The American Indian/Chapter 18

From Wikisource
Jump to navigation Jump to search
1383063The American Indian — Chapter XVIIIClark Wissler

CHAPTER XVIII


SOMATIC CLASSIFICATION

We come now to the man of the New World himself, instead of his works. While it may be more logical to begin here, the discussions under this head also call for a general understanding of the geography and relative distributions of all other characters. Further, this study of man presents certain technical difficulties which make it unsuitable for the first introduction to our subject. The fundamental difficulty has been to find a definite and consistent basis of classification. For example, the system developed by systematic zoology is not strictly applicable, because the existing peoples of the earth are very near to each other. Our experience with zoological classification leads us to expect that when we offer a morphological grouping of mankind, we shall base it upon descent. While this may be asking too much of a mere classification, we must not overlook the fact that here is the real problem in the case. Thus, in the New World, we must eventually know from what kind of men the original stock sprang and what elements, if any, have been later assimilated by it.

The most obvious method of approach to this solution is by morphological analysis of the main geographical groups. First, if we take the New World as a whole, certain common characters may be taken as significant. For one thing, the hair of the New World tends to be straight and black; indeed, the variations from this are so rare that we may consider straight black hair as universal. The only other part of the world where this is a distinguishing character is in Asia, particularly among Mongoloid peoples. A close view of the hair of mankind, as a whole, indicates great stability as to type, in consequence of which many anthropologists make it the basis of the first classification. We may, therefore, set down the above affinity with Mongoloid peoples as one very strong indication of common descent.

Louckeux Man and Woman showing Eskimo Characters
Boas, 1901. I

Bureau American Ethnology
A Pawnee Indian Geronimo, Apache
Fig. 89. North American Types
Photograph by Ehrenreich
Fig. 90. Brazilian Types

Fig. 91. Patagonian Types

A few exceptions to straight hair have been found in South America under conditions that makes it unlikely that they are of European origin.[1] However, a great deal more field-work will be necessary before this point can be made clear. If, however, it proves out that an element of wavy hair once entered the New World population, we shall have good ground for suspecting a non-Asiatic origin for at least one New World strain.

Skin color is rather an elusive matter, since its gradations do not admit of very precise definition. Some anthropologists see the basic color of the New World as yellow, others as brown. The yellow is clearly present in many tribes of Brazil and on the west coast of North America, but the remaining portions of both continents show populations ranging from dark chocolate to light brown. According to our own observation, this light color turns toward yellow, and the assumption of an original yellow race is fully justified. This, again, suggests Asiatic affinities, but just what may be the history of this dark strain in the yellow is not clear.

The nose has also been considered as Mongoloid, but as it presents great variety of form in both the North and the South and is not easily distinguished from the nose of the Pacific Islands and some other parts of the Old World, no great stress can be laid upon it, at least, until very carefully studied. Again, one of the most striking facial characters of Mongolian peoples is the eye-fold, or, in popular language, the "slanting eye." A number of observers claim to see faint traces of this in the Indian, but we should proceed with caution where the resemblances are so vague. Yet, a recent author[2] asserts its positive identification in the Andean region and also in parts of the Amazon country. In North America, it is prominent among the Eskimo and appears in Siberia, which fact gives us continuity with Asia.


BREADTH OF FACE

In popular belief, the aboriginal face is broad in respect to the width of the head, or "the cheek bones are prominent," resulting in what is sometimes characterized as a disharmonic face. This belief is, in the main, justified, when we review the measurements that have been made. The studies of Boas[3] and Jenks[4] upon mixed-blood Indians in the United States, show that the relative width of face among these is still greater than in case of whites, while reference to the accompanying table shows a very positive difference between the whites and full-blood Indians, in both North and South America. Thus, it is generally true that the New World peoples are characterized by broad faces. This feature is strongly accented in the Eskimo, but not sufficiently to place them in a class by themselves. Of Old World peoples, the broad faces are chiefly among Asiatics.

