Applications of Anthropometry

APPLICATIONS OF ANTHROPOMETRY One of the earliest statistical accounts of the physique of chil dren employed in factories appeared in 1833, containing observa tions made at Manchester and Stockport. Records are given of the stature and weight of 1,492 boys and girls. Out of that num ber 1,062 were engaged in factories, although some of the children were only nine years old. The factory hands exhibit inferiority both of height and weight, but the inferiority is not marked until the 16th year in respect of height and the 14th in respect of weight.

Nearly 75 years later a similar contrast was established when "neglected" boys were compared with those of the whole com munity employed as "controls." The contrast is much more definite at 13 to 14 years of age than at five to six years; but it does not apply quite equally to both sexes, for though the girls exhibit the contrast in more marked degree than the boys at the earlier stage, they are not as sharply contrasted at the later age.

Child-life in Scotland.

An exhaustive investigation into the child-life of urban and rural districts, carried out under the gen eral supervision of the Scottish committee for child-life investiga tion and the particular direction of Prof. Noel Paton, and pub lished in 1926, contains a wealth of statistical data. Anthropo metric observations lead to the conclusion that "up to between the first and second years the child of the agricultural labourer is on the average apparently growing more quickly than the urban child. Af ter this, there is a striking parallelism" (p. 1o1). Again (p. 104), "while there is apparently some delay in the rate of growth of the average town child between birth and 18 months, after that date the growth proceeds as rapidly in the town as in the country child." Chart IX. where "urban" boys (Glasgow) are compared with "rural" boys (Forfarshire, Island of Lewis, Ayrshire), shows that the stature of the urban boys is consistently lower than that of rural boys in the period under observation (5 to 15 years).

Instead of parallel lines of development it shows a divergence indicative of a temporary arrest of growth in the "urban" boys from 1 o to 12 years of age. This arrest interrupts the parallelism which would otherwise exist, and though the years of incidence do not tally exactly, this interruption recalls the contrast disclosed by Greenwood's comparison of "neglected" boys of 13 to 14 years of age with the general stock at that age. In the Scottish investi gations the "urban" girls exhibit the same sort of contrast with the "rural" girls as do the two groups of boys; but in the girls the divergence appears rather earlier. In respect of weight, the indication of a corresponding contrast is not altogether lacking, but is much less distinct than in regard to stature. Where the "deviation" occurs, an indication exists of a temporary arrest of growth-activity. Such a temporary arrest implies a periodic al ternation of an active with a resting phase; and that such peri odicity actually obtains, seems to have been fairly proved.

Munich Children.

Observations of large numbers of the school children at Munich were repeated over a series of years with somewhat fluctuating results. It appeared that the measures actu ally recorded (of stature and of weight respectively) manifested first (as in 1921) a reduction when compared with the data for years previous to 1914. Then, in 1924, the children who in that year attained the age of eight exceeded in stature those who reached that age a year previously by as much as 3.2mm. (2.64%) for boys, and 2.8 (2.34%) for girls. The respective weights showed a corresponding contrast. Prof. Rudolf Martin noted the im provement in physique thus revealed in the year 1924, and attrib uted this amelioration to improvements in the conditions of life at Munich. Anthropometry thus provided the data, and a correct interpretation is particularly important in view of the political bearing of the results announced in this investigation.

In a second research made by Prof. Martin, the children of "school age" at Munich were again the subjects of investigation, and on this occasion were compared with those of corresponding age at Chicago. For each of the ages represented by the years 6, 7, 8 . . . 13, the Bavarian children (Munich) fell short of the American children in respect of stature and of weight. Likewise, in each of the eight years in question save the first, the American children showed an extension of variation beyond the uppermost limit of the Munich children.

Physical Types and Disease.

The relation of disease to the physical conformation of the patient has been investigated by many authorities, among whom Dr. Shrubsall, in 1904, published evi dence that different types make different contributions to the sum of inmates of a general hospital. The association of disease with physical type was shown to be more definite in respect of com plexion (the blonde and brunette types respectively) than of a more definitely measurable character such as stature. Neverthe less, there is an association of blondness or brunette-ness with correspondingly contrasted varieties of stature, so that the rela tion of disease to stature is not negligible. Dr. Shrubsall laid stress on indications that the conditions of life in respect of hygiene exercise a sort of selection of particular types to the disadvantage of one and the advantage of another. He added that "to determine whether these phenomena are peculiar to London, or even true, for I have only been able to make some 50,000 observations, an anthropometric survey of the whole country is imperative, and its results would doubtless serve as a basis for any measures of public hygiene which might serve to restore the balance of power, should it be found wanting." That survey has still to be made.

Evidently the extension of such investigations would introduce anthropometric methods into the preparation of data eventually illustrating the significance of the death-rate in various regions. As regards expectation of life, the influences of profession and occupation are acknowledged already; nevertheless the reminder (p. 196) that morbidity is much greater than mortality is very apposite here.

Davenport has employed anthropometric data to distinguish various types of conformation, which he terms "build." He de vised an index based upon the factors of (a) weight and (b) stature conjointly, and he found that this index could be used to denote the characteristic "build" of a particular individual. He proceeded to study the association of the various types of "build" with diseases, and the observations relate to juvenile as well as adult subjects. The results confirm a popular impression. They exhibit prevalence of respiratory diseases in the class of individuals described as "slender" (with an index of "build" averaging 28).

