Similarly, there is no correlation in the rate of evolution either | living animals but only through the long perspective gained by of adjoining or of separated parts;. the middle digit of the foot of the three-toed horse is accelerated in development, while the lateral digits on either side are retarded. Many examples might be cited among invertebrates also. LAW OF THE INDEPENDENT ADAPTIVE EVOLUTION OF PARTS. FIG. 11-Diagram demonstrating that there are an, indefinite number of combinations of various adaptive types of limbs and feet with various adaptive types of teeth, and that there is no fixed law of correlation between the two series of adaptations. All these principles are consistent with Francis Galton's law of particulato inheritance in heredity, and with the modern doctrine of "unity of characters" held by students of Mendelian phenomena. Sudden versus Gradual Evolution of Parts.-There is a broad and most interesting analogy between the evolution of parts of animals and of groups of animals studied as a whole. Thus we observe persistent organs and persistent types of animals, analogous organs and analogous types of animals, and this analogy applics still further to the rival and more or less contradictory hypotheses of the sudden as distinguished from the gradual appearance of new parts or organs of animals, and the sudden appearance of new types of animals. The first exponent of the theory of sudden appearance of new parts and new types, to our knowledge, was Geoffroy St Hilaire, who suggested saltatory evolution through the direct action of the environment on development, as explaining the abrupt transitions in the Mesozoic Crocodilia and tho origin of the birds from the reptiles. Waagen's law of mutation, or the appearance of new parts or organs so gradually that they can be perceived only by following them through successive geologic time stages, appears to be directly contradictory to the saltation principle; it is certainly one of the most firmly established principles of palaeontology, and it constitutes the contribution par excellence of this branch of zoology to the law of evolution, since it is obvious that it could not possibly have been deduced from comparison of comparison of animals succeeding each other in time. The essence of Waagen's law is orthogenesis, or evolution in a definite direction, and, if there does exist an internal hereditary principle controlling such orthogenetic evolution, there does not appear to be any essential contradiction between its gradual operation in the "mutations of Waagen" and its occasional hurried operation in the "mutations of de Vries," which are by their definition discontinuous or saltatory (Osborn, 1907). VII.-MODES OF CHANGE IN ANIMALS AS A WHOLE OR IN GROUPS OF ANIMALS, AND METHODS OF ANALYSING THEM. 1. Origin from Primitive or Stem Forms.-As already observed, the same principles apply to groups of animals as to organs and groups of organs; an organ originates in a primitive and unspecialized stage, a group of animals originates in a primitive or stem form. It was early perceived by Huxley, Cope and many others that Cuvier's broad belief in a universal progression was erroneous, and there developed the distinction between "persistent primitive types" (Huxley) and progressive types." The theoretical existence of primitive or stem forms was clearly perceived by Darwin, but the steps by which the stem form might be restored were first clearly enunciated by Huxley in 1880 ("On the Application of Evolution to the Arrangement of the Vertebrata and more particularly of the Mammalia," Scient Mem. iv. 457) namely, by sharp separation of the primary or stem characters from the secondary or adaptive characters in all the known descendants or branches of a theoretical original form. The sum of the primitive characters approximately restores the primitive form; and the gaps in palaeontological evidence are supplied by analysis of the available zoological, embryological and anatomical evidence. Thus Huxley, with true prophetic instinct, found that the sum of primitive characters of all the higher placental mammals points to a stem form of a generalized insectivore type, a prophecy which has been fully confirmed by the latest research. On the other hand, Huxley's summation of the primitive characters of all the mammals led him to an amphibian stem type, a prophecy which has proved faulty because based on erroneous analysis and comparison. More or less independently, Huxley, Kowalevsky and Cope restored the stem ancestor of the hoofed animals, or ungulates, a restoration which has been nearly fulfilled by the discovery, in 1873, of the generalized type Phenacodus of northern Wyoming. Similar anticipations and verifications among the invertebrates have been made by Hyatt, Beecher, Jackson and others. In certain cases the character stem forms actually survive in unspecialized types. Thus the analysis of George Baur of the ancestral form of the lizards, mosasaurs, dinosaurs, crocodiles and phytosaurs led both to the generalized Palaeohatteria of the Permian and indirectly to the surviving Tuatera lizard of New Zealand. 