dmorphodon macronyx of the Lower Lias of Dorsetshire—is in all respects a typical and fully evolved member of the group, it can scarcely be doubted that their origin must be looked for at an earlier epoch of the earth’s history. Beyond the fact that some of the more specialised forms of the later formations have lost PEDIGREES AND RELATIONSHIPS 29 their teeth and diminished the length of the tail, it can scarcely be said that the group exhibits much signs of progres- sive evolution above its Liassic prototype. Indeed, there is no reason that it should, seeing that the species in question ap- pears to be thoroughly adapted in every particular for a life in the air and predatory habits. In some of the gigantic Cretaceous forms alluded to in the sequel, specialisation, in correlation with the enormous dimensions of the wings, shows itself in the welding of the scapula to the dorsal vertebra. Another form of specialisation is shown by the development of a ring of bones in the eye. The dinosaurs (Dinosauria), like the ichthyosaurs and plesio- saurs, commenced in the Trias and continued to flourish till the close of the Cretaceous period, when some of their members, as is more fully noticed in the sequel, underwent some most strange and bizarre developments. It is specially important to notice that all the known Triassic forms belong to the typical carni- vorous group (Dinosauria proper), commonly known as Thero- poda ; the more specialised herbivorous types being a later development. It is largely on account of this fact that all the groups are included in a single order in the present work ; although many naturalists prefer to consider the herbivorous forms as constituting an order by themselves. Another very remarkable circumstance is that some of the Triassic represen- tatives of the order had already assumed the bipedal mode of progression. This early assumption of the erect posture makes it somewhat difficult to accept a theory suggested by Professor H. F. Osborn, of New York, to the effect that birds were prob- ably an offshoot from the dinosaurian stock before the posture in question had been attained by any of the members of the latter. The huge herbivorous dinosaurs of the group Sauropoda are known to have been in existence in the early part of the Jurassic period (exclusive of the Lias); while the armoured, or stegosaurian, section of the second great herbivorous sroup (Ornithopoda) date from the Lower Lias; these Liassic forms being much less fully armoured than were the Jurassic and Cretaceous successors. The highly specialised iguanodonts date apparently from the Upper Jurassic and Wealden. To attempt to make a census of the known forms of extinct reptiles would be, in the first place, a matter of extreme diffi- 30 REPITEES culty, while, in the second place, if this were accomplished, it would be quite valueless, It would be difficult because the records are much scattered, and a number of generic and specific names have been assigned to specimens of so imperfect and unsatisfactory a nature that in many cases it is impossible to say whether or no they are mere synonyms. It would be valueless because the known forms of extinct reptiles must, from the very nature of the case, bear but a small proportion to the number of species that have existed. Nor is this all, for in order to make the census of the slightest use it would be necessary to divide these known forms into time-horizons, the number of which it would be extremely difficult to fix satisfactorily. Neither, in the present state of zoological science, is a census of the existing members of the class altogether free from some degree of uncertainty ; for very different views of the limitations of species are entertained by different naturalists. In accepting, however, the number of species given in the volumes of the British Museum Catalogues we shall certainly be on the safe side, as the author of the latest editions of such of those works as deal with reptiles is a naturalist who takes a broad view of the limitations of a species, and is not given to hair-splitting in this respect. It must be remembered, however, that these works are now somewhat out of date in respect to the number of known species of reptiles, the last volume of the catalogue of lizards having been published so long ago as 1887, while the one on crocodiles and chelonians was issued a year later, and the concluding volume of the snake-catalogue in 1896. Taking, however, the data given in these volumes, we have the following numbers for the five chief divisions of existing reptiles, namely :— Tuateras (Rhynchocephalia) : : : I Crocodiles (Crocodilia) ; ; 23 Tortoises and Turtles (Chelonia) : 201 Lizards (Lacertilia) . : : : 1616 Snakes (Ophidia) : : : 1639 This gives a total of 2480 species, a number which, by subse- quent additions to the list, may be safely increased to at least 2600. PEDIGREES AND RELATIONSHIPS 31 Although this number largely exceeds that of the Amphibia (frogs, toads, salamanders, etc.), which was 633 in 1882, and now, perhaps, approaches 700, it is incomparably inferior to that of the other three great vertebrate classes. For instance, so long agoas the year 1870, Dr. A. Ginther estimated the number of known species of fishes at 9000, and at the present day the estimate would be very much larger. As regards birds (in which, it should be observed, species are often separated on slight characters), the number in the passerine and “ picarian” groups alone was given in 1900 as 6487. Of mammals (to which the same remark applies) considerably over 7000 species, inclusive of extinct forms, were catalogued in 1899 by Dr. FE. L. Trouessart, and the number has since been increased. One feature is very noticeable in all the five vertebrate classes, namely the predominance in point of numbers of one particular order or group at the present day; the great majority of the representatives of such orders being of comparatively small bodily size, although to a certain extent fishes and, among reptiles, snakes form exceptions in the latter respect. Thus in the class Pisces we have the Teleostomi (bony fishes), in the Amphibia the Ecaudata (frogs and toads), in the Reptilia the Squamata (lizards and snakes), in Aves the Passeres (perching birds), and in Mammalia the Rodentia (rodents), as the overwhelmingly predominant group in point of numbers. Each one of these groups, it should be added, is of a highly specialised and essentially modern type. CHAPTER III HAUNTS AND HABITATS The haunts of reptiles in past times, and at the present day. The sea- iguana. Dearth of reptiles in polar and sub-polar regions. Haunts of tortoises and snakes. Tree-dwellers. Burrowers. Island forms. Geographical distribu- tion, difficulties of interpretation. Relation to temperature, moisture, etc. Hiber- nation and zstivation. URING the earlier and middle stages of their long and eventful history, when they were the dominant verte- brates other than fishes, reptiles, as will be more fully shown in the sequel, were adapted to occupy every station on the earth, and to fill every 7é/e in life. The ichthyosaurs and plesiosaurs, and later on the extinct sea-serpents (Pythono- morpha), together with numerous turtles and a peculiar group of crocodiles, were denizens of the ocean, in which the three first-named groups played the part now taken by whales in the scheme of creation.' The pterodactyles, or flying saurians, were, on the other hand, inhabitants of the air, where, so far as their power of flight permitted, they took the place now occupied by birds. It is, however, improbable that any of these creatures were adapted to a purely pelagic existence comparable to that of the albatross and the tropic-bird at the present; and, indeed, we do not even know whether they were capable of swimming on the surface of the sea after the fashion of modern sea-birds. Again, to what extent they were adapted to an inland life is likewise uncertain, the great majority of their remains having been obtained from shore-deposits, such as the Lias, although some have been found in the Chalk which was probably de- posited in deeper water. It is true indeed that in the Stones- field Slate of Oxfordshire remains of pterodactyles are found in association with those of mammals and other land animals, but 2 The Middle Jurassic Ofhthalmosaurus appears to have been more com- pletely pelagic than any of the other ichthyosaurs. 32 aw HAUNTS AND HABITATS 33 this by no means implies that they themselves were dwellers in an inland district. As a matter of fact, the Stonesfield beds are a lagoon-deposit, into which the remains of land animals and plants were washed down by rivers; and the pterodactyles were probably inhabitants of the shores of the lagoon itself. This inference is supported by the fact that pterodactyle re- mains are unknown from the purely fresh-water and terrestrial deposits of the Dorsetshire Purbeck. Consequently, it seems probable that pterodactyles were in the main frequenters of coast-lines, and that they did not take the place of birds in in- land districts. This is the more likely seeing that at any rate as early as the Upper Jurassic true birds (Arch@opteryx) were already in existence. On land, throughout the Mesozoic epoch, huge dinosaurs played the part of the great land mammals of the later Tertiary period and the present day; while rhynchocephalians, fresh-water tortoises, and crocodiles fulfilled the 7é/e of lizards and snakes and the modern representatives of-the last-named group. Whether there were arboreal reptiles (other than pterodactyles) during the Jurassic and Cretaceous periods, we have no means of knowing. At the present day all this predominance and exuberance of reptilian life have completely passed away. The dominion of the air, or such claim to the same as could be maintained by the pterodactyles, has been completely lost, not a single existing reptile having the power of true flight. Much the same may be said with regard to the “command ofthe sea” formerly possessed by the reptilian class, for at the present day there are no truly pelagic reptiles (if we except the mythical sea-serpent) save the sea-snakes, none of which exceeds five or six feet in length, and most of which are less. Turtles—both leathery and otherwise —are it is true in the main pelagic reptiles, but they go ashore to deposit their eggs, and are thus not comparable to ichthyo- saurs and plesiosaurs on the one hand and to whales on the other. The only other reptiles which can in any true sense be called marine are the Indian Cvocodilus porosus and the Galapagos sea-iguana (A mblyrhynchus cristatus); but these spend a large portion of their time on land, and the former is by no means exclusively a denizen of the coasts but is rather an amphibious land reptile which takes partly to a marine life. It should be 3 34 REPTILES added that the great water-monitor (Varanus salvator) of India will, when frightened, occasionally rush headlong into the sea. Apart from the air and the ocean, the reptiles of the present day have availed themselves of most stations suited to their mode of life over the greater part of the earth with the exception of the polar and sub-polar regions, although they attain their maximum development, both as regards bodily size and numbers, in the tropical and sub-tropical zones. Rivers and lakes, for example, are inhabited by various kinds of fresh-water tortoises, belonging to several distinct groups, among which the soft-tortoises, or 77zonychide, are ex- clusively aquatic, although they spend much of their time basking on sand-banks. Many snakes, too, such as the common British grass-snake (7vopzdonotus natrix), the North American water-mocassin (Axcéstrodon piscivorus), and the huge anaconda (Lumectes murinus) of South America, are in a greater or less degree aquatic; while several of the iguanas, and especially the basilisk (Baszl¢scus americanus) likewise spend much of their time in the water, as do some of the Old World monitor lizards. A large number of reptiles have found safety from enemies and an abundant food-supply by taking to an arboyeal life. Among such are the chameleons, many geckos, a large number of iguanas, and many Old World lizards. Among the latter special attention may be directed to the so-called flying- dragons (Draco) of the Malay countries, on account of the membranous expansions by means of which they are enabled to take long flying leaps from bough to bough. Many snakes, such as the whip-snakes and the various kinds of tree-snakes, are wholly arboreal; while others, such as the pythons and boas, spend much of their time on trees. Certain snakes are also stated to take short journeys in the air, the natives of Borneo attributing the power of flight to Chrosopelea ornata, C. chrysochlora, and Dendrophis pictus; and it appears that these reptiles really can descend from a height in a manner analogous to that practised by flying-squirrels. In the ventral scales of these snakes there is a suture on each side, and by muscular contraction these scales can be drawn inwards, so that the whole lower surface becomes concave, and the body may be compared to a split bamboo. HAUNTS AND HABITATS 35 Experiment has shown that the snake when made to fall from a height, descends with the body rigid, and that the line of fall is at an angle from the point of departure to the ground. It seems probable that the concave lower surface buoys the rep- tile up in its fall, since the fall of a split bamboo through the air is perceptibly slower than that of an undivided rod of equal weight. Other reptiles have taken to a more or less completely bur- rowing and subterranean life, frequently with loss of the limbs, and sometimes more or less abortion of the eyes ; thus coming under the designation of what many writers are pleased to call degraded animals, which is, of course, merely another term for specialisation and adaptation ina particular direction. Among such burrowing and worm-like creatures are several families of snakes—notably the blind-snakes, or 7'yph/opzde—and the am- phisbenas and slow-worms among lizards. Many other rep- tiles, such as the spiny-tailed lizards (Uromast¢zx) and the tuatera (Sphenodon) are burrowers, but without “ degradation” ; while among the skink lizards (Sczzczd@) almost every gradation from a fully limbed to a limbless burrower may be observed. Of the purely terrestrial forms other than burrowers, limit- ations of space preclude anything more than the very briefest mention in regard to haunts. It may be observed, however, that, unlike the smaller mammals, a very large number of forms —and notably the lizards of the genera Agama and Lacerta— are diurnal in their habits, and trust to escape from enemies by the lightning speed with which they retreat into the crevices of the rocks they frequent. Desert forms, on the other hand, of which there are many, probably trust largely to resemblance to their surroundings as a means of escaping detection by foes. A large number of lizards living among grass likewise owe their safety to a similar protective resemblance, as is more fully noticed in the sequel. Up to a recent date no reptiles were known to have adapted themselves to an existence in caverns; but about 1898 it was discovered that numbers of a species of Coluber frequent caves in the Selangor district of the Malay Peninsula, where they feed on bats by which the cave is tenanted. These snakes grow to eight or nine feet, and are paler-coloured than ordinary, while their colouring presents a remarkable resemblance to that of the walls of the cave; at first sight, suggesting that the reptiles are 36 REPTILES ancient cavern-dwellers. Such protective resemblance would be of no use when in the darkness of the inner parts of the cave, but many of these snakes are in the habit of taking up their quarters on ledges near the entrance, from which they seize the bats passing in and out. The cave-walls consist of yellowish crystalline limestone traversed by blackish veins running ina more or less nearly vertical direction. With these rocks the colours and markings of the snake harmonise in a remarkable manner, a blackish line along each side of the reptile’s tail simulating the veins in the rock. Young specimens have not been found within the cave, and, if it really be that the species breeds only in the open, the resemblance in colouring must be due to fading in each individual. This snake is known as Coluber tenturus. It has also been discovered that an allied member of the same genus—C. moellendorffi—takes up its quarters in a cavern in Tonkin. Although the cave is of lime- stone, the colouring of the snake is quite different from that of the Malay species, the upper parts being grey with dark spots and the tail ringed with red and black. Hence, it has been argued, the resemblance of the Malay species to its surroundings is accidental. As occupying a somewhat intermediate position between the subject of haunts, or “station,” and geographical distribu- tion, reference may be made to the abundance of reptilian life in certain so-called “oceanic” islands where mammals are al- together unknown. The best instance of this is afforded by the islands of the Galapagos group, off the west coast of South America almost on the equator, which are tenanted by a number of giant land-tortoises of the genus 7es¢wdo, as well as by two large species of iguana, each representing a genus by itself (Amblyrhynchus and Conolophus). Giant land-tortoises also inhabited the Mascarene and the Seychelle islands till they were more or less completely exterminated by man. It has been held that the Galapagos are truly oceanic islands, that is to say, they have been raised from the ocean without ever having formed part of a continent. But Dr. J. Baur, who visited the islands in 1890, wrote as follows :— “Tn this case there must have been a time when not a single organism existed on the islands. Only by accidental [LE TLL PEL ANV'IS VUdVa' TV HLOAOS AO UNSC AFd OdALSAL) ASTOLAOL-ANWT LNVID V HAUNTS* AND HABITATS 37 introduction from some other part of the earth could the islands be populated; but on such a supposition we are unable to explain the harmonious distribution, we cannot explain why every, or nearly every, island has its race or species. If some animals could be carried hundreds of miles to the islands, why are they not carried from one island to the other? But, besides this, how could we make plain the presence of such peculiar forms as the gigantic land-tortoises, for instance? According to the elevation theory, we can only think of an accidental im- portation of these tortoises by some current, because they are unable toswim. After the islands had been elevated out of the sea, it happened once, by a peculiar accident, that a land-tortoise was carried over. Alone it could not propagate.! This was only possible after a similar accident imported another specimen of the same species, of the opposite sex, to the same island. Or we could imagine that at the same time animals of both sexes were thus accidentally introduced. By this we could at least explain the population of a single island. But how did all the other islands become populated? To explain this we should have to invoke a thousand accidents. “The most simple solution is given by the theory of sub- sidence. All the islands were formerly connected with each other, forming a single large island; subsidence kept on, and the single island was divided into several islands. Every island developed, in the course of long periods, its peculiar races, because the conditions on these different islands were not absolutely identical.” In other words, in place of being “oceanic,” the Galapagos group is “continental,” and was once joined to the American mainland. And there is little doubt that a similar explanation must hold good with the other “ tortoise-islands”. In Tertiary times giant tortoises of the genus Zeszudo inhabited all the great continents ; the earliest known form being 7. ammon of the Lower Eocene of Egypt, which may have been the parent stock of all. It is from these continental tortoises that the island forms of to-day are derived. It may be mentioned that the great antiquity of giant tortoises of the genus Testudo affords an excellent example of the early date, as compared with mammals, of the dispersal ' Unless it happened to be a female with eggs (R.L.). 