Vascular troubles afflict more usually the very "fleshy" individuals, whose mean or average index of build works out at 48. The data collected by Davenport and his colleagues indicate that "fleshy" individuals prove more prolific than those of "slender" build. Moreover, the character or quality of fleshiness seems slightly to dominate that of slenderness in inheritance.

A more circumscribed type of enquiry involving the use of anthropometric data in the investigation of disease is exemplified in a report by Dr. Berry and Mr. Porteus, who investigated the dimensions of the head in mentally defective children, and in order to secure the means of comparison, carried out a most exten sive survey of normal individuals, including school children, uni versity students, and aged persons, to the number of over 6,000. The research was conducted in Australia, and as an interesting, but entirely accessory "control", about 6o aborigines were also measured. In reporting on the data, the authors admit that they can only consider the marked deviations from the normal as pos sessing diagnostic value, so effective are the sources of error. With this reservation (and their claims are not invalidated by the state ment made by Prof. Karl Pearson in Biometrika, vol. v. p. 1o5, as to the low degree of correlation between head-measurements and intelligence), they claim that the volume of the brain, calcu lated from the dimensions of the head as actually measured, dif fers significantly in the mentally defective from what it is in the normal groups of corresponding age used for purposes of control. Not only so, but besides these mentally defective children, the deaf and dumb children subjected to investigation, as well as the reformatory inmates and the aboriginal natives, all fail to present evidence of brains of the volume shown by the controls to be appropriate to comparable individuals of normal intelligence. The data illustrate these conclusions. The volume of the brain expressed in cubic centimetres, was estimated, and average values for several groups of boys or youths provided a sequence that may be represented as follows: A detail brought out rather strikingly in the preparation of the normal data for use is the definite transient phase of arrest which appears at least twice in the progress of brain-growth to its full amount. These periods of arrest affect the brain in boys and girls alike, and in the same years, namely from 11 to 12 and again from 13 to 14. In the girl a third resting stage is indicated more definite than either of those which precede it, and distinctive of the years from 16 to 18. Subsequently the female brain completes its growth to the extent of 2% of the whole amount. The calculation of the volume of the brain involved the use of a formula which had been worked by Dr. Lee under the supervision of Prof. Karl Pearson.

Dietary Investigations.

A memoir was published by Dr. Corry Mann in 1926 on the acquisition of "a numerical value for a basic diet, of poor quality, but adequate physiological value, in terms of nutrition, and to assess, similarly, the value of certain items of food when these are supplied as additional rations to such a basic diet." Anthropometric data (of stature and weight) were employed, for frequent records were made of weight and height. Boys of school age, with a low rate of sickness, living in an insti tution near London, were the subjects of the investigation. Sixty one boys who received only the basic diet gained an average of 3.85 lb. per boy and grew an average of 1.84in. per boy during 12 months. Forty-one boys who received in addition to the basic diet a ration of fresh cow's milk, pasteurized and homogenized, one pint, 388 calories daily, gained an average of 6.98 lb. per boy and grew an average of 2.63in. per boy during 12 months. The rela tive values of several diets, as measured by gains of height and weight, are exhibited graphically in figs. 3 and 4 of the reports. Besides illustrating the predominant effects of milk, these figures bring out a curious point in regard to the effects of (a) water cress and (b) casein when added to the basic diet. The effect on weight is very definite and indeed considerable when water-cress is introduced, but the effect on stature is negligible. (The actual numerical data would by themselves suggest that water-cress is deterrent to the attainment of the stature that would take place in its absence. But this conclusion should not be drawn without close scrutiny of the figures. The influence of a "random sampling" may be accountable, and the more probable conclusion is that the effect of water-cress is negligible in respect of stature.) Casein in these experiments seems to produce no significant effect on either weight or stature, despite the very marked addition it provided to the amount of protein in the diet (almost doubling it) and to the number of calories (increased from 61 to 91) .

The duration of the experiment, the number of individuals under observation, the number and minuteness of the analyses confer an exceptional importance upon Dr. Corry Mann's work. It affords guidance as to the fundamental requisites for such en quiries, including the necessity for exhaustive analyses of the diets employed.

Racial Characters.

In some of the preceding instances the element of "race" or "type" may evidently need consideration as a possible factor in determining differences detected by anthropo metry in the survey of a population. Anthropometric methods must therefore be applied to the study of the several divisions of mankind, and the wider aspects of the subject include the pre historic as well as the existing members of the Hominidae. More over, the scrutiny must be extended from mankind to the monkeys, and to any other animals whose general anatomical structure proclaims them nearly akin to these.

In a survey of what may be termed "Racial Anthropometry" the living subject and the absolute dimensions of the body evi dently claim the first place ; but from the first also regard must be had to the skeleton. The bones alone are available as the vehicle of information concerning prehistoric man; and the same is true of more than one recent type of humanity, e.g., the Tasmanians, the Mori-ori and the Caribs, who have become extinct within the last few centuries, or even within the memory of persons still living.

Anthropometric observations of living persons include measure ments of weight and of linear dimensions. The latter may be rectilinear, as in the case of stature, or curvilinear, and two sub varieties, viz., arcs and circumferences, are to be distinguished as coming within the latter description. Linear measurements of diameters or circumferences provide the means of calculating volume or capacity, and finally angular measurements consti tute a separate category which has representatives in nearly every scheme and schedule.