2. Adaptations to Alternations of Habitat, Law of Irreversi bility of Evolution. In the long vicissitudes of time and proces sion of continental changes, animals have been subjected to alternations of habitat either through their own migrations or through the "migration of the environment itself," to employ Van den Broeck's epigrammatic description of the profound and sometimes sudden environmental changes which may take place in a single locality. The traces of alternations of adaptations corresponding to these alternations of habitat are recorded both in palaeontology and anatomy, although often after the obscure analogy of the earlier and later writings of a palimpsest. Huxley in 1880 briefly suggested the arboreal origin, or primordial treehabitat of all the marsupials, a suggestion abundantly confirmed by the detailed studies of Dollo and of Bensley, according to which we may imagine the marsupials to have passed through (1) a former terrestrial phase, followed by (2) a primary arboreal phase-illustrated in the tree phalangers-followed by (3) a secondary terrestrial phase-illustrated in the kangaroos and wallabies-followed by (4) a secondary arboreal phase-ilustrated in the tree kangaroos. Louis Dollo especially has contributed most brilliant discussions of the theory of alter- or more persistent primitive type, which was contemporaneous nations of habitat as applied to the interpretation of the anatomy of the marsupials, of many kinds of fishes, of such reptiles as the herbivorous dinosaurs of the Upper Cretaceous. He has applied the theory with especial ingenuity to the interpretation of the circular bony plates in the carapace of the aberrant leather-back sea-turtles (Sphargidae) by prefacing an initial land phase, in which the typical armature of land tortoises was acquired, a first marine or pelagic phase, in which this armature was lost, a third littoral or seashore phase, in which a new polygonal armature was acquired, and a fourth resumed or secondary marine phase, in which this polygonal armature began to degenerate. Each of these alternate life phases may leave some profound modification, which is partially obscured but seldom wholly lost; thus the tracing of the evidences of former adaptations is of great importance in phylogenetic study. A very important evolutionary principle is that in such secondary returns to primary phases lost organs are never recovered, but new organs are acquired; hence the force of Dollo's dictum that evolution is irreversible from the point of view of structure, while frequently reversible, or recurrent, in point of view of the conditions of environment and adaptation. 3. Adaptive Radiations of Groups, Continental and Local.Starting with the stem forms the descendants of which have passed through either persistent or changed habitats, we reach the underlying idea of the branching law of Lamarck or the law of divergence of Darwin, and find it perhaps most clearly expressed in the words "adaptive radiation" (Osborn), which convey the idea of radii in many directions. Among extinct Tertiary mammals we can actually trace the giving off of these radii in all directions, for taking advantage of every possibility to secure food, to escape enemies and to reproduce kind; further, among such well-known quadrupeds as the horses, rhinoceroses and titanotheres, the modifications involved in these radiations can be clearly traced. Thus the history of continental life presents a picture of contemporaneous radiations in different parts of the world and of a succession of radiations in the same parts. We observe the contemporaneous and largely independent radiations of the hoofed animals in South America, in Africa and in the great ancient continent comprising Europe, Asia and North America; we observe the Cretaceous radiation of hoofed animals in the northern hemisphere, followed by a second radiation of hoofed animals in the same region, in some cases one surviving spur of an old radiation becoming the centre of a new one. As a rule, the larger the geographic theatre the grander the radiation. Successive discoveries have revealed certain grand centres, such as (1) the marsupial radiation of Australia, (2) the littleknown Cretaceous radiation of placental mammals in the northern hemisphere, which was probably connected in part with the peopling of South America, (3) the Tertiary placental radiation in the northern hemisphere, partly connected with Africa, (4) the main Tertiary radiation in South America. Each of these radiations produced a greater or less number of analogous groups, and while originally independent the animals thus evolving as autochthonous types finally mingled together as migrant or invading types. We are thus working out gradually the separate contributions of the land masses of North America, South America, Europe, Asia, Africa, and of Australia to the mammalian fauna of the world, a result which can be obtained through palaeontology only. 4. Adaptive Local Radiation.