38 REPTILES of reptiles, to which attention is more particularly directed in the paragraphs immediately following. The subject of the distribution of reptilian life on the globe at the present day is so extensive and complex, and involves the consideration of such a number of problems, that to accord it anything like adequate treatment would demand the greater part of the space accorded in the present work to reptiles. Obviously, therefore, only the vaguest sketch can be attempted. In the first place, it must be noted that, owing to the much ereater antiquity of reptiles as a whole than mammals or birds, their dispersal, or ‘‘ radiation,” has taken place at a much earlier date than that of either of the latter groups. This, then, would be sufficient to demonstrate that the zoological “realms and regions’ into which the surface of the globe has been parcelled out on the evidence of the present and past distribution of mammals and birds would not hold good for reptiles. This, however, is by no means all, for a study of the distribution of the existing orders of reptiles shows that there is little in common between them, and that each (doubtless owing to the different dates at which the groups have come into existence) has a “dispersal,” or ‘‘radiation,” of its own. Hence, not only cannot the distribution of reptiles be brought into line with that of mammals, but we cannot even map out a series of zoo- logical regions for reptiles as a whole. Certain facts in regard to the distribution of reptiles present problems which it is difficult to explain, in the present state of knowledge. Taking, for example, the northern hemisphere, of which the greater part is included in the Holarctic region of geographical zoology, with a Palearctic, or Old World, and a Nearctic, or New World, subdivision, we find an extensive community of mammalian types, due to the fact that in post- glacial times there existed a circumpolar mammal fauna, the members of which passed by way of which is now Bering Strait from one hemisphere to the other. On the other hand, no such circumpolar reptilian fauna existed in comparatively modern times, for the reason that reptiles cannot withstand the rigours of Arctic cold, and consequently could not pass from one hemi- sphere to the other across Bering Strait, in the same manner as the elk and the wapiti reached Alaska from Asia. The diffi- culty in the case of the reptiles is exemplified by the alligators, HAUNTS AND HABITATS 39 of which one species inhabits Eastern China and the other Eastern North America. This distribution seems at first sight to preclude the Bering Strait route, but the migration may have been earlier than that of the mammals, when the temperature was higher, At any rate, it is difficult to invoke the aid of land connections for modern types of reptiles which are not recog- nised in the case of mammals. A remarkable feature in the distribution of reptiles is the community between the faunas of Madagascar and South America, as exemplified by certain snakes, iguanas, and fresh- water tortoises. In the case of the tortoises the genus Podo- cnemzs is common to those two widely sundered areas, and the same community of generic type holds good for certain snakes. In the case of the tortoises it is possible to explain this re- markable distribution by supposing that different members of the genera in question travelled down the continents bordering the two sides of the Atlantic, as extinct members of the group are found fossil in the Eocene Tertiary rocks of the northern hemisphere. On the other hand, certain extinct tortoises (Miolania), of which no representatives are known in the northern hemisphere, occur respectively in Queensland and Patagonia. Their distribution it seems possible to explain only by a connection of the southern continents and islands in a southern zone; and this being so, it is difficult to deny that the Malagasy and South American species of Podocnemis have not travelled by the same route ;—a route which may perhaps best explain the distribution of the snakes and iguanas, despite the fact that representatives of the latter are met with in a fossil state in the European Tertiaries. It should be added that certain genera of tortoises, such as Pelomedusa and Ster- notherus, are peculiar to Madagascar and Africa, thus bringing, with the help of the aforesaid Podocnemis, the fauna of the latter continent into connection with that of South America, and so with Australia. Such are a few of the puzzling problems presented by the distribution of reptiles; but to dilate further on this subject is impracticable, and I accordingly pass on to glance at some of the leading features in the distribution of the more important groups of reptiles. As regards the Chelonia, of which the northern range in 40 REPTILES the western hemisphere is limited by latitude 50° and in the eastern by latitude 56°, the most remarkable feature is the limitation of the pleurodiran group (in which the neck is retracted by a horizontal flexure) to the southern hemisphere, where they occur in all the three continents, and in Australasia to the exclusion of the cryptodiran group (in which the neck is retracted by an S-like flexure in a vertical plane). Of the two chief families of Pleurodira, the Pelomedustd@ are common to Africa, Madagascar, and South America, while the Chelyide occur in South America and Australasia (exclusive of New Zealand), although there is no generic type common to the two areas. As regards the Trionychoidea, or soft river-tor- toises, the group is confined to Africa, Tropical Asia, and Eastern North America, to the exclusion of South America and Australasia; the distribution in America and Asia being thus comparable to that of alligators, although more extensive in the latter area. Of the terrestrial and fluviatile families of the Cryptodira, the snappers, Chelydride, are American, ranging as far south as Ecuador, but are represented in the Miocene of Europe. The Dermatemydide and Cztnosternide are North and Central American; the /Platysternide are confined to Eastern Asia, and the Yestudinide, especially as represented by the typical land-tortoises of the genus Zes¢udo, are cosmopolitan, with the exception of being unknown in the Australasian area. The Crocodilia are found in the warmer regions of all the great continents, inclusive of Northern Australia, as well as in the larger tropical islands. Crocodiles (Cvocodilus) in the New World are confined to Central America and the northern ex- tremity of South America, but elsewhere their distribution is coextensive with that of the order. The distribution of true alligators (A/égator) has been already referred to; but caimans or South American alligators (Cazmam) are natives of Central and South America. On the other hand, the long- snouted gharials (Garvialis and Tomzstoma) are restricted to India and the Malay countries. The Rhynchocephalia are confined at the present day to New Zealand, whose only other reptiles are geckos and skinks. Chameleons are an essentially African and Malagasy group, with outlying forms in the south of India and Arabia and in HAUNTS AND HABITATS 41 Ceylon, and also known (although perhaps introduced) in the south of Spain. Madagascar appears to be the headquarters of the group. The geckos (Geckonide, etc.) are practically cosmopolitan, exclusive of the colder regions of the globe. Another cosmopolitan family is that of the skinks (Sexc:d@) whose headquarters are Australasia, with representatives in New Zealand! The stellions (Agamzd@), monitors (Varintde@) and typical lizards (Lacert¢d@) are confined to the eastern hemisphere ; the latter having, however, a much more restricted range than the other two, not entering the Australasian region, and being likewise unknown in Madagascar. The latter island and New Zealand have no stellions or monitors, which are otherwise distributed over the greater part of the warmer zones of the eastern hemisphere. On the other hand, the iguanas (/gwanzd@) form an essenti- ally American group, which attains its maximum development in the tropical districts ; but it has three outlying generic types, two of which occur in Madagascar, and the third in the Fiji and Friendly Islands. Extinct forms, as already mentioned, are known from the European Tertiaries. The amphisbznas (Amphisbenide) have a wider range, occurring in America, the West Indies, Africa (but not Madagascar), and the Mediter- ranean countries; a very puzzling distribution, which is in no wise rendered more easy of explanation by the suggestion that these burrowing lizards are related to the tropical American family of tejus (Zezd@). Finally, the slow-worms (Axguzd@) are distributed over Europe, Northern Africa, Northern India, and the warmer parts of America: while their less specialised relatives the girdle-tailed lizards (Zonuride) take their place in Africa south of the Sahara and Madagascar. Apart from the tuatera of New Zealand, Australasia is seen to be inhabited only by geckos, skinks, stellions, and monitors ; and thus exhibits none of that marked relationship to South ‘In the volume on reptiles in the Cambridge Natural History it is stated (p. 5c0) that there are no skinks in New Zealand. There are, however, for ex- ample, two species of Lygosoma (see Cat. Rept. Brit. Mus. iii., pp. 271 and 272). 2In the work cited in the last note it is suggested that these eastern iguanas are not Iguanide at all, but have attained a resemblance to that family by “con- vergence”. If resort be had to such an explanation in all cases of difficulty, it would tend to show that our present schemes of classification are practically valueless, 42 REPTILES America displayed by its tortoises, or indeed to Madagascar. This latter island, on the other hand, lacks stellions, monitors, typical lizards, slow-worms, and amphisbznas, while it possesses, in addition to the cosmopolitan skinks and geckos (of which latter there are some very peculiar types) chamzleons and the African families Zonuride and Gerrhosauride. Lizards have the most northerly range of all reptiles, ex- tending in British Columbia to about latitude 56° and in the Old World almost to the Arctic circle. In the southern hemi- sphere they extend to the extremity of the American con- tinent. Snakes, on the other hand, stop considerably short of the above limits, both in the north and in the south. The sub- order is practically cosmopolitan, New Zealand being the only large island (other than those of the polar regions) in which it is wanting. Nor is this all, for dangerously poisonous species are to be met with throughout the distributional area of the group with the exception of Madagascar,! which enjoys complete freedom from such noxious creatures. A great contrast in this respect is presented by the Oriental region of geographical zoology (India, the Malay countries, and southern China), which is the home of members of the three most important groups of poisonous serpents, and has consequently the highest death-rate from snake-bite of any part of the world. The four families of burrowing snakes, namely 7yphlopide, Glauconiide, Ilystide, and Uropeltide, appear to indicate an ancient type, since some of them retain traces of the hind limbs, and it is therefore not surprising to find that, as a whole, they have a wide geographical range, although the last of the four is restricted to Southern India and Ceylon. The boas and pythons (Pythonzde), again, which likewise appear to be a com- paratively ancient group, have also a wide range in space, being almost cosmopolitan. Of the two subfamilies into which the group is divided, the pythons (Pythonine) are almost exclu- sively Old World types, large species of the typical genus occur- ring in Africa, India, the Malay countries, etc., and smaller forms in Australia. The one exception to the Old World distribution of this subfamily is the occurrence of Loxocemus bicolor (the 1Treland has, of course, the same immunity, but it can scarcely be recog- nised as a large island. HAUNTS AND HABITATS 43 sole representative of its genus) in Southern Mexico. On the other hand the boas (Bozve@) are mainly characteristic of tropical America, the home of the gigantic anaconda. There are, in- deed, certain Old World generic types (E7yx, Auygrus, and Casarea), but the remarkable feature in the distribution of the family is the occurrence of representatives of the American genera Loa and Corallus in Madagascar ;—a peculiarity paral- leled by the distribution of Podocnemzs among the tortoises. As regards the distribution of the family Colubrid@, which includes the majority of snakes, a brief notice must suffice, as many of the groups have a cosmopolitan range. Considerable interest attaches, to the distribution of the poisonous sub- family Elapine (which includes the cobras of India and Africa). This group ranges over the tropical and sub-tropical regions of both hemispheres (exclusive of Madagascar), but is particularly characteristic of Australia, where other snakes are represented only by a few pythons and blind snakes (7yphlopzde) and a small number of Co/uwbrine, or typical snakes. Nearly allied to the E/apine are the sea-snakes (/H/ydrophiine), which range from the Persian Gulf to Central America. Noteworthy, too, is the distribution of the Aszblycephalide, a small family of snakes allied to the Colubrid@, of which some forms are found in tropical America and the rest in the Oriental region. The vipers (Vzperzd@) have a nearly cosmopolitan distribu- tion, although absent from Madagascar and Australia; but whereas true vipers (subfamily Vzpercn@) form an exclusively Old World group, pit-vipers (Cvotaling) are represented in tropical Asia as well as in both North and South America. In this subfamily the rattlesnakes (Cro¢a/us) form a characteris- tic and widely distributed American group, although they have near relatives in the Old World, which are however unprovided with the distinctive rattle. As is evident from the statement already made as to their absence from the high north, reptiles are extremely sensitive to cold; and in temperate climates the whole of the species hiber- nate for a longer or shorter period in the cold season. With the exception of those adapted to a life in the desert, reptiles, and especially the sub-aquatic kinds, are intolerant of intense dry heat, and certain kinds “ zstivate,” or become torpid, during such periods, WA REPTILES A few words may be devoted to some of the features of desert reptiles—a subject which has been studied by Dr. Boettger, in the steppes of Transcaspia. In that district the winter, although comparatively short, is of great severity, while the summer is remarkable for its intense heat. The advent of spring clothes the ground with a carpet of lilies, tulips, and other flowers ; but their season is short, the heat of summer and the autumnal sandstorms reducing the country to a desert. The scattered vegetation has mostly narrow, or even needle-like leaves ; and the sand-heaps round the root of each plant form harbours for the lizards and snakes, the limited number of species of which abound in individuals. Among the most characteristic desert types are various skinks, such as Chalczdes and Teratosaurus, certain geckos which have forsaken the normal habitats of the group for this kind of existence, species of Phrynocephalus among the Agamzde, four members of the family Lacertide, and Varanus griseus representing the monitors. Neither are snakes wanting, the harmless kinds being represented by about half a score of species, among which the sand-burrowing Eryx jaculus is a well-known type ; while the venomous sorts include the asp or cobra, and the sand-viper (Echzs arenicola). Most of these reptiles are sandy brown in colour, marked with dark spots or stripes; and many of them exhibit other adaptations. The lizards and snakes are, for instance, markedly elongated, while the latter have an unusually large number of horny shields on the lower surface of the body. Some have the muzzle specially shaped for digging in the burning sand; and in others the scales are so arranged as to retain the sand when heaped up on the back. In most of the snakes the nostrils are protected by valves, or much reduced in size; but in the burrowing species they are situated on the top instead of in the front of the snout. The ears are also protected in a similar manner. Most remarkable of all these adaptations is the presence of a “window” in the lower eyelid of the skinks of the genera Maéduza and Adlepha- rus; the two eyelids being fused together in the latter. A similar arrangement obtains in certain typical lizards. None of these desert reptiles astivates. Crocodiles and alligators may be cited as examples of reptiles that aestivate when their haunts are desiccated ; burying HAUNTS AND HABITATS 45 themselves at such times deep down in the mud, where they remain till the return of moisture. Writing of the South American alligators, or caimans, the late Mr. H. W. Bates ob- serves that: “Like the turtles (Podocnemzs), the alligator has its annual migrations, for it retreats to the interior pools and flooded forests in the wet season and descends to the main river in the dry season. During the months of high water, therefore, scarcely a single individual is to be seen in the main river. In the middle part of the Lower Amazons, where many of the lakes with their communicating channels dry up in the fine months, the alligator buries itself in the mud and becomes dormant, sleeping till the rainy season returns. On the Upper Amazons, where the dry season is never excessive, it has not this habit, but is lively all the year round.” In the preceding paragraphs it has been mentioned that caimans and the great Amazonian tortoises make seasonal migrations. Similar migrations are undertaken by the European pond-tortoise (EHimys orbicularis) when its haunts are dried up. Whether under such circumstances it will estivate, I have no information. Another instance of seasonal migration among reptiles is afforded by the giant tortoises of the Galapagos Islands, which in the dry season make journeys into the interior in search of water. It does not appear, however, that any reptiles migrate south in winter in search of a warmer tempera- ture: in fact their rate of progress would prevent such marches. On the other hand, as already stated, all reptiles inhabiting cold climates hibernate. The pond-tortoise, for example, buries itself in mud in autumn, and does not reappear till the spring is well advanced ; while some of the American terrapins burrow in the banks for their winter slumber. As is well-known, the South European land-tortoises kept as pets in England retire for the winter months; but in their