Range of Stature and Weight (Extreme Instances) The highest average weight , recorded in Dr. Martin's list for females is that of south Russian Jewesses, and amounts to 63.5 kgm. (14o lb.). The erratic character of such data is shown by the fact that these women provide a greater average number than do the male south Russian Jews. This paradox is probably due to the greater obesity of the females. Among all human types indi vidual variations are considerable, amounting among the Bush natives of South Africa to a range of from 3okgm. to 59kgm. in males and 3okgm. to 45kgm. in females.

Such examples exhibit an extraordinary range of diversity. Their study raises a fundamental question as to its significance, and that question still lacks a final answer. As to the local diver sity of human stature, no doubt exists, and it is easy to proclaim the racial distinction of the Scandinavian from the Laplander, even though their territories lie adjacent the one to the other. Yet the gap can be bridged, and the intermediate examples are so numerous and so diverse as to establish a complete gradation between the extremes.

But the question remains, is mankind a single stock with an al most infinite number of sub-varieties, or are a limited number of fundamentally distinct types now linked together? The bonds of linkage would be such factors as admixture of types (with inter mediate characters in the offspring of such matings), local condi tions affecting growth favourably or the reverse, and other factors as yet imperfectly known. Kollmann postulated the original co existence of certain distinct "types." But there is no demonstra tive proof on either side. Against such "original diversity" the supporters of an "original unity" of human ancestry can still proffer their arguments, reopening the old problem of polygeny versus monogeny. This is so closely connected with the mode and precise path of human evolution from the immediately pre-human stages that its temporary loss of vogue as a subject of debate is bound to be succeeded by a renewal of interest. (See Although stature is naturally related to body-weight, the cor relation of the two characters is not complete ; the taller of two individuals need not necessarily be the heavier. Thus among the south Russian Jews, the women weigh on the average more than the men, though the men are notably taller (1,651mm. as against 1,S36mm.). And again, the geographical distribution of those human beings who are at once among the tallest and heaviest of their kind is capricious. They preponderate in areas so widely separated as Scotland, Montenegro, Rajputana, Sumatra, the Marquesas islands and Patagonia, with African groups in the Bahr-el-Gazel and Zululand. The last-mentioned come, curiously enough, into close relation with a remnant of the pygmy Bush natives of South Africa. Again the tall men of Galloway are set off by the diminutive weavers of Spitalfields, the interval separat ing them being almost equal to that of the extremes provided by the whole range of humanity. But the range of pygmies is almost equally extensive and as certainly interrupted and discon tinuous. To the pygmies of the Kalahari desert in South Africa there succeed those of the Congo State; but other representatives of pygmy size are found in Lapland, the Andaman islands, the Malay peninsula, the ,Philippine islands and New Guinea. Lati tude and longitude alike seem devoid of relation to such apparently fortuitous dispersals. (See PYGMY, and the appropriate sections in the articles on the places named.) Average stature Average weight Zulus (men) . . . 172 (approx.) Bush natives (men) . 155 (approx.) 89 lb.

Galloway (men) .

. • . 179 (approx.) 173 lb.

Spitalfields (men) .

. . . i55 (approx.) no lb.

But even if it be admitted that the measure of stature is not a sure indicator of body-weight, yet the taller individuals seem on an average to possess heavier brains.

Prehistoric Man.

In the present connection, the stature and bulk of the prehistoric representatives of humanity claim atten tion. The evidence comes necessarily from the skeleton, and con sequently estimates of body-weight are very liable to error in determination. Nor are estimates of stature derived from the bones by any means free from this drawback. Where skeletons have been found intact and extended "at length," direct measure ments have been sometimes practicable as the specimen lay ex posed to observation, and before it was disturbed in any other way. Such a measurement will differ notably from the stature of the individual in life; yet it will evidently furnish a useful control of estimates based upon the measurements of separate bones and the employment of formulae devised for the purpose of recon structing stature from such measurements.

Calculation is the only method available when the remains are fragmentary. Imperfections in the limb-bones will necessarily be detrimental to an estimate of stature based on their dimensions, and certain corrections have to be made, e.g., according to the sex of the skeleton. Should the estimate have to be made from the femoral length, it seems an easy proceeding, in view of the state ment at one time current that the femoral length is 25% of the stature ; but this rule needs amendment if the length of the femur is markedly above or below the average. Correction will be needed if the bone is part of the skeleton of a negro and not of European BY COURTESY OF THE CAMBRIDGE PHILOSOPHICAL SOCIETY FIG. 3.-DIAGRAMMATIC REPRESENTATION OF INCREASE IN STATURE, Fig. 3.-DIAGRAMMATIC REPRESENTATION OF INCREASE IN STATURE, IN CENTIMETRES, FROM BIRTH TO MANHOOD, 19 YEARS The rate of growth is greatest in the first 61/2 years, when the stature is approximately doubled origin ; and until evidence is adduced to the contrary, the need for correction must be assumed if the skeleton is of prehistoric an tiquity and not modern. In the case of prehistoric Europeans, the amount of the addition or subtraction may, however, be assumed to be very moderate. Thus by the employment of suitable formulae, a fairly close estimate of stature can be made upon data provided by single bones.