-On a smaller scale are the local adaptive radiations which occur through segregation of habit and local isolation in the same general geographic region wherever physiographic and climatic differences are sufficient to produce local differences in food supply or other local factors of change. This local divergence may proceed as rapidly as through wide geographical segregation or isolation. This principle has been demonstrated recently among Tertiary rhinoceroses and titanotheres, in which remains of four or five genetic series in the same geologic deposits have been discovered. We have proof that in the Upper Miocene of Colorado there existed a forest-living horse, with and is found in the same deposits with the plains-living horse (Neohipparion) of the most advanced or specialized desert type (see Plate IV., figs. 12, 13, 14, 15). In times of drought these animals undoubtedly resorted to the same water-courses for drink, and thus their fossilized remains are found associated. 5. The Law of Polyphyletic Evolution. The Sequence of Phyla or Genetic Scries.-There results from continental and local adaptive radiations the presence in the same geographical region of numerous distinct lines in a given group of animals. The polyphyletic law was early demonstrated among invertebrates by Neumayr (1889) when he showed that the ammonite genus Phylloceras follows not one but five distinct lines of evolution of unequal duration. The brachiopods, generally classed collectively as Spirifer mucronatus, follow at least five distinct lines of evolution in the Middle Devonian of North America, while more than twenty divergent lines have been observed by Grabau among the species of the gastropod genus Fusus in Tertiary and recent times. Vertebrate palaeontologists were slow to grasp this principle; while the early speculative phylogenies of the horse of Huxley and Marsh, for example, were mostly displayed monophyletically, or in single lines of descent, it is now recognized that the horses which were placed by Marsh in a single series are really to be ranged in a great number of contemporaneous but separate series, each but partially known, and that the direct phylum which leads to the modern horse has become a matter of far more difficult search. As early as 1862 Gaudry set forth this very polyphyletic principle in his tabular phylogenies, but failed to carry it to its logical application. It is now applied throughout the Vertebrata of both Mesozoic and Cenozoic times. Among marine Mesozoic reptiles, each of the groups broadly known as ichthyosaurs, plesiosaurs, mosasaurs and crocodiles were polyphyletic in a marked degree. Among land animals striking illustrations of this local polyphyletic law are found in the existence of seven or eight contemporary series of rhinoceroses, five or six contemporary series of horses, and an equally numerous contemporary series of American Miocene and Pliocene camels; in short, the polyphyletic condition is the rule rather than the exception. It is displayed to-day among the antelopes and to a limited degree among the zebras and rhinoceroses of Africa, a continent which exhibits a survival of the Miocene and Pliocene conditions of the northern hemisphere. 6. Development of Analogous Progressive and Retrogressive Groups.-Because of the repetition of analogous physiographic and climatic conditions in regions widely separated both in time and in space, we discover that continental and local adaptive radiations result in the creation of analogous groups of radii among all the vertebrates and invertebrates. Illustrations of this law were set forth by Cope as early as 1861 (see " Origin of Genera," reprinted in the Origin of the Fittest, pp. 95-106) in pointing out the extraordinary parallelisms between unrelated groups of amphibians, reptiles and mammals. In the Jurassic period there were no less than six orders of reptiles which independently abandoned terrestrial life and acquired more or less perfect adaptation to sea life. Nature, limited in her resources for adaptation, fashioned so many of these animals in like form that we have learned only recently to distinguish similarities of analogous habit from the similitudes of real kinship. From whatever order of Mammalia or Reptilia an animal may be derived, prolonged aquatic adaptation will model its outer, and finally its inner, structure according to certain advantageous designs. The requirements of an elongate body moving through the resistant medium of water are met by the evolution of similar entrant and exit curves, and the bodies of most swiftly moving aquatic animals evolve into forms resembling the hulls of modern sailing yachts (Bashford Dean). We owe especially to Willy Kükenthal, Eberhard Fraas, S.W. Williston and R. C. Osburn a summary of those modifications of form to which aquatic life invariably leads. The law of analogy also operates in retrogression. A. Smith Woodward has observed that the decline of many groups of fishes is heralded