native countries the period of torpor must be less prolonged. Here it may be mentioned that no reptile can stand being frozen; so that if they do not burrow deep enough, such hibernating tortoises are likely to perish in severe winters. British snakes and lizards undergo a long hibernation, retir- ing in autumn, and not reappearing till spring. In the case of the grass-snake it is not unusual for several individuals to occupy 46 REPTILES the same hole ; and occasionally a number of individuals have been found coiled up into a solid mass. The most extraordin- ary instance of such a collection of these snakes (which were not hibernating) was recorded a few years ago in the month of September near Llanelly, South Wales, where they took pos- session of a house. The reptiles crawled over the floors, infested the cupboards, curled themselves together on the furniture, while some individuals climbed the stairs and luxuriated in the comforts of the bedrooms. The human occupants of the house had done their best to rid themselves of these unwelcome visitors, and had waged a war of extermination against them. The snakes continued to come, however, although no fewer than twenty-two were slaughtered in one day. The eggs from which the twenty-two individuals were hatched were probably de- posited by the parent behind the oven, or in a hole in the back wall. On taking down a portion of the latter wall forty bunches, each containing thirty eggs, were discovered, all on the point of hatching. There were thus some twelve hundred snakes in an area of a few square feet. Similar large congregations of rattlesnakes are well-known in certain parts of North America about the commencement of the hibernating season. In some districts it is reported that the snakes used to collect in hundreds, or even thousands, in the den, to which they travelled from distances of thirty or forty miles. Whether all reptiles that hibernate fatten themselves preparatory for their fast, does not appear to be ascertained. It is stated, however, that chamzleons certainly do so; and that in North Africa at any rate they retire for the winter under ground, although how long they remain there is unknown. It is likewise not ascertained whether any of the tropical species zestivate. CHARTER: IV FOOD AND GROWTH Food. Mode of killing prey. Snake-eating snakes; egg-eating snakes. Fascination. Rate of growth. Age. Vitality. Regeneration. Chronic disease. Sloughing of skin. Peculiar associations. HE food of reptiles is very various; and while in some groups the nature of the diet is more or less similar in all the species, in other cases there is a remarkable diversity in this respect among closely allied forms. Crocodiles are typi- cally carnivorous reptiles, tearing the carcases of the animals on which they subsist with their powerful and sharp-pointed teeth before devouring them. From the nature of their teeth we may also infer that the typical dinosaurs, such as MWegalo- saurus, as well as the theriodonts, belodonts (Phyzosaurus), ich- thyosaurs, and plesiosaurs, were likewise carnivorous ; the food of the last two groups probably consisting of the contemporary mail-clad fishes. Such of the pterodactyles as were furnished with teeth were likewise in all probability to a great extent, if not altogether, fish-eaters. And if this be so, we can scarcely refuse to believe that their toothless brethren, in which the jaws were probably sheathed with horn, subsisted on similar food. At first sight, indeed,‘it may seem strange that this marked differ- ence in the armature of the jaws does not imply a correspond- ing difference in the food; but the case of the chelonians, in which a toothless horny beak is correlated in some instances with a carnivorous, and in others with a herbivorous diet, seems to negative this idea. For example, while the common green turtle (Chelone mydas) is herbivorous, the closely allied hawk- bill (C. zmbricata) is strictly carnivorous, feeding upon fishes and molluscs. Again while the land-tortoises (Zestudo) are purely vegetable feeders, the common pond-tortoise (Emys orbz- cularts) and some of its near relatives are carnivorous, subsist- ing on fishes, molluscs, worms, insects, etc. On the other hand, 47 48 REPTILES the large batagurs (Batagur, Cachuga, etc.) of the Indian rivers are as purely herbivorous. Indeed the only group of chelonians which is constant in the matter of diet appears to be the soft river-tortoises (77onychide), which are wholly carnivorous. Snakes of all kinds subsist on an animal diet, which may, however, consist of the entire bodies of land-animals, swallowed whole, of eggs, of fishes, or in the case of the burrowing species, of worms and other subterranean creatures. The pythons and boa-constrictors crush the bodies of the animals they have seized for food within their lithe coils until the former are reduced to the condition of a sausage, when, after being lubricated with saliva, they are swallowed whole. Exag- gerated stories are undoubtedly current as to the size of the animals a five-and-twenty-foot python is capable of gorging ; but we have no definite means of gauging the capacity of these voracious reptiles in this respect. The fibrous (instead of bony) union between the two branches of the lower jaw in all snakes is a special adaptation for the swallowing of such huge boluses. After a gorge of this nature a python, like other large snakes, requires a long period of quiescence, which is passed ina kind of semi-torpor, before it is ready for another meal. The following particulars in regard to the amount of food consumed during one year by snakes in captivity were published in the report of the Trivandrum Museum, Travancore, for 1903; the dates when the reptiles changed their skins being likewise given :— A Malay python 21 ft. long, ate 100 fowls, two hare-walla- bies, two bandicoots, one kangaroo, and onedog. It shed its skin on 30th September, 11th December, 1902, and 18th February, 12th April, and 9th June, 1903. An Indian python measuring 154 ft. in length, ate fifty-four fowls, two bandicoots, two dogs, two guinea-pigs, one heron, and two hare-wallabies. Its skin was shed on 22nd August, 9th October, 17th December, 1902, and 6th February, 1st April, and 2nd June, 1903. Another ex- ample of Python molurus, 84 ft. long, ate during the year nine- teen fowls, four bandicoots, and one dove; shedding its skin on 14th September, 26th December, 1902, and on 19th February, 8th April, and 19th June, 1903. Yet another, 9 ft. long, ate fifteen fowls, one bandicoot, and one hare-wallaby. This snake FOOD AND GROWTH 40 shed its skin on 20th August, 15th October, 1902, and on 13th January, 16th March, 1903. A king-cobra, length 84 ft., ate during the year forty-four rat-snakes, shedding its skin on 7th October, 25th December, 1902, 19th February, 29th April, and 17th July, 1903. A cobra (Vata tripudians), 54 ft. long, ate fifty-five rats and fifty frogs. It shed its skin on roth Novem- ber, 1902, and 19th February, 8th April, and 28th July, 1903. A Russell's viper consumed fifty rats in the year, shedding its skin on 24th September, 26th December, 1902, and 17th April, 1903. Finally, a rock-snake ate during the year sixty-seven frogs. It shed its skin on 13th February, 20th April, 13th June, and 12th August, 1903. As already stated, the king-cobra of South-Eastern Asia, which reaches a length of 15 ft., is in the habit of preying on non-venomous serpents of other species. A specimen of this snake measuring a little over 11 ft. has been seen in Burma carrying another member of its own species in its jaws, while on a second occasion a king-cobra was observed in the act of eating an ordinary cobra, and in a third case a cobra had been swallowed. Another king-cobra has been known to devour a banded krait ; while from the stomach of yet another specimen was taken the still more venomous Russell’s viper. At first sight these instances suggest that the devourer must be immune to the venom of the devoured; but such conclusions are not justified by the present state of information with regard to the action of serpent-poison. In captivity a python has been known to devour one of its own kind, and likewise the blanket with which it was supplied as a protection against cold ; although both these instances were probably due to the depraved habits so often developed among animals in captivity. But the snake-eating, or king, cobra, takes its name from its habit of preying on other snakes, especially the rat-snake, which also derives its name from the nature of its diet. Many snakes apparently devour the eggs of birds and other reptiles when opportunity occurs; and when the snake is large and the egg small, there need not necessarily be any special difficulty in the feat. The difficulty comes when the snake is small and the egg large. A special adaptation enables, however, the egg-eating snake (Dasyfelfis scabra) to accom- 4 £0 REPTILES plish this feat without undue difficulty. This reptile, a native of South and Tropical Africa, and less than a yard in length, is able to swallow a hen’s egg, while specimens a foot in length will “get outside” a pigeon’s egg. The means by which this is accomplished are simple and ingenious. The lower spines of some of the vertebra are so lengthened as to pierce the upper side of the gullet, on the surface of which they appear as small teeth-like knobs ; and when an egg is swallowed it is crushed or sawn through by these projections. After the egg, with some difficulty, is swallowed (the skin of the snake being dis- Fic. 3.—African Egg-eating Snake and its mode of swallowing food. tended almost to bursting), it gradually glides further and further down till it comes to the projections in the gullet, when the swollen part of the snake’s neck suddenly resumes its normal size; and, after an interval, the broken shell is dis- gorged. A similar structural peculiarity characterises an Indian snake belonging to a different group, and known as Elachistodon westermanni ; from which it has been suggested that this species also displays egg-swallowing propensities. If the suggestion be well-founded, we shall have an instance of the independent development in two groups of an adaptation to the same end. Monitor lizards (Vavanus) are likewise consumers of eggs, FOOD AND GROWTH 51 but since these reptiles are large, they have no difficulty in regard to swallowing such delicacies, which are taken into the mouth, when, with the head held well up, the shell is cracked and the contents allowed to flow down the throat. A Bengal monitor (V. bengalenszs) kept in the gardens of the Trivandrum Museum devoured in the course of a twelvemonth sixty rats, 10 lb. of beef, six eggs, and four guinea-pigs. Skinks form a large por- tion of the food of monitors in some districts. All chameleons are insectivorous, and the same is the case with a number of true lizards, such as the stellions (Agama) and the typical lizards (Lacerta). Many lizards, however, such as the stump-tailed skink (Zvachysaurus), subsist on a mixed diet; but, judging from the nature of their teeth, the large Australian skinks of the genus 77/zguwa are probably herbivor- ous. The American iguanas exhibit that diversity of diet in different groups already mentioned as a curious feature among reptiles. For instance, whereas the anolis iguanas of the genus Anolis are insectivorous, the true iguanas (/gwana), the basilisks (Basiliscus),and the Galapagos iguanas (Conolophus and A mbly- rhynchus) are herbivorous. The last-mentioned, or sea-iguana, is peculiar among the group in feeding on sea-weeds. A superficial resemblance in the teeth of the dinosaur of the Wealden to those of the iguanas suggested to Dr. Gideon Mantell, its describer, the name of /ewanodon for the former ; and it would appear that this resemblance is correlated with the nature of the food in the two groups, for the iguanodons, as proved by the manner in which their teeth became worn down, were certainly herbivorous. Another group of dinosaurs, with teeth of a somewhat similar type, as exemplified by the European Pelorosaurus and Hoplosaurus and the American Lron- tosaurus (Apatosaurus) and Dzplodocus, are likewise generally considered to have been herbivorous. In 1g1o Mr. J. Versluys suggested, however, that Dzplodocus and its kin subsisted on fishes. In support of this it is pointed out that the capacity of the body-cavity is not large enough to have contained a suff- cient supply of vegetable food. Moreover, it is argued that the length and mobility of the neck, coupled with its muscular power (as demonstrated by the prominences on the bones), and the small size of the head, apparently indicate rapid and definite 52 REPTILES movements in water. Then, again, the nature of the dentition and the form of the mouth-cavity appear adapted for a fish-diet. And it is accordingly suggested that these huge reptiles cap- tured fishes near the borders of lakes and rivers, which they swallowed whole without mastication. The number of repre- sentatives of the Chelonia which are vegetable-feeders has been already referred to, but it may be mentioned that the giant land- tortoises of the Galapagos, together with the land-iguana of the same islands, subsist on large cactuses, which form the chief vegetation of those islands. Lastly, the New Zealand tuatera (Sphenodon), the sole ex- isting representative of the Rhynchocephalia, subsists upon ani- mal food, although the nature of this seems to vary according to individual taste. Some specimens, for instance, consume insects and worms, and those which frequent the shore not improbably eat crustaceans. This suggests that the food of the extinct Triassic pavement-toothed tuatera (/7yperodapedon) may also have consisted of crustaceans and perhaps of molluscs. The bean-toothed reptiles (Placodontia) almost certainly sub- sisted on a diet of the latter type. Brief reference to certain peculiar modes in which some reptiles capture or kill their prey must suffice. An ingenious method of capturing flies and other insects is employed by chameleons, in which the tongue is developed into an elastic, trumpet-like organ which can be shot out to a long dis- tance in front of the mouth, and is furnished at its tip with a glutinous secretion for securing the prey. Crocodiles and _alli- gators, having seized their victims in their cruel jaws, hold them under water until they are drowned; and it would seem that these reptiles have developed the peculiar respiratory mechan- ism by means of which they are enabled to breathe while their mouths are under water for this purpose. Pythons kill their victims by encircling them in their coils and gradually crushing the life out of them; and vipers and other venomous snakes kill their prey by injecting poison into their tissues. In this connection it is interesting to notice that the poison of different groups of snakes is designed to destroy with the greatest rapidity the particular kinds of animal on which they severally prey. The venom of the sea-snakes, for example, acts much more powerfully on fishes than on land animals, while that of cobras FOOD AND GROWTH 53 and vipers exerts its deadly effect with the greatest rapidity on mammalsand birds. Here it may be noted that certain animals, such as the ichneumon and the hedgehog, appear to be immune to snake-poison. There is an ancient belief that snakes possess the power of “fascinating,” or, in other words, inducing a kind of paralysis in the animals upon which they are about to prey. For many years this idea has, however, been discredited by naturalists, and it may be said to have received its deathblow as the result of observations conducted in the menagerie of the Zoological Society of London, of which an account will be found in the Society’s Proceedings for 1908. According to these observations, it appears that in the majority of such supposed cases of fascination there is not the remotest pretext for believing in the existence of any such power. Many animals, however, are of an inquisitive disposi- tion; and in the case of the smaller mammals and birds, this is associated with the power of attention. If a movement be sudden or noisy, they start off at once; but if it be slow, silent, and stealthy, they remain motionless, although intensely watchful. If a snake be prompt in seizing that moment of watchfulness, it may secure its prey, but a human hand slowly advanced has just as much power of fascination. These observations likewise demonstrate that, except in the case of one particular group, animals display no special fear of snakes; the majority of species, such as frogs, rats, mice, guinea-pigs, rabbits, ruminants, and birds, being absolutely in- different to the proximity of a serpent—venomous or other- wise ; and even when the latter approaches them, avoid it as they would a stick when thrust in their direction. On the other hand, apes and monkeys—but not lemurs—display a marked instinctive dread and recognition of snakes. This is displayed whether the reptiles are venomous or harmless. When the recorders of these observations approached the cages in the monkey-house with a group of writhing snakes, “the monkeys at once fell back shrieking, whilst the lemurs crowded to the front of the cage, displaying the greatest interest and not the smallest perturbation when a snake was brought so close to them that its tongue almost touched their faces. We got the impression that had the lemurs been given the ‘4 REPTILES opportunity, they would at once have seized and tried to devour the snake. The South American monkeys showed fear in ir- regular and sometimes slightly marked form. Spider-monkeys were quite as excited and alarmed as any Old World monkey. Some of the larger Cedzd@ did not retreat, but uncovered their canines, and looked as if they were ready to show fight. The Old World monkeys recognised the snakes instantly and bolted panic-stricken, chattering loudly and retreating to their boxes or as high up as possible in the larger cages.” The writers conclude that in all probability “human beings have inherited this specific fear of snakes from their anthropoid ancestors, and that our inclination to attribute a similar fear of snakes to other animals is due not merely to erroneous observa- tion but to an ‘anthropoidomorphic’ prepossession ”. A remarkable provision for the purpose of attracting prey within reach occurs in the mata-mata terrapin (Chelys fimbriata) of Brazil and the Guianas. In this reptile the head and neck are fringed with warty appendages, floating in the water like some vegetable growth, whilst the rough, bossed carapace re- sembles a stone,—an appearance which evidently is of as great use to this creature in escaping the observation of its enemies as in alluring to it unsuspicious animals on which it feeds. It would seem that in this chelonian we have a double adaptation ; —one for attracting prey, and the other to harmonise with the surroundings. As mentioned in the first chapter, no reptile undergoes a metamorphosis, or transformation, comparable to that of am- phibians, so that the life-history of all the members of the group is comparatively simple, consisting in the main of a gradual and regular increase in size from birth to maturity. It is true, that the young of some reptiles are more brightly coloured than their parents, while the newly-born offspring of crocodiles are furnished with an “egg-tooth”. | These, however, are but trifling points of difference; and, in many cases at any rate, no one would have much difficulty in declaring the specific identity