VALUES OF FACE-BREADTH HEAD-BREADTH INDEX FOR MALES

(Compiled from Jenks[5])

Nationality Number Index
Scotch 50 90.34
French 100 90.85
Mixed-Indian 60 94.68
Mixed-Indian 8 92.14
Mixed-Indian 12 92.52
Mixed-Peruvian 24 90.40
Ojibway 24 97.19
Mexican 2300 94.90
Apache 148 94.30
Pima 51 98.63
Peruvian 124 95.30


FLATTENING OF THE FEMUR

Another anatomical trait supposed to characterize native Americans is an excessive flattening of the femur, or platymeria. A résumé of the literature, however, leaves one in doubt that any such variation can be definitely assigned to the New World as a whole, though some reports of Hrdlicka[6] suggest that Indians differ from whites and negroes. The same can be stated for flattening of the tibia, or platycnemia. If the theory of Manouvrier that the phenomenon is dependent upon the muscular development of the individual, is accepted, then these flattenings of the shafts have no racial significance. (For a brief résumé, see Duckworth.[7])

Another curious femoral variation is the presence of a third trochanter, which Duckworth believes is, to a certain extent, correlated with platymeria. The most convenient frequency table so far compiled is that of Deniker,[8] which shows a range from 13 per cent, for prehistoric Europe to 64 per cent, for the Fuegians of the New World. The great frequency of this trochanter among the Fuegians raises a suspicion as to the remainder of the New World, but satisfactory published data are wanting.

This about exhausts the list of widely diffused common characters for the New World as a whole. As we have seen, the somatic line-up is with the Mongoloid peoples, and, while we are facing this way, attention may well be called to other similarities asserted by certain observers. Posnansky[9] reports the Mongolian spot in the Andean area, and an anatomical peculiarity of the cranium in the maxillar or region of the processus frontalis which is absent in the European, but prominent in Mongolian crania. It seems best, however, to defer further discussion of this subject until we have considered the differentiation of internal New World types and their distributions.


HEAD FORM

Head form has received so much attention that it almost monopolizes the subject matter of physical anthropology. As to what extent this is justifiable, remains to be seen. The special literature for the New World contains a large mass of measurements, both on crania and heads of living subjects, from which we can form some idea of their classificatory value. The most conventional of these measurements is the cephalic index, or the maximum breadth of head divided by the maximum length and the quotient expressed in per cent. Ripley has plotted the distribution of this index on a map of the world,[10] but an inspection of this suggests that the New World cannot be sharply differentiated on the basis of the cephalic index alone. It is true, however, that the broadest heads are here accredited to Asia and the longest have minor

Fig. 92. The Cephalic Index


representation in the New World, which would give our hemisphere a kind of intermediate position. Yet, we should not give much weight to the specific boundaries indicated upon such a map because the cephalic index numerals have been arbitrarily grouped. A better method would be to plot according to each numerical unit of difference. To do this successfully, we must use a large map and enter in its proper place the actual cephalic index for each group of people. A condensed form of such a map for the New World is presented here (Fig. 92) *

* The cephalic indices appearing on this map are based upon skull measurements, but contain also those reported for living subjects. The special works upon craniometry give the accepted methods for reducing the latter to units of the former. The indices are from the published notes of all observers; hence, some allowance must always be made for errors of observation, though in the long run these should cancel out.

From this it appears that we have the most data for the United States, but, in general, all the larger parts of both continents are represented in some fashion. Notwithstanding such discrepancies, the data as given should approximately outline the general distribution of the cephalic index.

Perhaps the first impression we get from this map is one of great range and variability. The scattered index values may even suggest a random distribution. Yet, it appears that the lowest values tend to cluster around certain points, as the Gulf of St. Lawrence, southeastern Brazil, and southern Patagonia, while the highest seem to mass on the Pacific side of both continents. On the other hand, if we regard the numerical range of index units, it appears that the total New World series falls but little short of that for the Old World. In fact, both the lowest and highest cephalic indices recorded by Martin[11] are from the New World: Pericues (Lower California), 66.1; Californians, 89.7. The indices used in our map range as follows:—

71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87
1 2 5 1 5 3 9 14 10 8 8 3 6 11 2 2 2

Such lack of distinction between the two hemispheres in contrast to what we found true of other characters leads one to suspect that head form is exceedingly variable.