by the tendency to assume elongate and finally | of De Vries, "or the theory of saltatory evolution, which we may eel-shaped forms, as seen independently, for example, among next consider. the declining Acanthodians or palacozoic sharks, among the modern crossopterygian Polypterus and Calamoichthys of the Nile, in the modern dipneustan Lepidosiren and Protopterus, in the Triassic chondrostean Belonorhynchus, as well as in the bow-fin (Amia) and the garpike (Lepidosteus). Among invertebrates similar analogous groups also develop. This is especially marked in retrogressive, though also wellknown in progressive series. The loss of the power to coil, observed in the terminals of many declining series of gastropods from the Cambrian to the present time, and the similar loss of power among Natiloidea and Ammonoidea of many genetic series, as well as the ostraean form assumed by various declining series of pelecypods and by some brachiopods, may be cited as examples. 7. Periods of Gradual Evolution of Groups.-It is certainly a very striking fact that wherever we have been able to trace genetic series, either of invertebrates or vertebrates, in closely sequent geological horizons, or life zones, we find strong proof of evolution through extremely gradual mutation simultaneously affecting many parts of each organism, as set forth above. This proof has been reached quite independently by a very large number of observers studying a still greater variety of animals. Such diverse organisms as brachiopods, ammonites, horses and rhinoceroses absolutely conform to this law in all those rare localities where we have been able to observe closely sequent stages. The inference is almost irresistible that the law of gradual transformation through minute continuous change is by far the most universal; but many palaeontologists as well as zoologists and botanists hold a contrary opinion. 8. Periods of Rapid Evolution of Groups.-The above law of gradual evolution is perfectly consistent with a second principle, namely, that at certain times evolution is much more rapid than at others, and that organisms are accelerated or retarded in development in a manner broadly analogous to the acceleration or retardation of separate organs. Thus H. S. Williams observes (Geological Biology, p. 268) that the evolution of those fundamental characters which mark differences between separate classes, orders, sub-orders, and even families of organisms, took place in relatively short periods of time. Among the brachiopods the chief expansion of each type is at a relatively early period in their life-history. Hyatt (1883) observed of the ammonites that each group originated suddenly and spread out with great rapidity. Depéret notes that the genus Neumayria, an ammonite of the Kimmeridgian, suddenly branches out into an "explosion" of forms. Depéret also observes the contrast between periods of quiescence and limited variability and periods of sudden efflorescence. A. Smith Woodward ("Relations of Palaeontology to Biology," Annals and Mag. Natural Hist., 1906, p. 317) notes that the fundamental advances in the growth of fish life have always been sudden, beginning with excessive vigour at the end of long periods of apparent stagnation; while each advance has been marked by the fixed and definite acquisition of some new anatomical character or expression point," a term first used by Cope. One of the causes of these sudden advances is undoubtedly to be found in the acquisition of a new and extremely useful character. Thus the perfect jaw and the perfect pair of lateral fins when first acquired among the fishes favoured a very rapid and for a time unchecked development. It by no means follows, however, from this incontrovertible evidence that the acquisition either of the jaw or of the lateral fins had not been in itself an extremely gradual process. Thus both invertebrate and vertebrate palaeontologists have reached independently the conclusion that the evolution of groups is not continuously at a uniform rate, but that there are, especially in the beginnings of new phyla or at the time of acquisition of new organs, sudden variations in the rate of evolution which have been termed variously "rhythmic," "pulsating," 'efflorescent," "intermittent" and even "explosive" (Depéret). This varying rate of evolution has (illogically, we believe) been compared with and advanced in support of the "mutation law 9. Hypothesis of the Sudden Appearance of New Parts er Organs.