of a young reptile with its parent. A young gharial, for example, of half a dozen inches in length, is a miniature of its parents, with the exception that if it be a male, it lacks the protuberance at the extremity of the muzzle characteristic of adult males. PEATE IV THE MATA-MATA TERRAPIN (CHELYS F/ITBRIATA) NILE SOFT-TORTOISE (7RIONYVX 7RIUNGUIS) FOOD AND GROWTH ar The uniform growth distinctive of reptiles may apparently cease only with the extinction of life; and for this reason it seems unwise to discredit the stories of unusually gigantic crocodiles, pythons, and anacondas which are from time to time reported, although in some cases these are doubtless exaggerated. An idea is prevalent that the growth of reptiles is exceed- ingly slow, and this is doubtless true in many cases. For in- stance, the young of the scheltopusik, or glass-snake, are stated to be many years in coming to maturity. It is also practically certain that the giant land-tortoises of the Mascarene and Gala- pagos Islands take an immense time before reaching their full size. Again, the American painted terrapin is known, from the result of observation, to be very slow in its growth. A specimen, for instance, in which in its second year the shell measured 26°5 millimetres in length, took twenty-five years be- fore it attained a full average shell-length, that is to say 121 millimetres; and it is inferred that an example in which the shell measured 163 millimetres was much older. That the rate of growth of the first specimen was normal, is demonstrated by the circumstances that for many years it progressed parz passu with that of a number of other examples kept under observa- tion. These showed that during the first half-dozen years the rate of increase is so regular that specimens can be arranged in series corresponding to their age from their similarity in size. Up to this period the age of specimens is indicated by the number of lines of growth on the horny shields covering the shell; but after the seventh year these shields tend to become smooth so that the annual lines of growth become more or less obliterated, when the possibility of reckoning the age by this means be- comes proportionately difficult. On the other hand, statements as to the slowness of growth of the American alligator (A//@gator mitsstssippiensis) are not supported by observations on specimens in the New York Zoological Park. It was stated, for instance, by an American writer that: “Alligators grow very slowly. At fifteen years of age they are only two feet long. A twelve-footer may be reasonably supposed to be seventy-five years of age.” The observations at the New York Zoological Park showed that young alligators when first hatched measured eight inches 56 REPTILES in length; while when a year old their length was eighteen inches, showing an increase of ten inches in a twelvemonth. A year later (in August) their average length was twenty-three inches; but in the following March their average length was three feet nine inches. At the time when these last measure- ments were taken the alligators were only two and a half years old, during which time they had increased thirty-seven inches in length. Probably the rate of growth would not continue at this rapid pace, but even so an alligator of twelve feet in length need not be more than a dozen years old. It might be urged that the growth of specimens in captiv- ity is abnormally rapid ; but even if this were so, the rate ina wild state would be in excess of that given in the passage quoted above. The observer who recorded the dimensions of the New York specimens is, however, of opinion that the growth in the wild state is at least as rapid as in the case of captive specimens ' :— “From observations made in the South Carolina bayous by the writer, it would seem that the growth of wild alligators must be fully as rapid if not more so, than that of the specimens reared in captivity. The females construct their nests near shallows teeming with fish, and in an atmosphere of heat and humidity. The young reptiles probably grow more rapidly when wild than when confined. Of course hibernation must be considered in the case of the wild reptile. During this period growth must be very slow, or cease altogether. Yet the writer has always noted that reptiles in captivity, no matter how elaborate may be the facilities for their care, or the voracity evinced by the reptiles themselves, never grow so rapidly as those in a wild state. Repeatedly has this been observed by comparing the young of wild and captive-bred snakes, the ages of which are easily distinguished.” As to the duration of life in reptiles, information is imper- fect, although the most satisfactory dates are afforded among tortoises, some of which attain an age which may be counted by centuries instead of years. As we have seen, the painted terrapin is known to live more than five-and-twenty years. Again Gilbert White’s tortoise “Timothy,” which belonged to the species Zestudo zbera and whose shell is exhibited in the 1Seventh Annual Report, New York Zoological Society, p. 150. FOOD AND GROWTH 67 reptile gallery at the British Museum, is known to have lived fifty-four years at the date of its decease in 1794 ; but how much has to be added to this is uncertain, since there is no clue as to its age when it was brought to England. Some idea as regards the age attained by giant land-tortoises is afforded by certain specimens from the Seychelles. In the year 1766 five tortoises belonging to the species Testudo sumezrec were taken from their island by the Chevalier Marion de Fresne and carried to Mauritius, where two were living a few years ago, The most celebrated of the pair is the one at the Artillery Barracks, Port Louis, of which the shell measures about forty inches in length in a straight line. Since the dimensions of the shell are reported to have been practically as large so long ago as the year I810, it is certain that this tortoise must have been very old at the time of its arrival in Port Louis; and something over a century would probably be a moderate estimate of its age at that date. Accordingly, it would seem that the reptile cannot be much less than 250 years, and may be much more. Another aged specimen was the Colombo tortoise. Ac- cording to tradition, this patriarch, which died in 1894, was found in Colombo when Ceylon was taken over by the British in 1796; having been imported from one of the “tortoise- islands”.1 At that time it was doubtless an unusually large and old specimen, or it would not have been kept, and we may ac- cordingly allow it a minimum age of a couple of centuries. In a third case, that of the Egmont Island tortoise, the evidence as to the duration of its captivity is less satisfactory. It is re- ported to have lived on Egmont Island fora century and a half, but since the Chagos group (to which that island pertains) was not colonised from Mauritius till the early part of the nineteenth century, there is some doubt with regard to the statement. Nevertheless, it is certain that this monster must be of prodigious age. This tortoise was in the habit of burying itself and remain- ing dormant for half the year. Although nothing definite appears to be known with regard to other reptiles, there seems to be little doubt that crocodiles attain a great age; and many years ago when Professor A. Leith-Adams visited the celebrated “magar-pit,” or crocodile-pond, near Karachi, he was told that a large specimen of the Indian magar (Crocodilus palustris) was 1See Pearson, Sfolia Zeylanica, vol. xxvi., p. 108, 1910. 58 REPTILES supposed to be about two hundred years old, but this can only be taken as a general statement to the effect that it was of great age. As might be expected from their cold blood and low organ- isation, reptiles display great vitality and power of recuperation from injury, although, as already mentioned, they will not stand being frozen. The most striking instances of this vitality are recorded in the case of the turtles, in which the heart will con- tinue to beat for several hours after it has been removed from the body, while the flesh may be cut piece-meal from the body till little more than the skeleton remains before life becomes extinct. According to Sir J. Emerson Tennent, it was the custom in his time in Ceylon to cut pieces from the flesh of living turtles and sell them to customers as required. Again, as is mentioned elsewhere, it was the practice of the natives of the Galapagos Islands to cut a slit near the tail of the giant tortoises so as to reveal the interior of the body, and thus permit of ascertaining whether the reptile was sufficiently fat to be worth killing. If its condition was not found to be satisfactory in this respect, it was let loose, when it recovered without difficulty from the operation. Closely connected with their vitality is the power possessed by many or most reptiles of regenerating lost parts. The most familiar and striking example of this power is afforded by the ease with which many lizards grow a new tail, or part of a tail, after having discarded a large portion of this appendage as a protective measure. Since this phenomenon is described in a later chapter, it need not be alluded to further on this occasion ; but it may be added that the power of discarding the tail and growing a new one is also exhibited by the New Zealand tuatera. As regards chelonians, Sir J. E. Tennent states that in Ceylon when the horny plates are stripped off the hawksbill turtle (Chelone imbricata), after roasting the reptile over a fire, they are regenerated after the creature has been returned to the sea. It is affirmed in proof of this that no turtles are caught in a mutilated condition, but this, if the original state- ment be true, may be due to the circumstance that their sufferings are ended by death. On the other hand, Dr. Charles Hose states that turtles are caught in Borneo showing signs of having lost their horny plates, which have been replaced by thin and imperfectly grown ones of no commercial value. FOOD AND GROWTH 59 That tortoises and turtles can regenerate their horny plates to some extent, provided the deep-seated, or malpighian, layer of the underlying skin be not destroyed, is, however, amply demonstrated. On this subject Dr. H. Gadow writes as fol- lows in the volume on reptiles in the Cambridge Natural Hrs- tory :— “If part of the horny covering is badly bruised, torn off, or rubbed through, or if part of the shell is crushed, the underly- ing portion of the horny plates becomes necrotic, and the horny covering also dies so far as its malpighian layer is destroyed. Soon, however, the uninjured malpighian cells, around the margin of the wound, multiply, grow into and beneath the in- jured portion of the bone, and form a new horny layer, casting off the necrotic portion. After several months the deficiency is patched up; new bone has grown in the deeper remaining strata of the cutis, and the outside is covered by a continuous horny layer, without, however, reproducing the original con- centric moulding of the shields. In badly crushed shells some- times almost one-third of the whole shell is thus cast off and mended within one or two years.” Should terrapins, as not infrequently happens, lose their tails or limbs by a bite, the missing part is never reproduced ; the stump being merely sealed over. One species of terrapin, namely Clemmys leprosa of Europe, is subject to a peculiar disease when living in foul waters, although perfectly healthy when frequenting clear streams ; it is from this diseased condition that it derives its name of leprosa, or “‘leprous”. When living in slimy pools an alga makes its way into the cracks and crevices in the horny plates of the shell, and thus penetrates into the underlying malpighian layer of the skin, or even into the bone itself, which becomes gangrenous in large patches, while the whole shell has a dis- tinctly leprous appearance. The only other known instance of an alga flourishing in the superficial tissues of a vertebrate animal occurs in the sloths (Bradypodid@) among mammals; where it grows in the grooves traversing the coarse hairs, to which it communicates a distinct greenish tinge, believed to aid in rendering the animal inconspicuous in its leafy haunts. All vertebrates apparently change from time to time their epidermis and its appendages. In some cases, as in the human 60 REPTILES species, the change is gradual and imperceptible. In other instances the shift is more marked, as in the changing of the hair in many mammals in spring and autumn, and the “ moult” of birds. In certain reptiles—namely snakes and lizards—(as well as in amphibians) alone is the entire epidermis, inclusive even of that covering the cornea of the eye, sloughed at once, either, as in the case of snakes, entire, or in several pieces. In describing the change of skin in the common slow-worm, Dr. G. Leighton, in his “ British Lizards” writes as follows :— ‘The slough is exceedingly delicate and therefore torn with great ease, and can only be shed entire if the slow-worm is able to glide through soft material during the process of chang- ing. Any sharp projecting point rubbing against the sides of the creature will inevitably tear the slough before it is com- pletely removed. The process starts at the jaws, and the lizard gradually crawls out of the slough, leaving it turned in- side-out as a rule, though the terminal portion of the tail-slough may slip off unreversed. After sloughing, the slow-worm, like other reptiles, is more lively and feeds readily. The length of time between successive sloughings varies. Sloughing always happens after the slow-worm comes out of its winter-quarters, and is generally repeated at intervals of six weeks or so during the months in the year when active life is in progress. The colouring of the reptile is more brilliant after sloughing than at other times. The slough is never eaten by lizards [and snakes], as is the habit of some amphibians.” The number of times in the year in which certain Indian snakes shed their skins while in captivity is recorded on p. 48; and in these it would seem that the change is less frequent than in the slow-worm. In tropical countries it is no uncommon thing to find the complete slough of a python or other large snake from 6 to Io feet in length. In terrapins, and probably in other chelonians as well, a more incomplete, although at the same time a well-marked, change of the superficial layer of the skin takes place. In the case of the freshwater painted terrapin (Clemmys picta), for ex- ample, a thin transparent layer, like a film of mica, peels off all the heavy plates of the shell in autumn or at midsummer, when the brilliant colours underlying the newly formed plates ap- pear much more vivid than ordinary. FOOD AND GROWTH 61 Although no reptiles live in organised communities after the fashion of hamsters among rodent mammals and bees and ants among insects, there are a few instances of certain species living in companionship with other animals. In North America, for example, rattle-snakes frequently take up their re- sidence in the burrows of the “ prairie-dogs,” or prairie-marmots (Cynomys), which are often also tenanted by the curious little owls of the genus Sfeotzto. It used to be supposed that these strangely associated animals constituted a veritable “happy family,” but it is now ascertained that the snakes resort to the marmot-warrens for the sake of feeding on the young marmots. When there are no young ones, it would, however, seem that the snakes live in a state of harmony—or at all events of “armed neutrality ”—with the marmots. Another instance of a similar kind of association is afforded by the New Zealand tuatera (Sphenodon), whose burrows are sociably shared by petrels of various kinds. The petrel is stated to usually occupy the left, and the tuatera the right side of the inner chamber of the burrow, which, by the way, is inva- riably excavated by the reptile. The late Sir J. von Haast observes that while very tolerant of the bird and its young, the tuatera does not allow another reptile of its own kind to live in the same hole, which it is ready to defend by lying in such a manner that the head is placed where the passage widens out to form the chamber. Another curious association between a bird and reptile occurs in Egypt. Herodotus tells the story that a bird he called Zvochilus enters the open mouth of a basking crocodile to pick the fragments of food left between the teeth of the reptile. Fora long time this story was rejected as a “ traveller’s tale,” but modern observations appear to show that it is per- fectly true. At one time the bird in question was supposed to be the black-backed courser (Pluvialis egyptiacus), but it really appears to be the spur-winged lap-wing, or ziczac (/oflopterus armatus). An observer writing in 1870 stated that he sawa bird which he believed to belong to the latter species deliber- ately enter the mouth of a basking crocodile two or three times; and that during one of these visits the reptile actually closed its mouth, opening it again after a time to let the bird out. CEVA ARV SEX AND REPRODUCTION Some sexual features. Cries. Scent and scent-glands. Milky and other secretions. Significance of brilliant colouring. Habits in breeding season. Reproduction and care of eggs and young. Maternalextinct in pythons. Vipers swallowing young. Fertility of snakes. Incubation. NFORMATION is still deficient with regard to the be- haviour of the two sexes of reptiles during the breeding season, although a few observations have been recorded which show apparently that in the case of some species at any rate much excitement, frequently accompanied by pugnacity, is displayed at this period. A few kinds develop specially brilliant colours, or an intensification of the normal colouring ; but since many reptiles, or, at all events, many snakes and lizards, are at all times more or less brilliantly coloured, and likewise show variations in the brightness of their hues according to whether they have recently changed their skins or not, it is difficult to decide to what extent any or all of the brilliant tints displayed by certain species during the breeding season are due to what is commonly called sexual selection. In a large number of instances male and female reptiles are very similar to one another in general characters; but in other cases there are marked secondary sexual differences between the two sexes. For example, many male lizards are more brightly coloured than the females, and often show special patches of colour which are wanting in the latter. Again, in the family of the /ewanide there is often a greater development of the characteristic dorsal crests and gular pouches, or “ dew- laps,” in the males. For example, the male basilisk (aszlzscus americanus) alone possesses the tall sail-like crest running down the back and tail; and it is only the male of the horned iguana (Metopoceros cornutus) which carries the three horn-like scales on the forehead characteristic of the genus and _ species. 