To judge by such skeletons as have been recorded so far, the stature of prehistoric man was by no means uniform, and in gen eral it presents little difference from that of the modern inhabi tants of the same areas. Examples of the extremes of variation present themselves in the skeleton from Cro-Magnon, for despite the advanced age assigned to this individual, he is supposed to have measured 1,82omm. in height (Boule), while at the other end of the scale the skeleton from Chancelade of comparable antiquity denotes a stature of 1,S5omm. only (Boule). Comparison with the range of averages quoted above for modern men is not justi fiable, and yet the diversity of stature in prehistoric times is ren dered sufficiently evident. Both these examples (Cro-Magnon and Chancelade) represent mankind in the so-called Palaeolithic pe riod, and from the same period there are forthcoming other indi viduals so characterized as to fall between the two just men tioned in respect of stature. These data can be tabulated as follows : I. Les Eyzies. Old man of Cro-Magnon (Upper Palaeolithic) Boule's estimate of stature, 1,82omm.

Manouvrier's estimate of the same, 1,736mm. 2. Man of the Neanderthal (Middle Palaeolithic) Pearson's estimate of stature, 1,629mm.

3. Man of Chancelade (Contemporary with Cro-Magnon) Rahon's estimate of stature, 1,592mm.

Testut's estimate of stature, i,55omm.

In dimensions, at least, the man of the Palaeolithic period does not support the view that in prehistoric times primitive brutality was associated with gigantic bulk. Even beyond the confines of Europe, the prehistoric Rhodesian man with a stature of 1,83omm., tall though he may have been, does not notably exceed indi viduals among the existing inhabitants of the region. In China, again, the precursors of the modern inhabitants of Fengtien and Honan in the Aeneolithic period show no very outstanding dif ference in point cf stature from the existing inhabitants.

Body Ratios.

The stature of mankind is related to the body weight sufficiently closely to serve as a means of estimating the latter; and in view of the possibility of forming an estimate of the brain-weight from the dimensions of the skull, it appears that if there be available from a skeleton the skull and one of the thigh bones these remnants suffice for making these two estimates. (The thigh-bone can be used in estimating first the stature, and subse quently the body-weight of the individual.) The ratio of brain weight to body-weight has been termed the index of cephalization. Its value is approximate to - in normal healthy adult men. This may be taken as a distinctive human character with which to compare the ratio exhibited by an adult male gorilla, Estimates based on the dimensions of the appropriate parts of prehistoric human skeletons present no startling contrasts with such as modern humanity provides. An exceptionally interesting index is that derived from the skull-cap and the femur of the Javanese fossil Pithecanthropus erectus, for the value assigned to the index is in this instance 7 0, and it assigns to the fossil form a place intermediate between mankind and the higher apes. Thus the values of the ratio of brain-weight to body-weight are :—(r ) Normal adult man, 3-; (2) Pithecanthropus' erectus (Java), (3) Hylobates, small anthropoid ape, 1; (4) Gorilla, large anthropoid ape, The numerical value of the human ratio testifies to the abun dant provision of brain-material in mankind as compared even with closely allied mammals. But the ratio or index presents such a wide range of value in various animals as to require comment. Thus, for instance, the human figure 5 though well in advance of that for the gorilla, does not actually head the list. In this respect man falls behind many birds. At the other end of the scale come gigantic animals such as the hippopotamus, the whale and certain fishes. Some correction is thus seen to be necessary, and among the more obvious factors that of absolute size is evi dently important. The birds most favourably placed (index about are small, e.g., small finches. For a whale there has been cal culated an index represented by (Wilder). The same in fluence is manifestly active during the lifetime of the human individual, for as an infant he owns at birth the "superhuman" ratio of 6. Allowance for absolute dimensions must needs be made before any inference regarding cerebral endowment can be drawn from the figures; and in the above instances the human being, the Javanese Pithecanthropus and the gorilla are in this respect comparable, while the small anthropoid ape represents the general standard for such apes handicapped as generously as possi ble, by reason of the relatively diminutive size of this particular species. Even so, it fails to attain the human standard.

The "Canon" Error.

A fundamental error underlies many "systems of proportion," namely the assumption that a particular part of the body bears precise numerical relations to all the rest. Such an assumption is almost universal in artistic schemes, and is, in fact, implicit in the selection of many units or moduli, such as the finger, the foot, the cubit, the head or the vertebral column. That this matter should be disposed of once and for all is the more important since a definite "canon" is still commended as the basis of the comparison of one human type with another, an application wholly to be condemned. According to Duval, the works of the ancient sculptors present precisely the same varia tions as are met with in nature. In fact the rule was by no means obeyed by all or in every detail, and freedom from slavery to an artificial system resulted in the production of contrasted forms. Some of these are still in existence, so that contrasts can be drawn between such examples as the Running Fighter (Louvre) and the Antinous.

The anthropometric method of investigation recognizes no authority in the shape of a modulus or canon. The latter is the outcome of an assumption that remains unproved. In this respect, it invites comparison with the theoretical "archetype," a concept in morphological anatomy which enjoyed a transient vogue in the i8th and part of the i9th centuries.

Standards Required.