-The rarity of really continuous series has naturally led palaeontologists to support the hypothesis of brusque transitions of structure. As we have seen, this hypothesis was fathered by Geoffroy St Hilaire in 1830 from his studies of Mesozoic Crocodilia, was sustained by Haldemann, and quite recently has been revived by such eminent palaeontologists as Louis Dollo and A. Smith Woodward. The evidence for it is not to be confused with that for the law of rapid efflorescence of groups just considered. It should be remembered that palaeontology is the most unfavourable field of all for observation and demonstration of sudden saltations or mutations of character, because of the limited materials available for comparison and the rarity of genetic series. It should be borne in mind, first, that wherever a new animal suddenly appears or a new character suddenly arises in a fossil horizon we must consider whether such appearance may be due to the non-discovery of transitional links with older forms, or to the sudden invasion of a new type or new organ which has gradually evolved elsewhere. The rapid variation of certain groups of animals or the acceleration of certain organs is also not evidence of the sudden appearance of new adaptive characters. Such sudden appearances may be demonstrated possibly in zoology and embryology but never can be demonstrated by palaeontology, because of the incompleteness of the geological record. 10. Decline or Senescence of Groups.-Periods of gradual evolution and of efflorescence may be followed by stationary or senescent conditions. In his history of the Arictidae Hyatt points out that toward the close of the Cretaceous this entire group of ammonites appears to have been affected with some malady; the unrolled forms multiply, the septa are simplified, the ornamentation becon.es heavy, thick, and finally disappears in the adult; the entire group ends by dying out and leaving no descendants. This is not due to environmental conditions solely, because senescent branches of normal progressive groups are found in all geologic horizons, beginning, for gastropods, in the Lower Cambrian. Among the ammonites the loss of power to coil the shell is one feature of racial old age, and in others old age accompanied by closer coiling and loss of surface ornamentation, such as spines, ribs, spirals; while in other forms an arresting of variability precedes extinction. Thus Williams has observed that if we find a species breeding perfectly true we can conceive it to have reached the end of its racial life period. Brocchi and Daniel Rosa (1899) have developed the hypothesis of the progressive reduction of variability. Such decline is by no means a universal law of life, however, because among many of the continental vertebrates at least we observe extinctions repeatedly occurring during the expression of maximum variability. Whereas among many ammonites and gastropods smooth ness of the shell, following upon an ornamental youthful condition, is generally a symptom of decline, among many other invertebrates and vertebrates, as C. E. Beecher (1856-1905) has pointed out (1898), many aninials possessing hard parts tend toward the close of their racial history to produce a superfuity of dead matter, which accumulates in the form of spines among invertebrates, and of horns among the land vertebrates, reaching a maximum when the animals are really on the down-grade of development. 11. The Extinction of Groups.-We have seen that different lines vary in vitality and in longevity, that from the earliest times senescent branches are given off, that different lines vary in the rate of evolution, that extinction is often heralded by symptoms of racial old age, which, however, vary widely in different groups. In general we find an analogy between the development of groups and of organs; we discover that each phyletic branch of certain organisms traverses a geologic career comparable to the life of an individual, that we may often distinguish, especially among invertebrates, a phase of youth, a phase of maturity, a phase of senility or degeneration foreshadowing the extinction of a type. principles in this orderly process is that there appears to be in respect to certain characters a potentiality or a predisposition through hereditary kinship to evolve in certain definite directions. Yet there is strong evidence against the existence of any law in the nature of an internal perfecting tendency which would operate independently of external conditions. In other words, a balance appears to be always sustained between the internal (hereditary and ontogenetic) and the external (environmental and selectional) factors of evolution. BIBLIOGRAPHY.-Among the older works on the history of palaeontology are the treatises of Giovanni Battista Brocchi (1772dello studio... 1843 (Milan); of Etienne Jules d'Archiac, Histoire 1826), Conchiologia fossile Subappenina. ... Disc. sui progressi du progrès de la géologie de 1834 à 1862 (Paris, Soc. Geol. de France, 1847-1860); of Charles Lyell in his Principles of Geology. A clear in Founders of Geology, by Sir Archibald Geikie (London, 1897narrative of the work of many of the earlier contributors is found 1905). The most comprehensive and up-to-date reference work on the history of geology and palaeontology is Geschichte der Geologie und Paläontologie, by Karl Alfred von Zittel (Munich and Leipzig, 1899), the final life-work of this great authority, translated into Geology and Palaeontology to the end of the 19th Century." The English in part by Maria M. Ogilvie-Gordon, entitled "History of