62 SEX AND REPRODUCTION 63 Similarly, a dewlap is developed only in the males of the chamezleon-iguana (Axolis carolinensis). Among geckos the males are generally larger and distinguished by the presence of femoral pores. Among crocodiles the males of the Indian gharial (Garialis gangeticus) are distinguished by the presence of a large swelling at the extremity of the snout, enclosing the nostrils. When the nostrils are closed, this swelling can be blown out like a football: probably connected with this organ, which is doubtless of a sexual nature, is the presence on the base of the hinder part of the skull of a pair of large bony capsules, of the size of a hen’s egg, and connected with the respiratory passages. Perhaps, however, the most remarkable secondary sexual difference in reptiles occurs among the Chelonia in the terrapins and land-tortoises. In all these the male, which is generally much the larger of the two, has the centre of the plastron, or lower shell, more or less deeply hoilowed, whereas in the female it is flat or even slightly convex. The object of this hollow- ing of the plastron in the male is so obvious that it need not be particularised. Male tortoises have also longer tails than females. Closely connected with the sexual function are the cries uttered by many reptiles, more especially in the breeding-season. In the common 7estudo greca, for instance, the male, which then becomes unusually alert and active, makes a kind of piping noise when in pursuit of the female during the breeding season. Again, during the pairing-season the males of the giant land- tortoises of the Galpagos Islands utter a hoarse roar, or bellow, which can be heard at a distance of over a hundred yards. The female never uses her voice (even if she has one), and the male only at pairing-time; so that when the natives hear the sound, they know that the two are together. Certain terrapins, such as the pond-tortoise of Europe and the species of the Asiatic genus WVcorza, can give vent to a kind of whistle. It is, however, a comparatively recent discov- ery that the males of most members of the American family of terrapins typified by the so-called “stinkpot” (Czzosternum odoratum) are in the habit of producing musical notes after a fashion very similar to that obtaining among grasshoppers and crickets. These terrapins are furnished with two patches of 64 REPTMEES horny tubercles on the hind legs, which afford good specific characters. As in crickets, there is an active and a passive set of these tubercles, the lower patch being rubbed against the upper one, and thereby producing a musical note, as does the bow of a violin. The note is clear and distinct, audible at a considerable distance. As the tubercles are developed only in the males, they are probably used to produce musical notes solely during the breeding season, thereby informing the females of the proximity of members of the opposite sex. These musical instruments of the Cznosterntde are almost unique among vertebrates; the only other instances of the de- velopment of a somewhat analogous apparatus in that group occurring among the geckos of the genera 7eratosctncus and Ptenopus, which are enabled to produce musical notes by means of friction between the horny rings of their tails. This sound, which is like the chirping of grasshoppers, is, however, emitted by both sexes, and may be for the purpose of attract- ing those insects within reach. Very probably, the clicking cry of other geckos is, at least in part, of a sexual nature, and like- wise the chirp of lizards. Chameleons both hiss faintly and grunt ; but these sounds are probably uttered for the purpose of frightening enemies, as is certainly the case with the hissing of serpents and monitors. It is by no means easy, as in the case of cries, to determine in all instances whether scents (or their opposite) are for the purpose of attracting the individuals of a species or for repelling enemies. If, however, the sense of smell in reptiles is akin to that in ourselves, it may be presumed that agreeable scents are connected with the sexual function, while disagreeable ones are for the purpose of repelling enemies. Under the former category probably comes the musky odour exhaled by croco- diles from two pairs of glands, one of which is situated on the throat and the other near the vent. The scent, it has been suggested, leaves a musky aroma along the line of water traversed by the animal, by means of which another individual is enabled to ascertain its whereabouts. On the other hand, the exceeding ill-smelling odour exhaled by Clemmys leprosa and certain other terrapins and tortoises can scarcely be regarded in any other light than as a defence against enemies. In this instance the secretion, which is de- SEX AND REPRODUCTION 6s scribed as something like concentrated essence of fish, is the product of a pair of glands situated beneath the skin near the inguinal region, and opening on each side behind the bridge connecting the upper with the lower shell. Freshly caught specimens of this species, according to Dr. FI. Gadow, stink horribly when handled; but after they have been kept in con- finement for some time, they lose the habit of voiding the contents of the glands every time they are taken up, and thus become less objectionable pets. The North American ‘“stinkpot-terrapin” (Czxosternum odoratum) owes its ill-favoured name to the fetid secretion exuded by the inguinal glands, which can scarcely be regarded as intended otherwise than as a means of defence. Under the same category must doubtless come the habit displayed by certain lizards (such as the scheltopusik) and snakes (among them the common ring-snake) of ejecting when handled the ill-smelling contents of the cloaca. The milky fluid exuded by the American milk-snake and the blood squirted from the eyes of the so-called horned toad, to which fuller reference is made in the concluding chapter of this section of the present work, likewise come under the category of fluids discharged for defensive purposes, whether used in combat with rival males for the possession of some coveted female, or to repel the attacks of enemies of another kind, Many lizards inflate the body, the region of the mouth, or special laryngeal sacs, for the apparent purpose either of frightening enemies or as a means of sexual attraction, or perhaps for both together. Examples of this are displayed by the inflation of the body in Lacerta and Phrynosoma, in the expansion of the frills of Ch/amydosaurus, and the dilatation of the gular sacs of Metopoceros and other iguanas. Such effects might be enhanced, it is reasonable to suppose, by a swelling- out of the head and protrusion of the eyes. Such a function, according to Dr. H. L. Bruner in the American Journal of Anatomy, vol. vii. pp. 1-117, is, however, insufficient to explain the existence in the heads of both sexes of many lizards and snakes of an apparatus of muscles and vascular sinuses for producing excessive blood-pressure, and consequent swelling in this region. In lizards, at any rate, this mechanism 5 66 REPTILES is developed for the purpose of aiding in the shedding of the scales, and acts physiologically by accelerating lymph-move- ments, and thus promoting metabolism, and mechanically by stretching the skin over the soft parts. This being so, the probability is that the same factor holds good in the case of snakes and tortoises ; but in some instances the function may be modified for terrifying or sexual purposes, and it is probable that the ejection of blood from the eyes of the “horned toads ” (Phrynosoma) is a development of the same mechanism. From the foregoing it is clear that it is necessary to be guarded in framing any hypothesis as to the precise significance of brilliant coloration among reptiles. Though commonly as- sociated with sexual display, it does not seem always to be used as an accessory in this respect. A case in point is furnished by the painted terrapin (CZrys- emys picta). In the breeding season the male has been seen dodging the female, and making efforts to oppose her path. This end accomplished, the male closed up and immediately commenced to beat a lively tattoo with his long finger-nails upon her head and eyes, the movements being so rapid that nothing more than a blurred image was possible. So soon as possible, the female escaped these attentions, when the male set about repeating the performances, which were witnessed not once but many times before the pair were disturbed and made off. Here no use seems to have been made of the bright coloration but only one phase of the “display” may have been observed. Crocodiles find a sombre livery useful, since it affords con- cealment. Consequently the males present, except in the case of the gharial, no ornamental colours, nor any other marked secondary sexual character. Nevertheless, alligators, when en- deavouring to attract the females, splash, roar, and twist them- selves around on the water, with the head and tail raised, and the body inflated to its utmost extent, the effect of this being increased by the emission of a strong odour of musk from glands in the lower jaw. In many lizards the males display great pugnacity during the breeding season, and rivals never meet without a conflict. In Anolts carolinensis, for example, when two males meet they face one another, bob the head up and down two or three times SEX AND REPRODUCTION 67 expand a great throat-pouch they possess, lash their tails from side to side, and then, worked up to the requisite pitch of fury, rush at one another, rolling over and over, and holding firmly with the teeth. The conflict generally ends by one of the com- batants losing his tail, which is eaten by the victor. Dr. Gadow states that in the breeding season the males of a Malay lizard (Cadotes emma) “ are very pugnacious and change colour as they fight. At the time of courtship a curious per- formance is gone through by the male, the females remaining concealed in the foliage hard by. He chooses some convenient station such as a banana leaf, or the top of a fence, and advances slowly towards the female. His colour is then pale yellowish flesh-colour, with a conspicuous dark spot on each of the gular pouches, which are extended to their utmost. He stands up- right, raising the fore part of the body as high as possible, and nodding his head up and down. As he does so, the mouth is rapidly opened and shut but no sound is emitted. When he is driven away, caught or killed the dark spot disappears entirely from the neck.” That instances of this kind are by no means rare there can be no doubt, but records of actual observation are few. Those cited show the nature of the evidence so far collected, but on the whole there is less activity and intensity of feeling displayed by reptiles in their choice of mates than is the case with birds. Similarly, in the care displayed for their offspring reptiles are far behind birds. As already mentioned, reptiles are pro- perly oviparous, laying eggs from which the young are usually hatched by the heat of the sun or by that engendered by decay- ing vegetable matter, without any aid from either parent. In many species, however, such as the common viviparous lizard (Lacerta vivipara) the young burst the shells of the eggs immedi- ately after they are laid, or in some cases even before they are laid. Although in the latter instance the species may be said to be viviparous, the term ovo-viviparous is properly applicable to this mode of reproduction. At least one reptile, namely the Austra- lian stump-tailed lizard (7 vachysaurus rugosus) is, however, really entitled to be called viviparous, for not only are the young born free, but the hard calcareous shell characteristic of the eggs of reptiles in general is never developed. These differ- ences in the mode of reproduction are probably in all cases 68 REPTILES correlated with some special feature in the life-history of the species in which they occur, and are therefore mainly, if not entirely, adaptive, although in most instances the reason for such adaptation is not apparent. In the case of the sea-snakes (Hydrophiine), however, which are viviparous, the reason for the departure from the normal oviparous mode of reproduction is plain, as these reptiles, unlike turtles, never voluntarily come to land. The burrowing snakes of the families ///yszzde and Uropeltide, and possibly also the 7yphlopide, are likewise vivi- parous; and here again the reason for the specialisation is not very difficult to discover. On the other hand, it is not easy to see why the viper (like most of its tribe) should be viviparous, while the ring-snake is oviparous. Till a few years ago most of the viper tribe were believed to produce their young alive; that is to say, the eggs are hatched within the body of the female parent, so that, strictly speaking, these reptiles should be described as ovo-viviparous rather than viviparous. In the /ze/d of January I, I9I0, there appeared, however, a letter from Mr. C. Leigh in which it is stated that the Himalayan pit-viper (Lacheszs monticola) is ovi- parous. This viper, which ranges from the Eastern Himalaya to the Malay countries, is remarkable for the great thickness of its body as compared with the length; for while it usually attains a length of about 24 feet, its girth is sufficient for that of an average 5-foot snake. The ground-colour of the back is ashy-grey, upon which is a series of more or less regular dark blotches, with a sprinkling of black and traces of yellow. On the under-parts the general colour is likewise ashy, but with dark red mottlings. In some examples the dark markings on the upper surface are so pronounced that the reptile is scarcely distinguishable from the freshly-turned black soil of the tea- gardens, in which it is frequently found. It was while working ina garden at Kurseong, during the summer of 1909, that a coolie came across a cluster of snake’s eggs. On reaching out his hoe to investigate and pushing the grass aside, he discovered a viper of this species which made a snap at him, but was eventu- ally secured. The tropical American bushmaster (Lacheszs muta), and all the members of the African genus A/ractasprs » are also oviparous. An analogous instance is offered by the family ozde, in- SEX AND REPRODUCTION 69 which most of the boas produce living young, while pythons lay eggs, A still greater difficulty arises (as in the case of La- chests and Lacerta) when we find the two modes of reproduction occurring respectively in different species of the same genus. The North American banded water-snake (770fzdonotus fas- clatus) belongs, for example, to the same genus as_ the ege-laying ring-snake, and yet produces living offspring. In some instances among lizards all or most of the members of a family may be oviparous or viviparous as the case may be; but in other instances there is a strange mingling of the two types of reproduction. The iguanas, for example, appear to be, at least for the most part, oviparous, while the skinks are as markedly viviparous. On the other hand, the above-mentioned viviparous lizard stands out as a marked exception among an oOviparous group; while in the family Anguide we find that whereas the continental glass-snake, or scheltopusik (Ophzsaurus) lays eggs, the British slow-worm (Anguzs) gives birth to living offspring. All crocodiles, together with chelonians and the tuatera, are oviparous, Of the modes of reproduction of fossil reptiles we are, from the nature of the case, in most instances ignorant. By what may be regarded as a fortunate accident, we know, however, that the ichthyosaurs produced their young alive, after the manner of sea-snakes. The evidence for this ‘is afforded by certain skeletons (of females) from the German Lias, within the ribs of which are enclosed the remains of foetuses. This de- parture from the normal type of (oviparous) reproduction is evidently correlated, as in sea-snakes, with more or less com- pletely pelagic habits; and thus indicates, as might have been expected from the structure of their paddles, that the typical ichthyosaurs never came ashore, but led lives similar to those of whales; the pelagic habit being most developed in the specialised Ophthalmosaurus of the Oxford Clay. Whether plesiosaurs, pelagic crocodiles, and sea-serpents (mosasaurs) were likewise viviparous, or whether, like turtles, they came ashore to lay their eggs, cannot be positively stated. On the one hand, the less specialised type of paddle (as com- pared with the ichthyosaurs) might lead to the inference that these reptiles occasionally came to land. On the other hand, 70 REPTILES the snake-like form of the body in the mosasaurs, coupled with the weakness of the pelvis, and the extreme length of the neck in the plesiosaurs (which would appear to be exceedingly incon- venient to an animal on land), suggest a completely aquatic life for the members of those two groups. If this be a true infer ence, there can be little doubt that they were viviparous. One other adaptive modification in the ichthyosaurs may be mentioned. From their “coprolites” it is known that they were furnished with a spiral valve to the intestine; a feature paralleled by sharks, rays, and chimeras among fishes, Ichthyo- saurs also resembled sharks and many dolphins in possessing a dorsal and a caudal fin; the latter being, like that of sharks (and unlike that of dolphins), vertical. There are remarkable differences in the shape of the eggs of reptiles, some being spherical, while others are elliptical, with the two ends symmetrical ; in no case, however, do we find the typical “ egg-shape,” that is to say, one end rounded, and the other more or less pointed. Crocodiles and gharials lay ellip- tical eggs about the size of those of a goose; but the tuatera’s eggs are spherical. Pythons’ eggs are also spherical, while those of many other snakes are elliptical, and fastened together in bunches or strings. Many land-tortoises, again, lay spherical eggs, which in the case of the giant species are about the size of lawn-tennis balls; and the eggs of the marine turtles are simi- larly shaped. On the other hand, the fresh-water terrapins and batagurs lay elliptical eggs. The season for this diversity in shape is difficult to discover, seeing that it has no correlation with terrestrial or aquatic habits. As already mentioned, most reptiles leave their eggs to take care of themselves after they are laid. The sand-lizard (Lacerta agilis), for example, deposits its eggs, which usually range from five to eight in number, during July in a depression in sandy soil, where they are left to hatch by the aid of mois- ture and the sun’s heat. In cases, however, where sand is not to be found the eggs are deposited in leaves, vegetable mould, or rubbish. In the case of the green lizard (ZL. wvzrtdzs) the eggs, generally eight to ten in number, are developed for five weeks within the maternal oviducts, and are then laid in suit- able situations, where they remain four weeks more before they hatch. The ring-snake usually selects decaying vegetable SEX AND REPRODUCTION 71 matter, or, in gardens, manure-heaps, in which to deposit its clusters of eggs. Turtles lay their eggs in holes scraped out in the sand of tropical coasts well above high-water mark to the number of a hundred or more; the holes being carefully filled up with sand and the surface smoothed down. ‘The period of incubation, by the aid of the sun’s heat, is believed to be at least seven weeks. Turtles’ eggs have a parchment-like covering, unlike the hard calcareous shell of those of the giant land-tortoises, Crocodiles likewise generally lay their eggs in sand, and leave them to be hatched by the sun’s heat, although, in some cases at any rate, they are visited from time to time by the female parent, who assists the young ones in escaping from their prisons. In the case of the Nile crocodile (Crocodilus ni- loticus) in Madagascar, the eggs are laid in a pit in the sand from one and a half to two feet in depth ; the centre of the pit being rather higher than the margin, which is undermined, so that the eggs naturally roll into the shelter thus formed. The eggs are laid somewhat before daybreak; and after half the batch has been deposited, a layer of sand is spread for the reception of the moiety. When the laying is completed the pit is filled up with sand and the surface smoothed down. The position of the nest is, however, frequently betrayed by the mother taking up her position upon it at night. When the young are hatched (which occurs in about twelve weeks after the laying of the eggs), the nest is generally pulled to pieces, and the empty egg-shells left lying scattered about. This digging out of the nest is believed to be accomplished by the mother, who appears to be warned when the eggs are ready to be hatched by the piping cries uttered by the young crocodiles while still in the sheli. Apparently the eggs are dug out by the parent several days before they actually hatch. The process of breaking the shell is accomplished by the young crocodile with the aid of the “egg-tooth”; a two-cusped excrescence on the summit of the upper jaw at the extremity of the muzzle; this acts in drill-fashion, and does not disappear till some weeks after birth. Many other reptiles have a somewhat similar weapon, In tropical America the female of the black caiman (Catman niger) covers up her eggs with a mass of bushes, and 72 REPTILES a keeps guard close by till they are ready for hatching, when she probably assists the young ones to escape. In pythons the maternal instincts are still more developed, and the female incubates, or rather coils herself round (for no augmentation of temperature occurs), her pile of eggs until they hatch. The object of this action is probably in the main, if not entirely, to protect the eggs. A female python in the London Zoological Gardens brooded her eggs for some weeks in this manner in the summer of 1881. Unfortunately, the observations made in the foregoing in- stance were imperfect, practically the whole information ob- tained being that the snake coils herself round the mass of eggs to protect them and that no appreciable amount of extra warmth is developed during incubation. A second instance of a python laying and brooding eggs in captivity has been more recently recorded in Ceylon. In the autumn of 1904 a python (probably Python molurus) was received at Colombo from the Malay Archipelago. Although the measurements and weight could not be ascertained, it is estimated that the snake was about 28 ft. in length and weighed 2501b. Soon after its arrival from Singapore—that is to say, on 28th October—the snake laid about 100 eggs, which almost filled the box in which it was kept. On the following morning, by skilfully coiling her body around them, she had collected the mass of soft-shelled eggs into a heap in such a manner that they were almost completely covered without being pressed by the weight of her body. In order apparently to maintain a constant temperature the python from time to time partially uncoiled herself, so that the heap of eggs became visible. From 28th October, 1904, till 14th Janu- ary, 1905, she refused all nourishment, although tempting dainties were offered. On the latter date the python left the mass of eggs exposed, and it was feared that incubation had been unsuccessful. Closer inspection revealed, however, a young python half emerged from one of the eggs, into which it retired at evening. On the next day, 15th January, six young pythons were hatched, some of which died in a few hours, while the others soon became active, making darts at a cloth held near them. Eventually forty-five young snakes were counted, of which thirty-six were alive on 20th January. When first hatched the young measured 2 ft. to 2} ft. in length. The period SEX AND REPRODUCTION 73 of incubation of the Malay python is thus two and a half months, Another instance of a python (this time an African species) incubating in captivity occurred in the Jardin des Plantes, Paris, in the year 1841. A fourth case took place in the menagerie at the Tower of London in the year 1828. In this instance the snake had been more than two years in the col- lection when she produced fourteen or fifteen eggs, none of which, however, were hatched, although the mother evinced great anxiety for their preservation, coiling herself round them in the form of a cone, of which her head formed the summit, and guarding them from injury with solicitude. The eggs appear to have been shown to visitors by the keeper, for it is stated that they were only visible when she was occasionally roused, and raised her head, which formed the cover of the pile. Every time, however, she resumed her normal position as quickly as possible, allowing the spectator only a momen- tary glance at her treasures. If there be any truth in the story of its swallowing its young, the most remarkable instance of maternal care among reptiles occurs in the viper. On the face of it, this occurrence seems sufficiently improbable, and its possibility has been denied, Nevertheless, popular beliefs are often founded on fact. The question has been considered by Dr. G. Leighton ; and although he has been unable to cite any definite instance of the occur- rence of the phenomenon, he shows that some of the objections which have been urged against it are based on a misinterpreta- tion of anatomical facts, and demonstrates that there is nothing impossible in its taking place. As the gullet of an adder is perfectly capable of containing the body of a field-mouse, and as frogs are known to live for a considerable time after being swallowed by snakes, there is no reason why young adders should not be swallowed by their parent without being killed. The question remains, however, to be proved by positive evi- dence. “Of the possibility of the phenomenon,” writes the author, ‘‘we have not the slightest doubt, of the probability of it we have considerable doubt”. In the case of the python at Colombo about one hundred eggs were laid. In this great fertility pythons apparently are ahead of viviparous snakes, although exact information regarding the 74 REPTILES maximum number produced by different species of the latter is still required. In 1906 I received the skin of a female of the beautifully-coloured Gabun viper, or puff-adder (4772s gabonzca), together with several young ones taken from the body when the reptile was killed. These young ones measured about twelve inches in length, and the total number was said to be thirty. Thirty feet of young snakes is certainly a goodly family, and confirms current statements as to the abundance of these deadly reptiles in the forests of equatorial Africa, The number of young produced at a birth by adult females of the European viper is stated to be generally from twelve to fourteen although, occasionally reaching sixteen. Younger females, how- ever, produce only five or six at a time. The North American banded water-snake (7 vropidonotus fasciatus) seems to be an unusually prolific species, a female in the New York Zoological Gardens having on one occasion given birth to sixty-two young—a family so numerous that if literally overran the cage in a mass of writhing brilliantly coloured bodies, Usually, it seems, the eggs of reptiles are hatched within comparatively few weeks of being laid. A remarkable excep- tion is, however, afforded in the case of the New Zealand tuatera, which is in so many ways strange and aberrant. Although the eggs are laid during the summer (of the southern hemisphere), that is to say, from November till January or February, and contain fully developed embryos by the follow- ing August, they are not hatched till about thirteen months after being laid, thus coming into the world in summer. During the period of delayed development the embryos appear to undergo a kind of hibernation; the cavity of the nose becoming blocked by a growth of tissue which is absorbed as the time of hatching approaches. The eggs are not laid in the tuatera’s burrow but in a warm sandy spot, where they may receive the full advantage of the sun’s rays. CHAPTER? VI COLORATION AND ITS INTERPRETATION Colour in relation to environment. Stripes and spots in lizards. Colour- changes in relation to sex and age. Colours of young pit-vipers. Voluntary colour-changes. Green arboreal snakes. Desert-snakes. Sea-snakes. Bark- geckos. Lizards. Warning and protective colours. NLIKE the majority of mammals inhabiting the temper- ate and sub-arctic zones, reptiles do not exhibit a seasonal change of colour; and for the reason that they do not inhabit the sub-arctic regions, and that those of the colder parts of the temperate regions hibernate. It is noticeable, how- ever, that the young of some reptiles are more _ brilliantly coloured than the adults, while snakes display much more vivid hues immediately after changing their skins than at other times. Probably the colours of the great majority of reptiles are pro- tective, that is to say, they harmonise more or less completely with their environment, and thus render the creatures incon- spicuous. For instance, foliage-haunting reptiles such as chameleons and tree-snakes are coloured green; while the majority of reptiles inhabiting open ground are either mud- coloured or sand-coloured, as, for instance, crocodiles, stellion lizards, skinks, and desert-snakes. The brilliant coloration of many of the pythons, such as the Malay Python reticulatus, appears highly conspicuous in specimens in a museum, but it is probable that among the splashes of sunlight and shade of their native forests these bright-hued reptiles are inconspicuous, the tesselated pattern of the colour breaking up the hard outlines of the body. Again, as is shown later, sea-snakes display adaptive colouring ina marked degree. Before considering in detail these adaptations in general colouring, reference may be made to observations with regard to the occurrence of stripes and spots in certain lizards. Many lizards, especially some of the skinks, display a longi- 75 76 REPTILES tudinally striped type of coloration. Such stripes occur in the European wall-lizard (Lacerta muratlis); but they are by no means constant, and it appears that in certain cases a different colour-pattern may be assumed, according to age, and perhaps locality. Moreover, when such changes of pattern take place, they appear to do so in a definite order, which is generally as follows, First of all there are longitudinal stripes ; next these stripes break up into spots, which may again coalesce into cross- bars; while, as a final change, all markings may disappear, and leave the skin uniformly brown or sandy. Certain indi- viduals of this species undergo all these changes of colour and pattern as they pass from youth to old age; while others stop short at the second or even the first stage, and a few skip the first or second stage and begin life in the third or fourth. As a rule, these changes are most pronounced in the males; the females commencing the change at a later period of life than their partners, when they may either undergo the whole trans- formation or only pass through some of the stages. In all cases these changes of pattern and colour commence near the tail and advance forwards in a kind of wave-like manner, so that the shoulders of an individual of the species may be in one of the intermediate stages while the tail has attained the perman- ent sandy condition, Very similar changes also take place in certain tropical American lizards of the genera Cnemidophorus and Amerva, belonging to the family 7Zezzd¢@, which is related to the iguanas. In these lizards a pattern of longitudinal white stripes not infrequently breaks up into spots; this dissolution commencing at the tail, and gradually advancing towards the head. On the other hand, pale spots may by confluence produce longitudinal light stripes. Another mode by which such stripes are produced is the concentration of dark pigment along the borders of a pre-existing dark band, accompanied by the growth of new colourless tissue in the middle of such band ; thus giving rise to one white stripe flanked by a pair of dark ones. A modification of this process occurs in the case of the white dorsal streak met with in some lizards, which as it broadens tends to develop dark pigment along the middle line, and thus eventually becomes split into a pair of white stripes divided by a single dark one. Confluence of pale spots in a COLOERATION AND ITS INTERPRETATION 77 transverse direction, accompanied by the deposition of dark pigment along the forming borders, leads to a pattern of trans- verse bars, corresponding to the third stage in the colour-evo- lution of the wall-lizard; while by the disappearance of both spots and bars the final uniform coloration is reached. In these American lizards, as in the wall-lizard, there ap- pears to be no doubt that the longitudinally striped condition was the original one; the spotted type being a secondary con- dition, from which have been evolved the cross-barred and finally the uniformly coloured stage. And as these changes are more or less paralleled both among birds and mammals, it may be definitely accepted that stripes and spots are more ancient than a uniform colour. As regards the reason for these changes, it has been ob- served that if we start with the original and typical form, marked with six strongly pronounced white stripes, it will be found that in individuals frequenting sandy districts dotted with sparse tufts of grass there is a tendency to an increase in the number of stripes, of which there may be from eight to eleven. On the other hand, in neighbouring spots, where the vegetation is more abundant, the number of stripes is usually from seven to nine, and these show a tendency to break up into spots on the hind part of the back. As we pass into open tropical forest with much undergrowth the lateral stripes fade, while the others dissolve into spots which have a tendency to disappear on the loins in old individuals. Again, in races inhabiting open table-land with scattered spiny shrubs and hedges, the young form with six light stripes and pale spots on the dark bands tends to pass into a cross-barred type. Finally, in similar localities to the last, but with more mixed vegetation, all the lines become broken up into spots, in addition to those which existed in the intermediate dark intervals. These changes appear to be correlated with the varying distribution of light in the different stations these lizards re- spectively frequent. If this be so, the colour-changes are protective in their nature ; and it has been urged as a reason for the frequent elimination of the original white striping that the effect of such an arrangement is constantly interfered with by the movements of the animal. Further, it has been pointed out that it is far more natural for the lights and shadows to fall 78 REPOILES transversely, rather than longitudinally, on a cylindrical body. As to the question why any of these lizards retain their longi- tudinal striping, seeing that it is the natural course of evolution that they should be lost, it has been suggested this may be due to the circumstance that small-bodied reptiles living among sparse tufts of grass have their colour-pattern less interfered with than is the case with their larger relatives, or even that the incidence of sunlight enhances the effect of their linear marking. It appears indeed that these lizards cannot retain their stripes unless they live in situations where such a type of coloration tends to render them in harmony with their surroundings ; any more than they keep their pale spots amid desert sur- roundings, For it is a well-ascertained fact that when spots occur in desert-dwelling animals they are dark on a light ground, as is exemplified among reptiles by some of the sand- skinks (Chalcdes) and among mammals by leopards and servals; while among forest-dwellers light spots on a dark eround are the fashion, as in the fallow-deer and the Indian spotted deer. It may be taken, then, as probable that most of the uni- formly coloured or dark-spotted reptiles inhabiting desert districts, where their coloration is eminently protective, are derived from white-striped or white-spotted ancestors dwel- ling in situations where vegetation was more or less abundant. And if this be true, it follows that chamzleons and tree-snakes which have acquired a green livery to harmonise with the surrounding foliage have in all probability been evolved from striped or spotted ancestors by an analogous modification. Be this as it may, such green coloration in arboreal reptiles is evi- dently a specialised protective adaptation, as is likewise the colour-charge of chameleons and certain lizards. Exemplifying somewhat more fully the variations in colour and pattern which occur in many reptiles, it may be shown to what extent these are connected with sex and age in species other than those already mentioned. As an example of sexual colour-difference may be cited the terrestrial iguanas of the genus Sceloforus. In most of the numerous representatives of the group the males are larger and more brilliantly coloured than the females, from which they are further distinguished by the presence of bright blue blotches COLORATION AND ITS INTERPRETATION 709 on the under-parts. These blue blotches, and the brighter tints generally, being placed on the under surface of the body are more or less completely hidden when the reptiles are crouching on the ground, and only become visible when the head and body are elevated under the influence of excite- ment. At such times the blue or purple on the throat of the males is conspicuous, although that on the flanks is less notice- able. This type of colouring is therefore evidently a secondary sexual character. Generally the upper parts are some shade of brown, and adapted for protective resemblance when the lizards are crouching. Here it may be noticed that the bright blue spots on the flanks of the European eyed lizard (Lacerta ocellata) have the same situation as the lateral blue blotches of the males of Sce/oporus ; and if, as is probable, they are brighter in males than in females, they may be regarded as sexual colour- features. From their position, it may be inferred indeed that these blue blotches have little to do with concealment although, as already mentioned, the “vermiculated” black and yellow colour-pattern of the back of these lizards is protective. Still, as stated below, the whole colouring may at a short dis- tance blend into an inconspicuous blur. The American blue-tailed skink (Avmeces quinguelineatus) offers an example of pattern and colour-change dependent upon sex andage. In the young the tail is bright blue and the body blackish with five yellow stripes. As this lizard matures the tail fades to sombre grey, the body changes from black to brown, the stripes (in the males at any rate) entirely disappear, and the head, in the same sex, assumes a bright red hue. Formerly the adult was regarded as a distinct species. Again, in the aforesaid terrestrial iguana, Sceloporus szos- teromus, the adults show a broad dark lateral band running backwards from in front of the fore-limb along the flank, while in the young the large patch in front of the fore-leg is frequently disconnected from the lateral band. In 5S. d¢serzatus the throat, the middle of the under surface, and the lower side of the thigh in old males are often black, while in younger individuals they are greyish or bluish. In the case of one species