In this connection, it may be noted that investigations among the immigrant population of the United States, and also among the inhabitants of Porto Rico[12] reveal a rapid change in cephalic index. This suggests that there are some powerful environmental and physiological factors in the cause complex that produces head form. As compared with heredity, these are minor factors, but still strong enough to produce modifications of several numerical units in the calculated indices. Certain empirical observations led Boas to assume a change of five to six units possible in a few generations. Unfortunately, all these observations have been upon Europeans, but it seems a fair assumption that if the principle is found to hold for one, it will also hold for the other.

In the earlier parts of this chapter we found evidences of homogeneity, and though there is a perplexing range of variation in head form, the question still remains as to whether these fluctuations are not mere pulsations around a single norm. If such they prove to be, then we have homogeneity in head form as well as in other characters. This is a difficult problem, but some approach to its solution can be made by statistical methods.

First, on a priori grounds we have a right to expect that, if more than one type of cephalic index exists here in the New World, the observed indices for the successive groups of natives will tend to cluster around separate nodes, or norms, when we treat them as a series. To apply this test, we have collected all the observations that the literature of the subject brought to our notice and combined them into a single series. For precision, we have tabulated the indices for skulls and living subjects separately, and for further check upon the result we have tabulated absolute skull measurements for length and breadth of head (p. 305).

Close inspection of these distributions suggests that the two for the cephalic index will be found to approximate the normal, or symmetrical type, found in homogeneous groups. Also, the series for absolute length and breadth of head show fairly symmetrical distributions. From all this one may suspect that there is, after all, but one type of head form for the New World.

Fig. 93. Two Skulls from the Arctic Coast of North America and Two from Tierra Del Fuego, two views of each: The first and second pairs are Eskimo; the third, from a village site on Grande Island (150 miles N.W. of Cape Horn), and the fourth Yahgan, Furlong Collection

DISTRIBUTION OF GROUP AVERAGES FOR NEW WORLD NATIVES

Skull Index Head Index Skull Length Skull Breadth
68 1 68 150-2 2 125-7 1
69 0 69 153-5 2 128-30 5
70 1 70 156-8 1 131-3 7
71 2 71 1 159-61 7 134-6 13
72 6 72 0 163-4 5 137-9 30
73 10 73 0 165-7 11 140-2 27
74 7 74 2 168-70 9 143-5 21
75 10 75 0 171-3 13 146-8 11
76 8 76 2 174-6 22 149-51 6
77 15 77 3 177-9 13 152-4 5
78 13 78 3 180-2 18 155-7 7
79 11 79 7 183-5 17 158-60 2
80 12 80 15 186-8 7 161-3 2
81 14 81 14 189-91 5 164 1
82 4 82 14 192-3 1 ----
83 4 83 9 ---- 138
84 11 84 15 133
85 3 85 7
86 5 86 3
87 5 87 1
88 5 88 1
89 2 89 3
---- ----
149 100
Av. 174.8±9.15 Av. 141.3±7.00

It should be noted that we are here treating the successive social groups like individuals. Thus, if we wish to arrive at the head form for a social group, we take individuals at random and calculate from these observations the most probable average value. In exactly the same manner, we may take these values for groups in the New World and place them in a series. When we do this, we again find a typical distribution suggesting homogeneity. However, as it stands, this result should be taken only as a suggestion, for there are many complicating factors to be considered. The whole problem awaits further investigation, and so cannot be profitably considered here. That the general assumption we have made is consistent with other somatic data is obvious, for if the people of the New World spring from a single stock then the observed deviations of head form are but the normal variations of a biological character.