succession of life from the earliest times as it was known at the close of the last century was treated by the same author in his Handbuch der Paläontologie (5 vols., Munich and Leipzig, 1876-1893). Abbreviated editions of this work have appeared from the author, Grundzüge der Paläontologie (Palaeozoologie) (Munich and Leipzig, 1895, 2nd ed., 1903), and in English form in Charles R. Eastman's TextBook of Palaeontology (1900-1902). A classic but unfinished work describing the methods of invertebrate palaeontology is Die Stämme des Thierreichs (Vienna, 1889), by Melchior Neumayr. In France admirable recent works are Eléments de Paléontologie, by Felix Bernard (Paris, 1895), and the still more recent philosophical treatise by Charles Depéret, Les Transformations du monde animal (Paris, 1907). Huxley's researches, and especially his share in the his essays, The Scientific Memoirs of Thomas Henry Huxley (4 vols., development of the philosophy of palaeontology, will be found in London, 1898-1902). The whole subject is treated systematically in Nicholson and Lydekker's A Manual of Palaeontology (2 vols., Edinburgh and London, 1889), and A. Smith Woodward's Outlines of Vertebrate Palaeontology (Cambridge, 1898). Internal causes of extinction are to be found in exaggeration | test of fitness is supreme. The only inkling of possible underlying of body size, in the hypertrophy or over-specialization of certain organs, in the irreversibility of evolution, and possibly, although this has not been demonstrated, in a progressive reduction of variability. In a full analysis of this problem of internal and external causes in relation to the Tertiary Mammalia, H. F. Osborn ("Causes of Extinction of the Mammalia," Amer. Naturalist, 1906, pp. 769-795, 829-859) finds that foremost in the long series of causes which lead to extinction are the grander environmental changes, such as physiographic changes, diminished or contracted land areas, substitution of insular for continental conditions; changes of climate and secular lowering of temperature accompanied by deforestation and checking of the food supply; changes influencing the mating period as well as fertility; changes causing increased humidity, which in turn favours enemies among insect life. Similarly secular elevations of temperature, either accompanied by moisture or desiccation, by increasing droughts or by disturbance of the balance of nature, have been followed by great waves of extinction of the Mammalia. In the sphere of living environment, the varied evolution of plant life, the periods of forestation and deforestation, the introduction of deleterious plants simultaneously with harsh conditions of life and enforced migration, as well as of mechanically dangerous plants, are among the well-ascertained causes of diminution and extinction. The evolution of insect life in driving animals from feeding ranges and in the spread of disease probably has been a prime cause of extinction. Food competition among mammals, especially intensified on islands, and the introduction of Carnivora constitute another class of causes. Great waves of extinction have followed the long periods of the slow evolution of relatively inadaptive types of tooth and foot structure, as first demonstrated by Waldemar Kowalevsky; thus mammals are repeatedly observed in a cul-de-sac of structure from which there is no escape in an adaptive direction. Among still other causes are great bulk, which proves fatal under certain new conditions; relatively slow breeding; extreme specialization and development of dominant organs, such as horns and tusks, on which for a time selection centres to the detriment of more useful characters. Little proof is afforded among the mammals of extinction through arrested evolution or through the limiting of variation, although such laws undoubtedly exist. One of the chief deductions is that there are special dangers in numerical diminution of herds, which may arise from a chief or original cause and be followed by a conspiracy of other causes which are cumulative in effect. This survey of the phenomena of extinction in one great class of animals certainly establishes the existence of an almost infinite variety of causes, some of which are internal, some external in origin, operating on animals of different kinds. VIII.