of horned toad (Phrynosoma coronatum) all the markings are much more distinct in the young than in the adult ; the small spots on the raised portion of the scales of the head in the young gradually 80 REPTILES increasing in size until the whole crown becomes dark brown or blackish, crossed by irregular yellow lines marking the hind borders of the scales. In the allied P. platyrhinus the mark- ings on the hind part of the back are apt to become obsolete in the adult. Another type of variation has been recorded in the Mexican poisonous lizard (Heloderma suspectum), in which the dark rings on the tail and limbs of the young break up into irregular spots in the adult. The lizards of the genus Ger- rhonotus exhibit in a marked degree the more brilliant colouring of the young, and also show in that condition a distinct pattern of alternate light and dark bars which tends to fade out as maturity is attained. The age and sexual differences in colour and pattern in the genera Amezva and Cremidophorus have been already noted. Perhaps, however, the most remarkable colour-change of this type occurs in an African snake known as Grayia ornata. In the young of the phase described as Grayia ornata furcata the body is black with broad whitish or greyish transverse bars, which split on each side so as to form a series of reversed Ys; each Y being slightly speckled with black about the middle line. With advancing age the black ground changes to grey or brown; whereas the Ys show more and more black until they have only white margins, and even these eventually disap- pear. Consequently, the fully adult snake has black markings on a grey ground, in place of the light markings on a black ground characteristic of the young; in other words, the colour- pattern of the adult is precisely the reverse of that of the young. In regard to colour-variation in British lizards the following observations have been published by Dr. G. R. Leighton in his Life-History of British Lizards. “Very young specimens of Lacerta vivipara, the common lizard, are nearly black, while the adults are brown and spotted ; the under-parts red or orange in the males with black spots, while in the females these portions ars yellowish. The young of the sand-lizard, Z. agzl’s, are greyish-brown with spots, whitish underneath ; the adult male is typically green on the sides, the female more brown. In JZ. viridis, the green lizard, the yellow lateral stripes which are found in the young persist in some of the old females.” COLORATION AND ITS INTERPRETATION 281 The same writer refers to the much brighter colouring of young vipers, as compared with very old specimens. A much more marked change in colour occurs, however, in the Malay snake known as Wagler’s viper (Laches?’s zwag/er?), which in the young state is generally green while in the adult it tends to blackish. The protective nature of this colouring, at least in the young condition, is referred to in the sequel. This viper also exhibits great variation in colour apparently independent of either age or sex. Neither does it seem that these variations are altogether local, seeing that both the first and the third of the under-mentioned colour-types occur in Borneo, while the first and fifth are recorded from Celebes. These variations, according to the British Museum Ca/alogue of Snakes, are as follows :— “A. Green above, with white cross-lines edged behind with blue or purple, or with two dorsal series of small spots or cross- bars of the same colour ; a white line on each side of the head, passing through the eye, edged below with blue or purple; belly white or pale green, with or without black edges to the ventrals ; end of tail usually red or reddish brown. “ B®, Green above, with small black spots or cross-bands; a black streak on each side of the head, passing through the eye; yellow beneath, with or without black edges to the ventrals, with a series of small black spots on each side ; end of tail red, “C. Yellowish green above, the scales edged with dark bottle-green; dark cross-bands of the latter colour; some specimens dark bottle-green above, with scattered yellowish green dots; ventrals yellow, edged with dark green, with a series of round green spots on each side, or dark green with yellow spots ; end of tail dark green or blackish. “D, Green above, with the scales black-edged, with bright yellow black-edged cross-bands, or black with yellow cross- bands; head black, spotted with yellow; belly bright yellow or yellow and green; ventrals black-edged ; end of tail black. “#, Green above, with large brick-red, black-edged spots ; white beneath, with black spots and marblings powdered with brick-red ; end of tail red.” Such differences, whether they be individual, or character- istic of local races, demonstrate how easily new specific types can be evolved from existing forms. In the New York Zoological Park some remarkable colour- 6 82 REPTILES changes due to age have been observed in pit-vipers. These snakes are born with the tip of the tail, for the length of about an inch, of a brilliant sulphur-yellow. When food is in- troduced into the cage these young vipers communicate a writhing, twisting motion to their tails, causing the latter to resemble small worms, or maggots. Possibly nature has provided them with this dash of brilliant colour to attract small birds, lizards, or frogs within reach, as they lie coiled and difficult to discern from the surrounding vegetation. This feature has been observed in the copperhead snake (Azczstro- don contortrix), the water-moccasin (A. fzscevorus), and the fer-de-lance (Lachests lanceolata). After the first year, the yellow of the tail becomes very indistinct, and during the second year it disappears altogether. With some of the pit-vipers, the colours of the young are very brilliant, although they exhibit much the same pattern as the adult. |. Young moccasins show brilliant shades of red and yellow at birth. The adults display a dull pattern of varying shades of sombre olive; while old specimens exhibit no pattern at all, the body being dull green. Quite different from the pit-vipers are the young of some of the colubrine snakes. The young of the black snake (Lascanium constrictor) are pale grey, with blotches of brown or red along the back, and resemble the milk-snake (Ophzbolus doliatus tri- angulus). During their second year they become darker, and the pattern appears diffused. The third year shows hardly a trace of the spots as the black of the adult appears, although the sides indicate the marking of the young. The chicken-snake (Coluber guadrivittatus) is remarkable, as are most species of Coluber, in showing when young an en- tirely different pattern from the adult, both forms being strongly coloured, At the time of hatching the young chicken-snake is greyish, decorated with a regular series of oblong blackish saddles. As the reptile approaches maturity, the body-colours change to yellow, a dark stripe appears on each side of the saddle-like markings, and another on the side of the body. These stripes become very distinct before the saddles begin to fade. The latter change takes place usually during the third or fourth year, the mature form being uniform yellow or brown, traversed by four longitudinal stripes. ECOLORATION AND ITS INTERPRETATION 83 Apparently no reptiles assume a special coloration in the breeding-season comparable to the “nuptial plumage” of many cock birds, or to the brilliant tints of the male crested newt at the season in question. The fact that chameleons, which are for the most part uni- formly coloured, have the power of changing the hue of their skins according to the nature of their environment has long been familiar, but it is less well-known that certain lizards are endowed with the same faculty. The colour-change in chame- leons is due to the presence of pigment-bodies deep down in the skin, which can be kept in that situation, or can be brought near to the surface at the will of the reptile. When the pigment is deep-seated, the skin appears dirty white or yellowish ; when, on the other hand, it is allowed to come near the surface, the colour becomes dark or even blackish; while in the intermedi- ate condition a greenish hue results from the diffusion of the dark colour through the outer layers of the yellowish skin, and by the iridescent character of certain elements in the latter. The colour of the common chameleon is practically inde- scribable, so great and so frequent are its liabilities to change, although there is great individual variation in regard to the frequency, duration, and extent of the changes. The normal coloration may be described as greyish green, with a number of darker specks and two series of pale brown patches on the sides of body, and a single patch near the ear. At night the prevalent hue is creamy yellow, with irregular blotches of yellow. Under the influence of excitement, as when in the act of striking a fly, the normal diurnal tints are intensified, the pale brown patches deepening to a full maroon, while the green is dotted with spots of golden yellow. Passion seems to cause these golden spots to turn blackish green. Very noteworthy, as showing the economy with which natural phenomena are conducted, is the fact that in many instances it is only the ex- posed side of the body that turns green when the animal is moving or sitting among foliage in daylight; the concealed side being yellowish white. Direct sunlight leads to a general darkening of the colours; and it is remarkable that in some in- stances the normal green hue, with or without the dark blotches, is retained during the night. As chameleons form a subordinal group (the Rhiptoglossa) 84 REPTILES of the Squamata, it is evident that their adaptive green and changeable livery has been acquired independently of that of the chamzleon-iguanas (Axo/7s), which belong to the family leuanide. \n the case of the Florida chamezleon-iguana (Anolis carolinensis) the extreme ranges of variation in colour extend from dark brown to pea-green; the former hue (in captive specimens at any rate) being assumed in daylight and the latter at night. The brown condition is induced by the migration of pigment-granules from the centres to the terminal branches and processes of the “ melanophores” ; the green state, which is one of rest, being the result of the withdrawal of the granules to the centres of the larger bodies. In three fun- damental points this colour-change differs from that of the chameleons ; differences which are only natural to expect in phenomena of totally independent origin. When a brown anolis is placed in the dark, it invariably turns green in about five-and-twenty minutes, and when a green one is exposed to the light, it nearly always changes to brown in the short period of about four minutes. Marked differences in the length of time occupied in these changes have been recorded ; and the change from brown to green is often slower at the commencement of an experiment than after it has been continued for some time. This leads to the conclusion that the period of colour-change is shortened by exercise, or practice; and it has been inferred that a low temperature induces a migration of the pigment- bodies towards the surface of the skin, while a high one is con- ducive towards their withdrawal. This rule appears to be constant for all reptiles subject to a colour-change. On the other hand, light sometimes follows this rule, and sometimes acts in the opposite manner. For example, in certain monitors a strong light acts like a high temperature in causing a withdrawal of the pigment-bodies, while a dim light, or darkness, acts in the opposite way. On the other hand, in the African chamzleon and in the chamzleon- iguana strong light acts like low temperature, inducing the mi- gration of the pigment-bodies to the skin, while dim light, or darkness, causes their withdrawal. In other words, while in the first case light and heat act in unison, in the second case their results are antagonistic. These lizards are arboreal, hopping from bough to bough COLORATION AND ITS INTERPRETATION 85 like tree-frogs ; and their changes in colour are very rapid in a state of nature. Other tropical American lizards belonging to the family Teude, or tejus, and the genus A mezva, also possess the power of changing colour, although apparently to a more limited ex- tent ; and in this case also the power seems to have been inde- pendently acquired. A fourth instance of a similar colour- change occurs in the Indian and Malay variable-lizard (Ca/- otes versicolor); and since this species is a member of the family Agamzde, which has no near relationship with iguanas or tejus, the independent origin of the phenomenon may be considered as beyond doubt. These lizards, which in the Malay Peninsula are miscalled chameleons by Europeans, are arboreal and pugnacious in their habits. The normal colour of the skin of the body is brownish, but when the lizard is feeding or ex- cited this changes to pale yellow, while the head and neck be- come suffused with brilliant red. When the male is courting the female its hue becomes yellowish flesh-colour, with a con- spicuous dark patch (which completely disappears when he is disturbed) on each of the throat-pouches. As already mentioned, the coloration of a large number of snakes (including pythons and boas) appears to be of a pro- tective nature, but only a few instances of special adaptation of this kind can claim notice. Among these, reference may be made to the occurrence of a green colour, to harmonise with their leafy environment, in several groups of those reptiles, as there appears to be evidence that this has been acquired in- dependently in at least two of these groups. The American wood-snakes of the genus Herpetodryas, which belong to the typical subfamily of the great family of Colubrid@, are purely arboreal, and typically of a more or less uniformly olive-green tone of colour. The sipo, or Brazilian wood-snake (//. cavz- natus), for instance, is normally bright verditer or olive-green above, with a tinge of brown on the back, and greenish or bright yellow below. Curiously enough, however, probably owing to some difference in its habits—in the West Indies it becomes blackish or blackish brown above, with the under parts steel-grey. Nearly allied are the Old World and Australian tree-snakes of the genera Dendrophis, Dendrelaphis and Chlor- ophis, the latter of which is confined to Africa and takes its 86 REPTILES name from the green colour characteristic of the group in general. Seeing that these snakes are nearly allied to the American Herfetodryas, it would be rash to affirm, despite their wide geographical separation, that their green livery has been independently acquired. Whatever doubt may exist in this case, it is practically certain that the arboreal Indian whip-snakes (Dryophzs) and their American and Malagasy re- latives of the genus Phzlodrvas have developed their green colour apart from the members of the above-mentioned groups. For these snakes belong to a section of the Colubrzd@ in which the hind teeth in the upper jaw are grooved for carrying poison, whereas those of Herpetodryas, Dendrophis, etc., are solid ;—a difference implying a widely sundered ancestry. The long and slender bright green whip-snakes are almost impossible to detect when among the foliage of their native forests, and are even difficult to see when coiled among the branches of a small shrub in captivity. One species of viper (Lachesis waglert) of the Malay countries is also bright grass-green in the young state, although later on the scales develop black edges, and in the adult condition the whole colour becomes blackish. This viper has a prehensile tail and feeds largely on birds, and is thus to a great extent arboreal. That the green colour in the young is protective and has been developed independently from that of all the colubrine tree-snakes is manifest; but why the creature should turn black in old age is a puzzle. Possibly the adult may lie more on the larger boughs than the young. Many desert snakes present a remarkable protective resem- blance to their surroundings. In no case is this more pro- nounced than in the horned viper (Cerastes cornutus) of North-eastern Africa and Arabia, which is sandy coloured, and in the day-time is in the habit of lying buried in the sand with only the eyes, nostrils and the horn-like processes on the head exposed, when it is practically invisible. The European rhino- ceros-viper (Vzpera ammodytes) presents a similar intimate re- semblance to its desert surroundings. Although conspicuous when removed from its natural surroundings, the Cape puff- adder (vtes arietans), with its yellowish or orange-brown ground-colour marked by chevron-shaped dark barrings, is almost invisible when on sandy and stony ground. The last instance of protective coloration among snakes to PLATE V PROTECTIVE RESEMBLANCE IN REPTILES: BARK-GECKOS OF MADAGASCAR A.—THE LICHEN BARK-GECKO (UROPLATES FIMBRIATUS LICHEN) B.