But while recognition of a pre-deter mined form of representation is refused, this does not imply re jection of all forms of representation. Anthropology willingly recognizes a standard of proportion which is the outcome of actual observations, and represents the mean of measurements made on many individuals with unvarying precision. Such a standard is indispensable when human beings are being compared with each other, or even when the human form is brought into account with those of the higher apes ; and in certain instances the standard may usefully take the rigid form of what mechanics colloquially term a "jig." Thus Dr. Percy Stocks (Annual Report of the Chief Medical Officer of the Board of Education, 1926, p. 105), reporting on "Goitre in the English School-child," recommends the employ ment of a "standard gauge" for thyroid-breadth. By its use, the number and proportionate frequency may be learned of those children possessed of a thyroid gland exceeding the standard breadth of 42mm.

Schematic Representation.

In the accompanying figures (I and 2) there will be observed indications of the respective lengths and widths of particular parts of the body. The stature of all is reduced to a common measure so that contrasts in the relative proportion of parts may be facilitated. For the construction of such diagrams, particular measurements, of which a list is ap pended, were made on a number of individuals. And from the collected data, certain averages or mean values have been derived.

Comparisons and contrasts need consideration in respect of the male and female (fig. 2) ; next in turn the immature stages call for consideration (fig. 3) ; and finally there are added schematic outlines of an anthropoid ape (fig. 4) depicted in the same way as the foregoing figures. The contrast between the newly born infant and the adult in respect of the proportionate weight of the brain has been the subject of comment in an earlier paragraph. Here the contrast affects the proportionate length of the limbs. The difference in respect of the relative lengths of the upper and the lower limb is a matter of common knowledge and is in the present instance capable of expression in a numerical manner. Ob servations on a large anthropoid ape (gorilla) have been used in the preparation of the schematic figure with which to compare the adult human form. But in this matter, again, contrasts are much more impressive than similarities. These contrasts include (a) the relation of the length of the torso to the stature as a whole, (b) that of the span of the outstretched arms to the stature, (c) that of the upper limb to the lower limb, and (d) that of the forearm to the upper arm. The differences are very strongly marked, and the other anthropoid apes would serve to illustrate them almost as well as the gorilla ; in certain particulars they would provide an even more decided contrast. Thus the apes present a very con siderable diversity and are themselves the subjects of a gradation in characters, the line of which in some details is directed towards and continuous with that of human types.

Growth.

Anthropometry is also the mainstay of many im portant investigations involving the study of the growing indi vidual, particularly the details of the normal process, since these provide the "control" necessarily employed and indispensable in researches upon the factors influencing development such as those mentioned above. Among the various paths of research there must be recognized that followed by artists such as Direr, whose work on the proportions of the human figure includes reference to infants and adolescents. Records have been kept by certain par ents of the dimensions (whether of stature or weight) of their own children, e.g., the data collected in the r 7th and i8th cen turies by M. Gueneau de Montbeillard, and the illustrious Comte de Buffon, who inaugurated the method which, after a long inter val, received a wider recognition and application. Quetelet, com bining the measurements quoted by Buffon with those made by himself, points out the sudden acceleration of growth manifest at the i 5th year in boys, and the corresponding acceleration in girls a year or two below that age. Without attempting to detect evidence of any other marked change during the period covered by the record, Quetelet repeats that the rate of progress is, in fact, subject to variation. He adds that physical development, like that of the moral side of human nature, advances in a series of bounds. But despite his recognition of occasional acceleration, he seems to have preferred to lay stress on a more general phenomenon, viz., the apparently constant retardation of velocity from an early period onwards. It was reserved for Dr. Godin to make further discoveries of the rule of periodic variation. He observed during a period of five years a large number of adolescents (from 13 to 18 years of age) . He measured i oo of these youths at intervals of six months during that period, and he claims that great importance attaches to the facts (a) that the same subjects were kept under observation throughout the years in question, and (b) that the measurements were repeated frequently. Of the abundant harvest of data gathered by Dr. Godin the figures relating to the average increments of stature of his at 14, i 5, i 6 and i7 years of age respectively will serve to illustrate one of the most important outcomes of the research. The data may thus be set out :— The sudden augmentation of the increment at the i 5th year is the point on which stress is laid. Quetelet's figures fail to exhibit this local increase, perhaps owing to the selection deliberately made by Quetelet of his subjects. However, in respect of rhythmic variation in the rate of growth, Dr. Godin's figures, supported by those of Carlier and others, represent the actual sequence of events more faithfully than do Quetelet's, and the details are won derfully illuminating. In the six months between 15 and 151 years, the lower limb contributes to the augmented growth in a char acteristic and distinctive fashion. Thus out of a total increment of 56, the lower limb provides 19 in the following proportion :— the thigh, 12 the leg, 6; and the foot (instep), i. To realize the peculiar and distinctive nature of such a combination, the incre ments for the other half-years must be considered, and the whole series as set out in the memoir may be given as follows : Age . . . . . 132-1-4. 142-15. Meas. thigh . 9 4 8 12 Leg. 2 9 0 6 Foot (instep) . . 3 0 3 I Minm. circum. of leg . 6 2 6 2 Age . . . . . 151—r6. 16-162. 161-17. 17-171.

Meas. thigh . 2 3 3 0 Leg. 4 to 3 Foot (instep) . . 2 0 2 0 Minm. circum. of leg . 5 2 3 0 Proportional Growth.—The essential feature (viz., the alter ation of the activity in thigh and leg respectively) is sufficiently evident. But at one epoch, viz., that from 15 to 151 years, the two segments for once combine forces and though the leg contributes relatively little, still a certain combination does exist, and the result is the very marked augmentation of stature just noted.

