or less exclusively fertile among themselves; in other words, when crossed with individuals of even allied groups, they produce offspring which are more or less infertile inter se, in the first or the second generation. Such a group may be provisionally regarded as a species. But it is also often found that, within the group, there are certain subordinate aggregates which may be differentiated from one another by the same kind of tests that were employed in forming the group itself, though these traits are present in a markedly inferior degree. In other words, the subordinate aggregates are marked off by peculiar distinct hereditary characters, and they show greater fertility in the second, third, and succeeding generations, inter se, than when crossed with other subordinate groups. Such a subordinate group may be said to form a subspecies or race.

Application to Man.-I. Like Characters: Formation of Types.Essentially unsound are all classifications based on a single character, whether it is the pigmentation of skin, hair, and iris, the texture of hair with shape of transverse section, the nasal index, the cephalic index, or the geometrical varieties of the cranial or the facial form. Nor does it help to employ single characters successively in continued sub-division, e.g., first dividing by hair, sub-dividing by pigmentation, sub-dividing still by cranial and facial form, or in the reverse order. This dislocates natural affinities, and frustrates a sound serial arrangement.

It is necessary to adopt biometric methods in studying characters and variations, and to find the mean or means by co-ordination and seriation. Averages are apt to be misleading, and conceal the differences of type that may exist in a group, except where very extensive observations have been made under a variety of conditions. The range of variations in a character is as important an index as the character itself, and the variations should be studied, not merely among the adults, but with reference to sex and ontogenetic (including embryonic) development, as well as to reversion and retrogression. These are of great value in determining the pure stocks in a heterogeneous mixture as well as racial affinities and distinctions.

We may arrange the types of physical race in several ways

(1) We may classify them as primary, secondary, tertiary, and so on, but this can not be properly done until the effects of environment and miscegenation have been studied by the biologist, and not by the statistical empiric, as has often been the case. When sufficient data are available, we may represent the formation of the physical types by a modified genealogical tree (with devices for intercrossing and retrogression), or by symbols and formulæ analogous to those of organic chemistry (as in arranging isomers, polymers, &c.). But even chemistry is becoming increasingly evolutionary, and the descent of the elements

(and their seven or sixteen races), with the position, say, of the helium-argon group, will shortly be discussed as hotly as the affinities of the Mediterranean

race.

(2) A second way would be to arrange the types in space (or, more simply, on a plane surface), the distances along different directions marking the degree of affinity as estimated by three (or two) groups of correlated characters (cf. the horizontal and vertical rows of the periodic classification in chemistry).

(3) A third way would be to conceive an ideal type as the goal towards which the normal development of the organism is tending, and to place the actual types round this as a centre, at distances corresponding, more or less, to their approximation to the ideal. The difference between the second and the third method is that while the former is statical, the latter is dynamical. Though the third method is not quite feasible, an occasional application of this test of normal or standard development is a useful corrective.

General Remarks on Morphological Characters.-The morphological characters most useful in distinguishing types of physical race are not necessarily of zoonomic interest. Many of the marks are nonadaptive and useless. As Topinard notes, the facial angle is a rational character in craniometry, but the nasal index of which no rational (or zoonomic) explanation is available, is far more valuable as a racial mark. Secondary parts best furnish such distinctive traits. Again, most of the morphological distinctions do not connote vitality, or a high or low place in the normal scale of development. The head and the foot do not vary among races according to their order of superiority. A long head (a so-called simian character), or a long foot, is not a characteristic of inferiority. Taking prognathism (the true or sub-nasal prognathism), while all races are prognathous, some of the neolithic European races were less prognathous than modern Europeans (e.g., Parisians); and the Polynesians of the purest blood, and (probably) the Tasmanians come nearer to the white races than the yellow races or the African Negroes (Topinard). As Weisbach remarks, each race has its share of the characteristics of inferiority. As regards the ideal of normal development for the human body, it is disputed whether the infantile or the adult condition of man makes the nearest approach to it. The young of the anthropoid apes and of man are somewhat alike, but the adult in both cases falls away from this, not in the same direction but along collateral lines, the deviation being much greater in the adult ape than in the adult man. On the whole, as Havelock Ellis notes, "the yellow races are nearest to the infantile condition" (in brachycephaly, scanty hair, proportion of trunk and limbs); "Negroes and Australians are farthest removed from it (though not always in the direction of the ape); the Caucasian races occupy an intermediate position. In the nose" (and also in the well-developed calves as contrasted with the Negro's spindle-legs) "they are at the farthest remove from the ape; in the hairy covering they recede from the human and approach the ape. The lowest races

are in some respects more highly evolved than the white Caucasian races." From the ensemble of osteological characters, it appears that the Australians, the South Sea Islanders, and the Negritos have affinities to the Pithecanthropus erectus, the Polynesians to the Orang, the Negroes to the Gorilla, the Mongols to the Chimpanzee, and two of the original European types, the Neanderthal man and the Aurignac man, to the Gorilla and the Orang respectively (Klaatsch).