—THE COMMON BARK-GECKO (UROPLATES FIMBRIATUS) . COLORATION AND ITS INTERPRETATION 87 which space admits of allusion occurs in the sea-snakes, forming the subfamily Hydrophiine of the Colubride. As mentioned above, these snakes are noticeable for their com- pressed form, at least in the region of the tail. The majority are coloured almost exactly like mackerel; that is to say, the back is dark blue, from which descend bars of the same colour, separated by whitish interspaces of similar width; while the under surface is of the same silvery hue as these light inter- spaces. Obviously, as in the case of mackerel, the mottled blue of the upper parts harmonises with the rippled surface of _ the ocean and thus renders these snakes invisible from above ; while their light under-parts when projected against the light of the sky render them equally inconspicuous to enemies from below. In one large species the dark upper surface is orange- brown instead of blue; the object of this departure from the normal type of coloration being at present unknown. Tree-geckos in general, by their mottled hues of grey, white, and black, present a marked protective resemblance to the bark of forest trees ; but the resemblance is carried to its highest degree in one variety of a Malagasy species known as the lichen bark-gecko, Uvoplates fimbriatus licheniuin (see Pl]. V.). In this species, to which allusion will again be made, the upper surface is dotted with irregular whitish lichen- like markings upon a slate-coloured ground, so that the resemb- lance to lichen-coloured bark must be almost perfect when the creature is at rest with its flattened head, body, and tail closely pressed to the bough. Indeed the only part of the reptile that would appear in the least conspicuous would be its large and bloodshot eyes, and even these may perhaps assimilate in nature to the end of a recently broken twig. A few geckos, such as Naultinus elegans and Thelsuma madagascariensis, are coloured green to resemble foliage, the former being wholly of this tint, while the latter has the hind part of the back spotted with red, As already mentioned, a number of ground-lizards are coloured slaty grey to harmonise with the prevailing hues of the rocks or soil on which they habitually dwell ; but two in- stances may be cited where this protective coloration is speci- ally noticeable. These are the two lizards commonly known under the names of the horned toad (Pkrynosoma cornutum) 688 REPTILES and the moloch (Moloch horridus). In the former the resem- blance to dark-coloured soil is striking ; the skin being slaty erey with flecks of white and black. On the other hand, the moloch is coloured to harmonise with sandy soil containing brown pebbles or fragments of stone ; the ground-colour being buff, with a number of oval cholocate blotches. When half- buried in soil of the above description, the creature must be practically invisible; the illusion being enhanced by the horn-like excrescences dotted over the head, body, and tail. It is, however, by no means only such dull-coloured lizards that harmonise in hue with their surroundings. Few creatures are more brilliantly coloured than the eyed lizard (Lacerta ocellata) of Southern Europe, with its back of mingled flecks of gold and chocolate, the large bright blue “eyes” on the flanks and the brown head and tail. And yet there can be little doubt that these harlequin colours fade into a confused blue when this reptile is lurking among grass and other low herbage. Again, certain small snakes have their skins mottled with bright red and black ; a combination which although conspicuous in cap- tured specimens, harmonises with soil composed of red sand mingled with dark pebbles. A similar type of coloration occurs in that American reptile known as the poisonous lizard, or gila monster (Fleloderma suspectum), which is not uncommon in parts of Arizona and the neighbouring districts. If this reptile inhabit country with a soil of the above-mentioned description, its colouring must be of a protective nature, but such descriptions of its habitat as have come under my notice are silent on this point. On the other hand, the gila monster may present an example of “ warning colour” ; that is to say, its bright colour- ing may be conspicuous and serve to warn enemies (by inherited experience) of its dangerous and poisonous nature. In concluding this part of the subject, it may be pointed out that no reptiles display that peculiar protective arrange- ment of colouring so commonly developed in large mammals living in the open, which consists in having the under-parts light and the back dark, thus neutralising the effect of the shade cast by the body. And the explanation of this is not far to seek, for no existing reptiles have the body sufficiently elevated above the ground to make this type of coloration effective. It is not meant by this to assert that no reptiles COLORAHION:- AND ITS INTERPRETATION 89 have light under-parts—they are in fact strongly marked in crocodiles and many snakes; but in these cases the light- ness is due to the absence of any necessity for colour, these parts being concealed, and not to a special protective adap- tation. CHAPTER Vil ADAPTATIONS ADAPTATIONS TO THE GENERAL CONDITIONS OF THE ENVIRONMENT: Ter- restrial types. Arboreal types. Climbing types and effect on tail and feet. Running types. Bipedal types. Flying types. Swimming types. Sail-backed lizards. Limbless types. Burrowing types. Modification of the eye and ear. Dermal armour. IZARDS and crocodiles present what may be called the + ordinary or typical form of body among reptiles—a form so familiar to all that nothing in the way of de- scription is necessary. If, however, we examine a large series of reptiles in a museum (exclusive of those which have lost the limbs and assumed a snake-like form of body) we shall find that there are two distinct and well-marked modifications of this typical shape; the one pertaining to species which are terrestrial, and the other to such as are more or less completely arboreal in their mode of life. In all the purely terrestrial types, such as the common agama lizards, or stellions, of the Old World, for instance, the body is more or less markedly depressed, or flattened from above downwards, and much ex- panded laterally, so that the sides form comparatively sharp edges, while the limbs, which are generally short, thick, and powerful, are widely sundered from one another at their points of origin from the shoulders and haunches. Only a moment’s reflection is required to show that such a construction of body and limbs is obviously the one best adapted to enable the owner to escape detection from enemies by squeezing itself as flatly and closely as possible upon the rock or sand upon which it happens to be resting, more especially when (as is almost in- variably the case) its colour harmonises with that of its sur- roundings. The above-mentioned agamas, or stellions, for instance, are generally coloured some shade of mottled grey, brown, and blackish, so as to harmonise closely with the lichen- go PLATE VI MOLOCH LIZARD (YVOLOCH HORRIDUS) le “HORNED TOAD” (PHRYNOSOMA CORNUTUM). TO SHOW DEPRESSED FORM OF BODY CHARACTERISTIC OF TERRESTRIAL TYPES OF LIZARDS id BEARDED LIZARD (AMPHIBOLURUS BARBATUS) OF AUSTRALIA ADAPTATIONS gI spotted rocks upon which they delight to bask in the sun. On the other hand, desert-haunting species are more usually of a sandy tint. The extreme development of the depressed type of bodily form occurs in species which dwell on sandy or other soft soil, where the body can be closely pressed to the ground; this supreme development being found in species belonging to dif- ferent family groups, thus demonstrating that its origin is de- pendent on adaptation to habits and environment and has nothing to do with zoological affinity. The best example of this is afforded by the two lizards respectively known as the moloch (Moloch horridus) of the deserts of Western and Southern Australia, and the American so-called “horned toad” (Phrynosoma cornutum), which inhabits similar situations in California. The former—the sole representative of its genus— belongs to the family Agamzde, while the latter, which is one out of about a dozen species of the same genus, is a member of the /guanide. Both have the body so depressed and flattened that it is nearly oval in shape; and in both, the head, body, and tail are covered with a number of short spines, evidently either for protection or to aid in concealment. In the moloch the colour is pale brown blotched with chocolate, while in the horned toad the prevailing tint is a mixture of yellow, brown, grey, and black. When squatting closely down in the sands of their native deserts, with which the colour of each harmonises, both these lizards must be difficult to detect ; and when they are recognised, their prickly coats must render them difficult to pick up. Any non-scientific person seeing the two species side by side would almost certainly declare that they must be near relations ; whereas their superficial resemblance to one another is a case of parallelism in development for the purpose of adaptation to their surroundings and consequent concealment from enemies (Pl. V.). The iguanas of the genus Sce/oporus are also depressed terrestrial forms, quite unlike the arboreal compressed representatives of the family. The spiny-tailed lizards (Uvomastix) of the deserts of the Old World, which belong to the same family as the moloch, may be cited as other examples of the depressed type of body, although the depression is not carried to the same degree as in the instances described above. The spiny-tails are burrowing 92 REPTILES lizards, which dwell in holes: these they invariably enter head- first, and the spiny tail consequently blocks the entrance in a most effectual manner to all would-be intruders—another in- stance of structural modification and adaptation for protective purposes. Crocodiles also offer an example of the depressed type of bodily form: but since these reptiles are to a great extent aquatic in their habits, the large and powerful tail has assumed a compressed form, so as to act as an oar or rudder in swimming ; its efficiency in this respect being increased by the elevation of the two edges of the upper surface into crests. Although this depressed type of body is characteristic of purely terrestrial species, it would be a mistake to suppose that it is entirely confined to reptiles with such habits. On the con- trary, it occurs in species habitually dwelling in trees, or fre- quenting walls or vertical rock-surfaces ; and, strictly speaking, therefore, this type of bodily contour may be met with among arboreal reptiles. It is, however, better to term species with this habit trunk-haunting or wall-haunting reptiles, for they are not arboreal in the sense in which an iguana or a chameleon is, that is to say, they do not dwell on the small branches and amid the leaves, but cling tightly to the bark of the trunk and larger boughs. The best examples of trunk-haunting reptiles with this depressed and expanded type of body are afforded by certain geckos, such as the Turkish gecko (/lemmidactylus turcicus) and that variety of the Malagasy bark-gecko known as Uroplates fimbriatus lichenium. In the latter not only is the body greatly depressed and expanded, but the short and trowel-like tail is modified in the same manner; and since, as already stated, the colouring accords in a marvellous degree with lichen-clad bark, the reptile must be practically invisible when clinging to the trunks it frequents. A further develop- ment of the depressed type of body is presented by the fringed gecko (Ptychozoum homalocephalum) of the Malay countries, in which the sides of the head, body, limbs, and tail are bordered by a thin membrane of considerable width. When clinging to a trunk, this membrane causes the outlines of the body to merge imperceptibly into the bark. If, however, it be true that it also serves as a parachute, this fringe must have a double function, namely as an aid in protection and as an organ of flight. The arboreal type of bodily form, as presented in its ADAPTATIONS 93 most characteristic development by iguanas (/gwanid@) and chameleons (Chameleontid@), is precisely the reverse of the _terrestrial, that is to say, in place of being depressed, the body is compressed or flattened from side to side, so that the back and belly take the form of more or less sharp ridges, while the sides are extensive flattened surfaces. The tail partakes in a greater or less degree of the same modification ; while the limbs, which are often of considerable length, must necessarily be separated from their fellows of the opposite side by a com- paratively small space at the shoulders and haunches. Now it is clear that a reptile of this shape is admirably adapted to escape detection when standing on or clinging to a bough, as it may be easily mistaken for a broken branch, or, if coloured green, for a leaf or bunch of leaves. But chameleons (and for aught I know, iguanas also) go one better than this, for when a stranger approaches the tree or shrub on which they may be resting, every one of them promptly moves to the opposite side of the branch, when its thin body is more or less com- pletely eclipsed. In the foregoing paragraph it has been mentioned that iguanas belong to one family of reptiles and chameleons to another; from which we see that, like the depressed type, the compressed form has been independently developed in differ- ent groups. This, however, is by no means all, for we find among reptiles of the latter type an instance where species be- longing to different families have acquired a superficial re- semblance analogous to the one existing between the horned toad and the moloch lizard, and due to parallelism in adaptive development. The species between which this resemblance is most marked are the Indian chamzleon (Chameleo calcaratus) on the one hand, and the chameleon-lizard (Ganyocephalus chameéleontinus) of the Malay countries on the other; the former typifying a family by itself, while the latter isa member of the Agamide. Both species display the same helmet-like form of the head, the laterally-compressed body, with a sharply keeled back, and the long tapering tail. The general colour of the two is likewise very similar. A certain difference is notice- able with regard to the spines on the back, which are more distinct in the chamzleon-lizard; but such a difference might be specific. Careful examination will show that the two reptiles 94 REPTILES are really very different. ‘The chameleon, for instance, has a eranulated skin, while that of the chamezleon-lizard is scaly. The latter species also lacks the telescopic eye of the chamzleon, and the toes of each foot are not divided into opposing groups for grasping, nor is the tail prehensile. The chamzleon’s pro- trusile tongue has, moreover, no parallel in the lizard. The resemblance between the two reptiles is, in fact, super- ficial, and doubtless correlated with their mode of life, both being purely arboreal creatures, of slow and sluggish habits, and feeding upon insects. The strange thing about the matter is that the resemblance should be as close as it is, seeing that chameleons are unknown in the countries east of the Bay of Bengal, and therefore that it cannot be due to mimicry of the one species by the other. The suggestion might arise that chameleons once inhabited the Malay countries, but there is no evidence of this. Moreover, the chameleon-lizard belongs to a rather large genus, some of the members of which inhabit India and Ceylon, where chameleons are found ; but these species do not present anything like the same resemblance to chameleons as is shown by the chameleon-lizard and a few allied Malay species. In connection with chameleons, it may be mentioned that these reptiles exhibit a modification in the structure of the feet unique in the class, although paralleled among birds by cuckoos and certain allied groups. Not only are the limbs of chame- leons relatively long and slender, but two of the toes of each foot are permanently opposed to the other three; the first three toes in the fore-foot being opposed to the other two, while in the hind limb the inner division includes only the first and second toes, to which the other three are opposed. Additional aid in climbing is afforded by the prehensile tail, the tip of which can be curled downwards round a branch. Here, then, we have a function parallel to what exists among arboreal mammals, such as spider-monkeys, tree-porcupines, and opossums. ‘The tail of the pythons and certain other snakes is also endowed with the power of prehension, Geckos display a different modification of the toes for the purpose of enabling them to cling to walls, cliffs, and even ceilings, upon the latter of which they run back-downwards, like flies. This clinging function is effected by means of a number A.—CHAMAELEON LIZARD. TO SHOW SIMILARITY OF ARBOREAL B.—COMMON FORMS IN WIDELY CHAMAELEON DIFFERENT TYVES AVAPLATIONS 95 of spaces analogous in their action to suckers on the under side of the feet; this surface on each toe being divided into such spaces by means of a series of tranverse plates in the skin. When the foot is pressed upon a flat surface the soft, and yield- ing plates are squeezed flat and the air between them is con- sequently driven out. Elevation of the centre of the foot, which in some cases is webbed, consequently produces vacuums between the plates which act as suckers; these vacuums being rendered more effectual than would otherwise be the case in the non-webbed species by the presence of a number of minute hairs on the edges of the plates. This sucker-like structure is pre- sent on the feet of many if not most geckos. In certain species, such as Zeratosaurus scincus, of Persia and Turkestan, which have forsaken their climbing habits in favour of an existence on desert sands, the clinging apparatus has been lost or modi- fied, so that the foot has reverted to a more normal type. Another modification in connection with climbing is ex- hibited by the hind-feet of the iguanas, which attains its maxi- mum development in the chamzleon-iguanas of the genus Azo/zs. In these reptiles the outermost, or fifth toe, is widely separated from the other four, which are elongated, and branches off at the root of the foot, so that its point of origin is higher up than that of the rest. In Avxo/is it looks quite distinct from the rest of the foot, and the whole foot is designed to form an efficient climbing instrument. A few lizards, such as some of the American iguanas, and more especially the frilled lizard (Chlamydosaurus king?) of Australia, whose bodies are formed on the compressed arboreal type, have deserted to a greater or less degree their original climbing habits, and taken to an existence on open sandy ground, where they run at times on the hind-legs, in the up- right posture with the fore-limbs folded and hanging by the sides of the body ; the large throat-frill of the Australian species being on such occasions folded up like a badly-made umbrella. The frilled lizard cannot, however, maintain this running gait for any length of time; and after a bit will either turn to bay at the foot of a tree, or run up the stem of the latter in the manner of a climbing lizard. The frilled lizard is a member of the Agamzde. The same cursorial type occurs in certain smaller carnivorous representatives of the extinct dinosaurs, 96 REPTILES such as Compsognathus longipes, from the Upper Jurassic lithographic limestone of Bavaria and Ornithomimus altus of the Cretaceous rocks of North America; and it has consequently been suggested that the frilled lizard has inherited the upright posture and running gait from dinosaurian ancestors. Nothing could be further from the truth, lizards being widely sundered from dinosaurs. ‘The acquisition of the running habit in the frilled lizards, in certain iguanas, and in the carnivorous dino- saurs, on the contrary, is another instance of the independent development of similar adaptations. From a study of the skeleton of the fore-limb of the dinosaur Ovzzthomimus palzon- tologists have come to the conclusion that the fore-foot was en- dowed with grasping and seizing power and consequently capable of acting as a hand; and since these reptiles, from the structure of their teeth, were evidently carnivorous, it has been suggested that they were in the habit of capturing on the wing the contemporary lizard-tailed birds, such as Avch@opteryx, whose flight was probably slow, heavy, and low. From the foregoing cursorial type, in which the upright posture was assumed either occasionally or permanently, the transition is easy to the giant herbivorous dinosaurs, such as Lguanodon of the European Wealden and a number of allied forms from the Upper Jurassic and Cretaceous strata of North America, in which the bipedal posture was habitual, although the gait was a walk, as is indicated by the impressions of the footsteps of these reptiles found in the sandstone of Hastings. The larger kinds of iguanodon (which take their name from a fancied resemblance between their teeth and those of modern iguanas) stood approximately twenty feet in height, and walked on their hind-limbs, with perhaps some support from the long and heavy tail, which probably acted as a counter- poise to the head and fore part of the body. Apparently the general shape of the trunk approximated more or less closely to the compressed arboreal type, although there may have been some greater degree of expansion in the region of the chest to allow of the free play of the fore-limbs which were evidently used as arms. The iguanodons and their allies were by no means the only large dinosaurs which habitually walked in the upright posture. The carnivorous J/ega/osaurus, for example, which when in this posture stood about a dozen feet in height, PATRATCE VTL AUSTRALIAN FRILLED LIZARD (CHAMYDOSAURUS KING/) FROM A STUFFED SPECIMEN SAME IN RUNNING POSITIONS PHOTOGRAPHED FROM LIFE ~ ih, - = > e- J sf