But the tabulation reveals more than this. The contribution of the arch of the foot is at this critical epoch at a very low level, though in other periods it is of appreciable amount. Thus in the lower limb alone contributions are made by at least three distinct "elements" and usually in a sequence involving alternate phases of activity and restfulness. Moreover, the alternatives are themselves alternate with those of the neighbouring segment of the limb.

Another point is that while a given segment, for instance the leg, is making a small or even negligible increase in length, such inactivity does not affect the growth of this part in its entirety. Evidently the limb may be increasing in some other direction at this period. Reference to the tabull ted data relating to the growth of the circumference of the leg will leave no doubt on this point, for the active periods of growth in length alternate with compa rable phases of activity in respect of the thickness of the limb. There is a suggestion of a similar relation in the records of length and circumference measured on the forearm. And the conclusion thus formed regarding the limb as a whole applies not improbably to the bones which form its framework.

A final result of the observation is the indication that at the age of 151 years (in the type of youths measured) the thigh has near ly completed its growth in length. Here the observation of Carus (1854), confirmed by Humphry in 1858, relating to the propor tionate length of the lower limb as a whole at the i 5th year, is recalled appropriately to mind.

The changing rate of growth, and the alternate activity in ad jacent parts of the body detected by Dr. Godin, constitute a periodicity requiring comment from another point of view. The sum total of such phenomena has been compared to those presented by certain chemical reactions, and having regard to the admitted va riation in the rate of progress of some reactions, and to the chemi cal analogies, if not the identity, of the metabolic processes at the base of animal growth, the comparison does not seem inappropri ate. In one particular comparison there are brought together (i ) The rate of growth of the individual, and (2) The rate of change in a chemical reaction of the kind in which that velocity attains its maximum when just halfway towards its completion. To bring the progress of zoological development into line with this, the rate of change should be shown to be most active when the animal under consideration has just attained one half of the bulk it will eventually assume. Dr. Robertson has published observations on such a comparison since 1911. In 1924 attention was redirected to his work by Dr. Cruickshank and Mr. Miller in their report to the Medical Research Council (Special Report Series No. 86) on the estimation of foetal age. The investigations are relevant to the last-mentioned subject, since if the periodicity of growth were fully known, estimates of foetal age based upon the data of weight and body-length could be made more closely approximate to the real value by the employment of more suitable corrections than are at present available. Consequently the subject has practical bearings in addition to its theoretical interest. Dr. Robertson found some evidence in support of the view identifying the phe nomena of growth with those of the particular kind of chemical action referred to above ; yet the agreement does not appear to amount to identity, for reference is made to several phases of activity in the progress of human development, alternations which do not necessarily appear in the chemical reaction. The whole of this side of the subject requires fuller investigation.

Roberts' Work.

Roberts drew attention to the stress laid by Quetelet upon the distinction of the "average" man from the "mean," i.e., the most frequent observation or event of the series. But he dissented from Quetelet's deliberate opinion that so small a number as 3o individuals suffice for the establishment of an an thropological type, for he disputed the premise of that conclusion, namely the belief that the human type is so uniform that a small sample will represent the whole mass of humanity. Roberts pro claimed the necessity for renewing enquiry into variations, and for studying their frequency, their several degrees, and their possible causes. He pointed out the fact overlooked by Quetelet and first observed by Bowditch (in 1877) that at the ages of 13 and 14 years, the young girls in Great Britain and in the United States exceed in both their height and weight the boys of the correspond ing ages; and he remarked upon the contrast between urban and rural populations recorded by Quetelet for Belgium, inasmuch as the greater stature assigned by Quetelet to the town-dwellers is the exact opposite of the conditions obtaining in Great Britain. He insisted upon the importance of collecting large numbers of obser vations for the establishment on a wide basis of the relative height and weight of individuals for each year of age. He called attention to the remarkable range of variety in respect of such a dimen sion as height when a large number of observations are available for comparison.

More recent researches distinguish two particular periods of growth-activity in respect of stature, namely from 5 to 7 years, and from 13 to 16 years respectively. With these two periods alternate others when circumferential growth is predominant. Hammond has published figures which indicate a comparable alter nation in the parts of other animals. The comparison of the growth-curves of Japanese and New Britain children and adoles cents with those of European children shows the earlier age at which the female (in the non-European races) first surpasses the male in stature, the corresponding recovery by the male being made at the 13th year by Japanese, while in New Britain that event is actually delayed to the commencement of the 17th year, i.e., beyond its arrival in Europeans.