In a general discussion of the whole subject, Boas[12] states that as the population expanded over the New World, it scattered out into more or less isolated local groups, whose inbreeding soon differentiated varieties of head form and other features. In other words, these were the natural random fluctuations around the fundamental type and are probably not permanent characters, all of which is consistent with their erratic geographic distribution. It may be, therefore, that the longer-headed Algonkins and Patagonians are merely the result of greater marginal isolation rather than survivors of a previous long-headed population.

At the outset, we stated that the New World could make some claim to an intermediate position in head form. If we take the averages of the lengths and breadths of head tabulated on p. 305, we find a length of 175 cm. and a breadth of 141 cm. Martin[13] places the extremes of the world at 143–225 for length, and 101–173 for breadth, from which it appears that our natives are grouped around the middle values. Further, since the extremes in our tables fall short of those for the world as a whole, we are justified in the conclusion that in absolute dimensions of head the New World is truly intermediate, but that in respect to head form as expressed by the cephalic index, approximately the whole known range is found.

The investigations of the Australian school of anthropologists have demonstrated the great comparative significance of the height of the head.[14] We may suspect, therefore, that the native of the New World will show some distinctions in this character, but, as is often the case, sufficient data are not available for a satisfactory conclusion. Martin[15] gives the range of absolute height of skull as 125 to 143 mm. and cites certain Californians as 129, Dakota as 131, and Eskimo as 135. This is not very promising as a definitive character, since we have about the whole range in the New World alone. Further, when we consider the distribution of the recorded values they seem to occur at random over both continents and when all are treated as components of a single series, we get the normal frequency curve. The suggestion is, therefore, that we have in the height of the skull for the New World a character that shows considerable range of variation, but which, on the other hand, is of one distinct type. However, at present this must remain a mere suggestion.

ORBITS AND NASAL SKELETON

A great deal of attention has been given to the form of the eye sockets or orbits, and the nasal skeleton. The former takes a quadrilateral form whose varying breadth and height can be measured. The relation of the latter to the former is expressed as the orbital index. The tabulation of such data as we found in the literature presents a symmetrical distribution with an approximate node at 87. As in the case of other characters we find the orbits of each local group to vary, but all to cluster about a single norm. Since there are several ways of measuring the orbit and the several authors are not often explicit as to the method employed, the above result should be regarded as tentative. Both Deniker[16] and Martin[17] have compiled standard lists of this index, which furnish the additional series in our table. The series we have for the New World is of the symmetrical type which gives us greater confidence in its mean value. Martin[18] regards the Fuegian as in the lowest class, with the Tasmanian and Australian, while the Patagonian and Indians in general have very high indices, as also do the Eskimo and Mongolians. It appears, therefore, that the New World as a whole falls in the same class as do Asiatics, but that, on the other hand, high values are not unusual elsewhere. Thus, we cannot be sure that the orbit is a definitive character.

The skeletal structure of the nose affords further opportunity for measurement. The reader interested in the technical details of the subject may consult the very concise and convenient statement of Deniker,[19] together with a tabulation of nasal indices for representative groups from all parts of the world. This index is analogous to that for the orbit, expressing the relation between the height of the nasal bones and their width. From the recorded indices, it appears that the white race is at one end of the series and the negro at the other.[20] Just where our natives will fall is not yet clear, but in last analysis our conclusion must rest upon qualitative observations, and it is the opinion of our leading anthropologists that there is a recognizable form for the nasal opening of New World peoples as a whole.

DISTRIBUTION TABLE FOR THE ORBITAL INDEX

The New World All Races
Deniker[21] Martin[22]
76 1
77 1
78 0
79 1 2
80 2 0
81 0 3 1
82 1 0 2
83 3 2 3
84 3 6 5
85 2 4 4
86 3 4 6
87 9 4 1
88 6 4 7
89 6 6 4
90 4 3 2
91 2 3 4
92 1 6 1
93 2
94 0
95 1
96 0
97 1

In conclusion, then, we find in the New World face and head many indications of somatic homogeneity strongly suggesting unity of origin. A more positive statement, while greatly to be desired, must await new searching investigation, for the problem is peculiarly intricate and the available data, so far, inadequate.