Physiological and Pathological Characters: General Remarks.— The characters relating to metabolism and reproduction are of greater bionomic value than any of the morphological ones. The number of red corpuscles and the amount of hæmoglobin in the blood, the pulserate, the vital capacity, the muscular strength, the amount of urea in the urine, are different in different races. But they depend in part on the quantity of proteid consumption. This has been conclusively established by clinical researches in India into the metabolism of peoples with a vegetarian diet. Indeed, some of the morphological characters (e.g., pigmentation of skin, hair, and eye, amount and distribution of hairy covering, &c.), are themselves due to physicochemical processes connected with the metabolism (as well as the secretions) of the organism. The racial differences in muscular force and in vital capacity (as measured by the dynamometer and the spirometer), like those in stature and weight, depend on conditions of nutrition and habitat (including climate), though the costal breathing of civilised as contrasted with the abdominal breathing of uncivilised women has arisen from conventions of dress. The depth and range of the voice also furnish racial characters. In the lower races (as in women), the larynx is less developed than in the higher, and the voice is shriller. Still the Germans are not at the top; the Tartars appear to have even louder and more powerful voices. Thus sexual selection (if this is the origin), like natural selection, does not always work advantageously for the so-called higher races, nor in all directions.

The resistance to particular local diseases that marks particular races may have been due to the elimination of the more susceptible through that selective mortality, which, in the view of Karl Pearson and Archdall Reid, is the most effective instrument of natural selection among the races of men.

Acclimation appears to depend in part on the quantity of water in the organism, the tropics requiring more water than temperate countries (Kochs). On the other hand, cold climates require more proteid than hot. Pure or primitive stocks are less easily acclimated than civilised (or mixed) stocks; the latter are more cosmopolitan. Loss of vital energy owing to chemical changes in metabolism, incapacity to resist diseases of bacterial origin (the phagocytes in the

blood being without the supply of the requisite opsons), and, finally, sterility or diminishing fertility of the germ-plasm due to changes in the environment, food, and habits of life-these are the circumstances that set a limit to the cosmopolitanism of a race, and baffle successful acclimation and colonisation.

II. Stability of Characters and Type.-Both morphological and physiological characters change with change of environment. The chemical changes due to the new conditions of climate or nutrition act upon the "hormones" and enzymes, stimulate cell-growth, induce changes of form in the somatic tissues, and, sometimes, affecting the germ-plasm, become hereditary. This is not merely in the fungi, algæ, flowering plants (Klebs), or in protozoa, sponges, sea-urchins (Roux, Herbst), or in insects (Weismann, Tower, &c.), but also in domesticated animals and in man. The rate of change and the amount vary, being less in pure stocks of long standing and more in mixed or recent stocks.

Evidence is gradually accumulating to show that other morphological characters, e.g., the changes rung on a few simple varieties of geometrical form, in the structure of the hair, the face, the orbits, the nose, the cranium and the pelvis, are not so stable as some physical anthropologists would fain believe. That remarkable osteological changes of this description may be induced in mammals &c., by the action of environment, has indeed been long known (e.g., in the niata cattle, the Java ponies, the Gangetic crocodiles, not to mention oysters and crabs). And the recent careful inquiry of Professor Boas into the anatomical characters of United States immigrants, under the direction of the Immigration Committee of Congress, shows that profound changes of head-form (cephalic index) occur under the influence of American environment, in the American-born descendants of immigrants as compared with the foreign-born immigrants of the same races; that the amount of change in the American-born depends in part on the period of their immigrant mothers' stay in America before their birth; that the rate of change decreases as this period increases, and finally that the changes make the most divergent types (e.g., East European Hebrews and Sicilians) converge and approach to a uniform type in this respect. The cephalic index in man, even if it were not otherwise open to dispute as confounding real distinctions of shape, seems to be unstable under special conditions. That the changes of head-form in American-born children are persistent and hereditary under American conditions may be presumed from he fact that they are in the direction of the normal American ype. That there may be a reversion with a return to European conditions cannot be urged as an objection against one who denies

1

the racial significance of this cephalic character. The persistence of the Neanderthal type or the Aurignac type, so far as this is a fact, may be due to the operation of similar conditions, or the absence of special modifying agencies, or, in some instances, to atavism, reversion or freaks.

Proto-man.-A proto-human type with primitive characters must be assumed as the starting-point, a generalised type from which all the pure primary stocks of Man may be derived by further differentiation and specialisation along different collateral lines in special environments.

The Proto-man as a more generalised form possessed this (phylogenetic) variability in a greater measure, and his skull, cerebral mass and cerebral convolutions have shown striking changes; in other words, the evolution of man has been rapid and continuous in the direction in which he differentiated from the anthropoids. For example, the cranial capacity of the gorilla is about 450 c.c.; of the Pithecanthropus erectus, in Upper Pliocene, about 900 c.c.; of the Neanderthal man, in middle Pleistocene, about 1,250 c.c.; and of the Cro-Magnon man of the lower alluvium about 1,500 c.c. The progress was most marked in the earlier stages, and gradually slowed down.

All this cerebral change is the index of a rapid psychic variation. Even in the case of the higher animals, the psychic (and social) characters are of "zoogenic" value, influencing the course of animal evolution and the origin of species among the higher vertebrates (birds and mammals), e.g., through sexual selection, gregarious impulses, instincts of species-preservation, mutual aid and sympathy. It is these psycho-social characters of the organism that chiefly differentiate Man from the animals. They ensure the exercise of that foresight, control, and co-ordination which are the chief marks of bionomic progress. Besides, what is of vital importance, these psycho-social characters (and therefore the Racial types of Man whereof they are constituent elements) are marked by that greater range, variability and plasticity of response (ie, of the internal factor in organic evolution), which is the concomitant of all higher and more complex organisation. As such, they furnish some new developments, especially an extending range of wants, and the phenomena of choice and conscious control which condition the operation of natural selection, and determine its direction, though they do not by any means suspend it. Hence it is that no view of civilisation is sound or adequate which considers Race and and Racial types statically, and not dynamically as growing, developing, progressive entities.