Heredity and Head

Form.—Another group of investigations relates to facts of heredity, including illustrations provided by the proportions of the head. The well-known contrast between heads of rounded form (brachy-cephalic) and long heads (dolicho cephalic) is usually expressed numerically by the percentage pro portion of the width in terms of the length. This "cephalic index" gives a convenient means of comparing heads of contrasted pro portions, and in regard to its indications, its values in excess of 8o denote heads of definitely rounded form, while the long heads own values below 75. The rule or order of inheritance in respect of the parental characters has not been very extensively investigated as yet. It has been suggested that suitable posing of the infant in a cot may be a potent factor in determining the final proportions (round head if the child lies habitually on its back, and long head if it lies on the side). Without accepting this pronouncement, the possibility that such influences are by no means entirely negligible should be kept in mind. Direct evidence as to the comparison of parent with offspring in respect of head proportions distinguishes the researches of Puccioni, who shows that the contrast between the two parents is repeated, though with slightly diminished em phasis, in the children, the evidence of blending being very slight. Puccioni's family yields numerical values of the index as follows : Parents: father 78•1 mother 86.7 Children: elder 82•3 (io years) younger 76.6 (71 years) These figures are of interest from several points of view. For instance, both children are boys and the mother would seem to have transmitted characteristic head proportions to one but not to the other. The latter does not merely differ from his mother here in, but has accentuated the distinguishing head form of his father. In this connection, notice should be taken of the youth (71 years) of this individual and of the consequent possibility that maturity may effect changes, among which further accentuation of long headedness is very probable. Again, it appears that although in regard to the maternal index the offspring provide some evidence of regression, nevertheless even such evidence appears only when the mean value of the children's indices is recorded without refer ence to the increased long-headedness of the younger child, which constitutes the most striking feature of the record. Nor does any support emerge for the_ "melting-pot" theory of F. Boas (191I) , in which environmental factors hold a prominent place, since in the present instance the environment was unchanged. As to the Mendelian aspect of the inheritance of the form of the head, dis cussed by Frets (1925), on the basis of an investigation of over 3,00o persons, it was found that the cephalic index (expressive of the form of the head) is hereditary, though modifying factors affect this general tendency. Frets lays stress on the difference between large and small heads in respect of this tendency. For in stance, where heads are large, that of brachy-cephalic proportions is dominant over that of dolicho-cephalic proportions ; but among small heads that of dolicho-cephalic proportions is the dominant. In submitting two alternative schemes to account for the actual se quences disclosed by the records collected by him, Frets remarks that the dominance of brachy-cephaly is stronger in the female parent, whereas that of dolicho-cephaly is stronger in the male.

Work in the British Isles.

Very remarkable extensions of Roberts' work have been made by the committees appointed by the British Association for the Advancement of Science 83 ), by the London School Board (19o4), and by the London County Council (1909) . These extensions comprise the collection of measurements representing the height and weight of school children. Roberts' tables record measurements representative of the physique of individuals at each year of age from the 5th to the 23rd in succession, and thereafter discontinuous data carry them on to the loth year. The collection of information relating to the years subsequent to the "school age" necessitates recourse f rom schools to universities, the public services and institutions, or to factories. First among the university sources, Cambridge com menced contributions in 1833, when the average stature of students is recorded as 1,768mm., and the corresponding weight as about io stone (after making necessary corrections). About half a century later, the collection of data was revived at the in stance of Galton, and on the basis of about 2,000 observations, it appeared that the average stature was approximately 1,75omm.

The collection of statistical data made at Cambridge during the middle period mentioned above, was undertaken by a committee of the Philosophical Society of that university at a time when an anthropometric laboratory was opened at South Kensington under the supervision of Galton himself. The information collected at Cambridge included observations of a physiological kind such as those relating to eyesight and muscular and respiratory energy. Among the measurements were comprised those of the three prin cipal dimensions of the head, and in i888 the late Dr. Venn drew upon these in the preparation of a very striking comparison. Com bining the measurements in question so as to furnish a single numeral expressive of the size of the head, he enquired into the size thus presented by various groups of students distinguished not only by differences of age but by their proficiency in examina tions. Over ff,000 individuals were available for the purposes of the enquiry, and the striking feature of the comparison may be stated as follows. The scholars, grouped with men distinguished in examination tests, revealed at the commencement of their resi dence heads with dimensions distinctly greater than those of their associates. The latter nevertheless acquired during residence a remarkable compensation for this initial handicap ; for while the heads of all increased in size during the period in question, the smaller heads grew so much more vigorously than those of the first group that the original margin of superiority was reduced by at least one half.

Valuable work of a corresponding nature is being carried on by Dr. A. H. Mumford, Medical Officer of Manchester Grammar school.

Work in America.

In the University of Pennsylvania, the director of the department of physical education sets an example of systematic thoroughness. In the preparation of records, the needs of the individual are kept in mind, in view of the contin gency that he may seek or need advice regarding the state of his health, or the selection of a particular form of athletic recreation. But while the student may receive benefits, he also incurs liability under that system; and the status accorded to it by the authorities is summed up in the director's remark that "It is quite possible for a student to have his degree withheld because he has not ful filled his requirements in physical education." Measurements of Recruits and Prisoners of War.—In the combatant services of States, recruiting necessarily involves the practice of anthropometry, if only for the purpose of securing uniformity, or of determining graduation or rejection. Conscrip tion makes available a large amount of material in the form of youths called up for service year by year. Such material has been studied in various countries, but the World War naturally led to a vast extension of recruiting and conscription. In Great Britain it caused the anthropometric study of an enormous number of individuals ; and by reason of the prolongation of the war, the scrutiny affected an ever-increasing range of years. The Report to the Ministry of National Service, vol. i. (1920) surveys the results up to 1919. The "Grades" into which the subjects were divided are well known, as is also the revelation of a large proportion of individuals of poor physique in certain districts. The actual obser vations include measurements of height, weight and chest-circum ference; and while the results* show the need for remedial action, it does not appear that the state of affairs indicates a recent or rapid deterioration (Shrubsall, 1924). Considerations based on the anthropometry of Cambridge undergraduates played an im portant part in determining the principles actually employed in graduating the individuals according to their physique.