-UNDERLYING BIOLOGICAL PRINCIPLES AS THEY It follows from the above brief summary that palaeontology affords a distinct and highly suggestive field of purely biological research; that is, of the causes of evolution underlying the observable modes which we have been describing. The net result of observation is not favourable to the essentially Darwinian view that the adaptive arises out of the fortuitous by selection, but is rather favourable to the hypothesis of the existence of some quite unknown intrinsic law of life which we are at present totally unable to comprehend or even conceive. We have shown that the direct observation of the origin of new characters in palaeontology brings them within that domain of natural law and order to which the evolution of the physical universe conforms. The nature of this law, which, upon the whole, appears to be purposive or teleological in its operations, is altogether a mystery which may or may not be illumined by future research. In other words, the origin, or first appearance of new characters, which is the essence of evolution, is an orderly process so far as the vertebrate and invertebrate palaeontologist observes it. The selection of organisms through the crucial test of fitness and the shaping of the organic world is an orderly process when contemplated on a grand scale, but of another kind; here the Among American contributions to vertebrate palaeontology, the development of Cope's theories is to be found in the volumes of his collected essays, The Origin of the Fittest (New York, 1887), and The Primary Factors of Organic Evolution (Chicago, 1896). A brief summary of the rise of vertebrate palaeontology is found in the address of O. Marsh, entitled "History and Methods of Palaeontological Discovery" (American Association for the Advancement of Science, 1879). The chief presentations of the methods of the American school of invertebrate palaeontologists are to be found in A. Hyatt's great memoir "Genesis of the Arietidae" (Smithsonian Contr. to Knowledge, 673, 1889), in Hyatt's "Phylogeny of an Acquired Characteristic (Philosophical Soc. Proc., vol. xxxii. 1894), and in Geological Biology, by H. S. Williams (New York, 1895). In preparing the present article the author has drawn freely on his own addresses: see H. F. Osborn, The Rise of the Mammalia in North America (Proc. Amer. Assn. Adv. Science, vol. xlii., 1893), "Ten Years' Progress in the Mammalian Palacontology of North America" (Comptes rendus du 6 Congrès intern. de zoologie, session de Bern, 1904). "The Present Problems of Palaeontology (Address before Section of Zool. International Congress of Arts and Science, St Louis, Sept. 1904), "The Causes of Extinction of Mammalia" (Amer. Naturalist, xl. 769–795, 829-859, 1906). (H. F. O.) PALAEOSPONDYLUS, a small fish-like organism, of which the skeleton is found fossil in the Middle Old Red Sandstone From British Museum Guide to Fossil Reptiles and Fishes, by of Achanarras, near Thurso, Caithness. It was thus named studied in detail in 1903 by Professor and Miss Sollas, who | which are employed by geologists to mark three stages in the succeeded in making enlarged models of the fossil in wax. development of life on the earth, are based primarily upon the The skeleton as preserved is carbonized, and indicates an eel- fossil contents of the strata which, at one point or another, have shaped animal from 3 to 5 cm., in length. The skull, which been continuously forming since the very earliest times. The must have consisted of hardened cartilage, exhibits pairs of precise line in the "record of the rocks" where the chronicle nasal and auditory capsules, with a gill-apparatus below its of the Palaeozoic era closes and that of the Mesozoic era openshinder part, but no indications of ordinary jaws. The anterior as in more recent historical documents—is a matter for editorial opening of the brain-case is surrounded by a ring of hard cirri. caprice. The early geologists took the most natural dividing A pair of "post-branchial plates" projects backwards from the lines that came within their knowledge, namely, the line of change head. The vertebral axis shows a series of broad rings, with in general petrological characters, e.g. the "Transition Series" distinct neural arches, but no ribs. Towards the end of the body (Übergangsgebirge), the name given to rocks approximately of both neural and haemal arches are continued into forked Palaeozoic age by A. G. Werner because they exhibited a transiradial cartilages, which support a median fin. There are no tional stage between the older crystalline rocks and the younger traces either of paired fins or of dermal armour. The affinities non-crystalline; later in Germany these same rocks were said to of Palacospondylus are doubtful, but it is probably related to have been formed in the "Kohlenperiode " by H. G. Bronn and the contemporaneous armoured Ostracoderms. others, while in England H. T. de la Beche classed them as a Carbonaceous and Greywacke group. Finally, the divisional time separating the Palaeozoic record from that of the Mesozoic was made to coincide with a great natural break or unconformity of the strata. This was the most obvious course, for where such a break occurred there would be the most marked differences between the fossils found below and those found above the physical discordance. The divisions in the fossil record having been thus established, they must for convenience remain, but their artificiality cannot be too strongly emphasized, for the broad stratigraphical gaps and lithological groups which made the divisions sharp and clear to the earlier geologists are proved to be absent in other regions, and fossils which were formerly deemed characteristic of the Palaeozoic era are found in some places to commingle with forms of strongly marked Mesozoic type. In short, the record is more nearly complete than was originally supposed. REFERENCES.