Fig.94. Types from Central Brazil: Bakairi and Tumayana
Von den Steinen, 1897. I

BODILY PROPORTIONS

Among the many other aspects of somatological description that promise well, is bodily proportion. This is usually expressed in diagrammatic form, as in Fig. 95. Though there are many published measurements of New World natives, in few cases do they give sufficient data for the construction of such diagrams. In fact, the four we have here about exhaust the subject. Yet, they show a striking similarity in form and proportions, suggesting that the New World native is of one bodily type.

This method of comparison seems to have been devised by Thomson[23] from whom three of the diagrams in Fig. 96 are taken. If these are accurately constructed, there are obvious differences between the Indian and other races. However, our present purpose is to show that even in bodily form there is evidence of New World unity.


SUMMARY OF SOMATIC CHARACTERS

In this brief review of a large subject, we have sought, first of all, the characters truly definitive of New World man. The difficulties besetting such a quest are now apparent, but a summary of some kind is necessary to progress. Accordingly, on the basis of the preceding and the available studies, the following is offered:—

1. The hair is straight, of medium coarseness, and black. Body hair scant. The skin color is a brown, ranging from yellowish to chocolate tones. Eyes tend to be dark brown.

2. The head is quite variable as to length and breadth, but appears to approximate an intermediate position in the world's series. On the other hand, the face is broad in proportion to the head. As to facial angle, there is moderate prognathism, and this, as well as size of mouth, thickness of lips, proportions of nose, size of teeth, takes an intermediate rank between whites and negroes. According to Hrdlicka, the upper incisors are markedly concave.[24] Finally, there is a slight slant to the eye, reminding one of certain Asiatics.

3. In bodily proportions we note first an intermediate position between whites and negroes with respect to length of arms and legs. The hands and feet are relatively small. The capacity of head and estimated brain weights also give intermediate values.

Fig. 95. Diagrammatic Representation of Average Bodily Forms for the Eskimo (Duckworth, 1900. I), the Quechua (Ferris, 1916. I), the Bakairi (Ehrenreich, 1897. I), and the Yahgan (Deniker, 1900. I)


RELATIONS TO MANKIND IN GENERAL

Assuming for the time, at least, the single origin of all New-World peoples, we may now turn to the question in which all are interested, viz., the relation of the Indian to mankind in general. As has often been stated, the affinities of New World man are with Mongolians and, to a less marked degree, with Polynesians. With the former we have close parallels in hair, form of eye, breadth of face, and bodily proportions. With the Polynesians, the agreements are chiefly in pigmentation and to some extent, in the hair. Hrdlicka[25] has formulated a convenient statement of the problem, which may serve as our point of departure:—

Fig. 96. Bodily Forms from Various Races—Negroes, Australians, and United States Soldiers, from Thompson, 1880. I; Amherst Students, from Hitchcock, 1888. I. There is some doubt as to the accuracy of the United States Army data, but the variation shown here serves to emphasize the homogeneity of the New World types


The conclusions, therefore, are: the American natives represent in the main a single stem or strain of people, one homotype; this stem is identical with that of the yellow-brown races of Asia and Polynesia; and the main immigration of the Americans has taken place, in the main, at least, gradually and by the northwestern route in the earlier part of the recent period, after man had reached a relatively high stage of physical development and multiple secondary differentiations. The immigration, in all probability, was a dribbling and prolonged overflow, likely due to pressure from behind, or want, and a search for better hunting and fishing grounds in the direction where no resistance of man as yet existed. This was followed by multiplication, spread, and numerous minor differentiations of language due to isolation and other natural conditions, and by the development, on the basis of what was transported, of more or less localized American cultures. It is also probable that the western coast of America, within the last 2,000 years, was on more than one occasion reached by small parties of Polynesians, and that the eastern coast was similarly reached by small groups of whites, and that such parties may have locally influenced the culture of the Americans; but such accretions have nowhere, as far as we know to-day, modified the native population.