A remarkable side-issue of the war was the anthropometric ex amination of prisoners. An example had been set already by Japan, where Prof. Koganei published in 1903 the results of his anthropometric study of Chinese soldiers, taken prisoner in the war against that country. In Great Britain Prof. Parsons reported on some of the captured Germans, while in Austria a large and systematic investigation was made of Russian prisoners, who rep resented a considerable number of ethnic types. Since the cessa tion of hostilities in 1918, the extensive anthropometry of which they were the occasion has likewise come to an end in some coun tries. On the Continent conscription persists, and in Switzerland the survey of recruits has proceeded without interruption. Judg ing by recent reports, the survey is becoming less mechanical.

Interest increases progressively in the actual history of the recruit during his engagement. The employment of particular measures or combinations of measures expressive of the general condition of each individual is intimated by the most frequent references to "indices" of fitness, such as Pignet's. And even though it be neces sary to admit the shortcomings of some of the coefficients pro posed, the very fact that they owe their origin to an attempt to increase the significance of the recorded data should be taken into account and set against those weaknesses. But not all coefficients are open to such adverse criticism, and alternative methods of employing the data may yield valuable results. Thus the methods devised by Prof. Dreyer for assessing physical fitness seem to be proving their value, and receive confirmation of their reliability as the area of their employment extends.

Gaol Population.

The inmates of prisons provide another source of information regarding physical development, and from this Quetelet drew some of the statistics used by him in 1835. At a later date, Bertillon advised and used anthropometric methods for the identification of habitual offenders and, indeed, of any per son under arrest. Such methods together with the descriptive rec ords of the complexion, hair colour and eye colour are accessory to the practice of dactyloscopy or study of fingerprints.

The instruments of anthropometry include measuring-rods of different sizes. Of these the smaller are commonly fitted with arms, and are called calipers. A flexible measure (for circumfer ential measurements) and a weighing-machine complete the out fit for the greater number of the records usually made. But spe cial instruments are required for certain measurements of the head, as also for the estimation of muscular power, and again for the measurement of "lung-capacity." Descriptions of the various instruments, with illustrations and directions as to their use, appear in various manuals among which may be cited the Report of the British Association Committee for 1008, or again, Martin's AntlLropometrie (1925) or Richtlinien fiir Korpermessungen by the same author (1924).

The list of measurements of admitted utility has increased con siderably since the schedule included the primary measures of stat ure and weight. The latter still retain their value and, in fact, constitute the sole basis of many schemes still in progress, but the additions are numerous. The British Association Committee of 1875 dealt with four measurements properly so-called, viz., height, weight, girth of chest and span of arms; but to these data observa tions on the colour of hair and eyes, visual acuity, heating capacity and strength of arm were added. In 1888 Dr. Venn reported obser vations on Cambridge students, with data from a list comprising stature and weight, together with three head-measurements, and others relating to visual acuity, breathing capacity and strength of arm. In 1902 a committee appointed by the British Association (in succession to that of 1875) to advise on anthropometric inves tigation in the British Isles, opened enquiries which led in 1908 to the publication of a report approving of 8o specified measure ments. This report includes sections on physiological and psycho logical measurements as well as the methods of recording photo graphic data. A recent critic (1924) describes the report as crude, and in certain respects it is evidently capable of improvement. Daring the period of its preparation, efforts were made to secure some degree of international consistency in regard to these mat ters, and two reports, published in succession, the first in 1906 and the second in 1912, exhibit the extent to which agreement had then progressed. These reports are not in entire agreement with the British Association's report of 1908, though the amount of diver gence is not great. The international agreement itself might be again revised with advantage to all concerned.

The last few years have seen the birth and growth of an entirely novel extension of anthropometry, viz., to a matter previously physiological in nature, termed comparative serology. As the name suggests, the object of comparative study is the blood-serum. Moreover, the practical application involves the pitting of one serum against another, and the outcome of the encounter con fers distinct "grades" on the competitors. Tests carried on in large numbers in various localities (mostly in Central Europe) point to variations in frequency of the possessors of different grades of blood-serum, and the confirmation of these results will be very welcome.

Anthropometrie (187o) ; Roberts, An thropometry (1878) ; Galton, Natural Inheritance (1889) ; Karl Pearson, The Chances of Death (1897), Biometrika—passim (to date) ; Godin, Recherches sur la croissance des diverses parties du corps (19o3) ; Greenwood, Health and Physique of School-Children (1907) ; Martin, Lehrbuch der Anthropologie (1914) ; M. Boule, Les hommes fossiles (1923) ; Davenport, Carnegie Institution of Wash ington, Publication No. 329; No. 35 of the Dept. of Genetics; Body build and its Inheritance (1923) ; H. H. Wilder, Pedigree of the Human Race (1925) ; Frets, The Cephalic Index and its Heredity (1925) ; Shrubsall, British Association for the Advancement of Science —Presidential Address to Section H. (1924) ; H. H. Wilder, Laboratory Manual of Anthropometry (1921) ; A. Hrdlicka, Anthropometry (192o) ; Martin, Anthropometrie, from the Handbuch der Sozialen Hygiene and Gesundheitsfursorge (1925) ; Sir G. Newman, Annual Report of the Chief Medical Officer of the Board of Education (Eng land and Wales) for 1926 (1927) ; G. Keynes, Blood Transfusion (1922) ; A. A. Mumford, Healthy Growth (1927). (W. L. H. D.)