-R. H. Traquair, paper in Proc. Roy. Phys. Soc. Edin., xii. 312, (1894); W. J. Sollas and I. B. J. Sollas, paper in Phil. Trans. Roy. Soc. (1903 B.). (A. S. Wo.) PALAEOTHERIUM (i.e. ancient animal), a name applied by Cuvier to the remains of ungulate mammals recalling tapirs in general appearance, from the Lower Oligocene gypsum quarries of Paris. These were the first indications of the (From the Paris gypsum.) Restoration of Palaeotherium magnum. (About nat. size.) occurrence in the fossil state of perissodactyle ungulates allied to the horse, although it was long before the relationship was recognized. The palaeotheres, which range in size from that of a pig to that of a small rhinoceros, are now regarded as representing a family, Palaeotheriidae, nearly related to the horsetribe, and having, in fact, probably originated from the same ancestral stock, namely, Hyracotherium of the Lower Eocene (see EQUIDAE). The connecting link with Hyracotherium was formed by Pachynolophus (Propalacotherium), and the line apparently terminated in Paloplotherium, which is also Oligocene. Representatives of the family occur in many parts of Europe, but the typical genus is unknown in North America, where, however, other forms occur. Although palaeotheres resemble tapirs in general appearance, they differ in having only three toes on the fore as well as on the hind foot. The dentition normally comprises the typical series of 44 teeth, although in some instances the first premolar is wanting. The cheek-teeth are short-crowned, generally with no cement, the upper molars having a W-shaped outer wall, from which proceed two oblique transverse crests, while the lower ones carry two crescents. Unlike the carly horses, the later premolars are as complex as the molars; and although there is a well-marked gap between the canine and the premolars, there is only a very short one between the former and the incisors. orbit is completely open behind. In other respects the palaeotheres resemble the ancestral horses. They were, however, essentially marsh-dwelling animals, and exhibit no tendency to the cursorial type of limb so characteristic of the horse-line. They were, in fact, essentially inadaptive creatures, and hence rapidly died out. (R. L.*) The The Palacozoic or Primary era is divided into the following periods or epochs: Cambrian, Ordovician, Silurian, Devonian, Carboniferous and Permian. The fact that fossils found in the rocks of the three earlier epochs-Cambrian, Ordovician, Silurian -have features in common, as distinguished from those in the three later epochs has led certain authors to divide this era into an earlier, Protozoic (Proterozoic) and a later Deuterozoic time. The rocks of Palaeozoic age are mainly sandy and muddy sediments with a considerable development of limestone in places. These sediments have been altered to shales, slates, quartzites, &c., and frequently they are found in a highly metamorphosed condition; in eastern North America, however, and in north-east Europe they still maintain their horizontality and primitive texture over large areas. The fossils of the earlier Palaeozoic rocks are characterized by the abundance of trilobites, graptolites, brachiopods, and the absence of all vertebrates except in the upper strata; the later rocks of the era are distinguished by the absence of graptolites, the gradual failing of the trilobites, the continued predominance of brachiopods and tabulate corals, the abundance of crinoids and the rapid development of placoderm and heterocercal ganoid fishes and amphibians. The land plants were all cryptogams, Lepidodendron, Sigillaria, followed by Conifers and Cycads. It is obvious from the advanced stage of development of the organisms found in the earliest of these Palaeozoic rocks that the beginnings of life must go much farther back, and indeed organic remains have been found in rocks older than the Cambrian; for convenience, therefore, the base of the Cambrian is usually placed at the zone of the trilobite Olenellus. (J.A.H.) PALAEPHATUS, the author of a small extant treatise, entitled Пepi 'AniσTшv (On "Incredible Things "). It consists of a series of rationalizing explanations of Greek legends, without any attempt at arrangement or plan, and is probably an epitome, composed in the Byzantine age, of some larger work, perhaps the Aure. Tŵv μvoikŵs eiputrov, mentioned by Suidas as the work of a grammarian of Egypt or Athens. Suidas himself ascribes a ПTepi 'AπiσTшv, in five books, to Palaephatus of Paros or PALAEOZOIC ERA, in geology, the oldest of the great time Priene. The author was perhaps a contemporary of Euhemerus divisions in which organic remains have left any clear record. (3rd century B.C.). Suïdas mentions two other writers of the The three broad divisions-Palaeozoic, Mesozoic, Cainozoic-name: (1) an epic poet of Athens, who lived before the time of |