Such conclusions must, in last analysis, rest upon a satisfactory classification of mankind as a whole, but as stated at the outset, this has proved a difficult problem, and the reader will find the literature of the subject very perplexing, each investigator proposing a different scheme. Upon analysis, however, we find these schemes have a great deal in common and, hence, are fairly intelligible. In the first place, they fall into two groups according to the controlling concepts. One of these classifications is purely objective and empirical, no regard being paid to ancestral relations or other related factors, while in the other, the ideal is to form a classificatory scheme that will express genetic relationships. The exposition of these systems does not fall within the scope of this work, but a few of their main features may be noted.

Taking men as they come, experience shows that the most definitive objective characters are hair, skin color, and head form. Of these, the first is most persistently transmitted from parent to offspring, and so tends to remain constant. In the main, three kinds of hair are recognized: straight, wavy, and woolly, each having a distinctive cross-section and associated peculiarities.[26] With these as the points of departure, the people of the world fall into three groups:—

1. Straight hair (leiotrichi). The Asian-American group.

2. Wavy hair (cymotrichi). The Polynesian-European group.

3. Woolly hair (ulotrichi). The Australian-African group.

It is important to note that, though this grouping is strictly based upon the hair, the majority of all classifications recognize these three great clusters of mankind, their differences arising from transferences of the more doubtful examples from one class to the other. A convenient summary of the main characters for each is offered by Giddings:[27]

  I. The Australian-African Group.

Characteristics: black skin, dolichocephalic (long-headed), prognathic, woolly or frizzly-haired (cross-section of hair very elliptical).
Area of distribution: Australia and Africa south of the equator.

 II. The Polynesian-European Group.

Characteristics: fair skin, mesocephalic, orthognathic, straight or wavy hair (cross-section slightly elliptical).
Area of distribution: broad zone from Polynesia northwestward through southwestern Asia and northern Africa and most of the continent of Europe.

III. The Asian-American Group.

Characteristics: yellow or red skin, brachycephalic (broad-headed), narrow-eyed, lank or straight-haired (cylindrical in cross-section).
Area of distribution: eastern Asia and western America, chiefly north of the equator along the semicircular shore-line of Asia and America.[28]

This investigator regards the second or middle group as the main stem, the generalized ancestral stock, from which the other two diverged and specialized. The American aborigines are regarded as a diverging branch of the Asiatic group. This classification, therefore, assumes a genetic relation between its main division by selecting the Polynesian-European type as the generalized, or ancestral type. Unfortunately, in this case, as in most others, there is at hand no convincing proof. The grouping by hair and associated characters, has proved of great convenience and is in-so-far justified; but it cannot meet the needs of the case. Recently, Duckworth[29] offered a morphological analysis of man upon the basis of which was proposed a different grouping. In this scheme, the generalized type is designated as the Eurasiatic, embracing almost the entire population of Europe, Asia, and the New World. It will be observed that in distribution, this type covers the greater part of the earth in one continuous mass. What remains are the minor outlying groups, as the Australians, Africans, Andamanese, Polynesians, etc., which are treated as diverging stems from the main body. The one new feature of this grouping is that it correlates morphological distinctions with the facts of distribution, the significance of which we shall consider shortly. On the other hand, the morphological grouping is based upon the same principles of classification as are employed in zoology. Hence, Duckworth's scheme is worthy of the most careful consideration. The vital question is, however, the reconstruction of the generalized ancestral type. This is a broad problem whose solution will rest largely on general paleontological conceptions and the specific study of the primates. Its discussion falls outside the limits of the chapter, but a brief glance into the special literature of the subject will show the strong support that can be found for this idea. We may, therefore, adopt as our working hypothesis the view that the Eurasiatic type, to which the bulk of the native population of the New World belongs, is the best living representative of the general type hominidæ, and that the remaining groups of mankind are such as by wide dispersion and greater isolation became more divergent, or specialized. This idea is also consistent with culture data, for it is exclusively among this generalized main body that the great civilizations of the world arose; the highly specialized outlying groups, though showing sparks of true genius now and then, never rise to such a level.

The next point to consider is the distribution of mankind over the earth. In the synthetic treatment of such a large problem, we need the greatest possible array of corrective data. While such data cannot entirely mitigate the weakness of all interpretations based upon incomplete series of observations, they, at least, serve as friendly guide-posts. If, instead of focusing our attention upon man alone, we take in the whole gamut of mammalian dispersion, we greatly increase the number of these corrective aids. Thus, a brief perusal of general books on mammalian life[30] suggests at once the existence of a veritable swarming center in the heart of Asia. It is not, of course, contended that all mammals arose there, but that a surprisingly large number of the most distinctive families can be successfully localized within the general limits of that continent. This is supported by specific data when we restrict our

Fig. 97. Lines of Dispersion for the Primates. Matthew, 1915. I

Fig. 98. Phylogenetic Relations of the Living and Extinct Groups of Primates. The circles indicate the size and known geological range of the several groups, the dotted lines their most probable derivation. Their supposed relations to certain Insectivora and intermediate extinct groups are also indicated. Matthew, 1915. I

view to the primates alone, respecting whose Asiatic origin there is no dissenting voice among those competent to speak. The most convenient presentation of the subject is in Matthew's paper on "Climate and Evolution," whose sketch map we reproduce here, together with the chronological scale essential to its interpretation.[31] From this we see, in a general way, how the many diverse forms of primates have swarmed out of the Old World, one after the other. We have, therefore, no alternative but to conclude that the same factors controlling the dispersion of the whole division of primates apply also to man, and that his dispersion will, in the main, follow the same beaten tracks. With this as a point of departure, it is not very difficult to form a satisfactory statement of the order and direction of dispersion for the known species and varieties of hominidæ.

When we consider the map in Fig. 97, it appears that the New World is geographically more remote than either Africa, Europe, or the insular area, and has but one point of contact, but that point is in direct line with the assumed center of dispersion. On mere geographical grounds, we should locate the extremes of expansion in Patagonia, Greenland, Cape Colony, Ceylon, and Tasmania. To this list may also be added the extreme western parts of Europe, as the Canaries and British Isles. These are veritable blind alleys in which primitive man was ultimately brought up short and where some of his descendants are still marking time. It so happens, however, that the fluctuating polar ice cap hovers over the single narrow bridge to the New World and the part it has played in the story of the primates is truly wonderful. The reader familiar with the recent literature on Paleolithic man need not be told that the pulsation of this ice cap is proving the only reliable time clock for human culture. At four intervals, at least, this ice field crept down into North America and into the Old World, each time greatly modifying the distribution of mammalian life.

Though we now have a fairly complete outline of man's relation to these events in Europe, we are quite in the dark as to what happened in Asia and America. The United States school of anthropologists, led by the veteran Holmes, has successfully combated all claims to an interglacial man in America, but this negative result is difficult to harmonize with the cultural career of the aborigines we have just sketched. No doubt it was the consciousness of this that led Boas[32] to formulate the dissenting view: viz., that man reached the New World during an interglacial period. Many investigators agree that ten to twenty thousand years is all that can be allotted for the lapse of time since the last retreat of the ice in the New World. This, of itself, might give time enough to account for the growth of American culture, but the ice receded gradually, in fact, still hangs about the connecting bridge, so that a much more recent date must be set for the last opening of communication. Reference to our chapter on chronology will show that this leaves a very narrow margin for the development of aboriginal culture. Yet, exactly the same difficulty confronts us in the Old World, for it is clear that the earlier phases of Egyptian development also leave us a narrow margin, if we assume that all the higher cultures we know were developed since the dawn of the Neolithic in Europe. The important difference, however, lies in that whereas we know that man was even at that time a very old inhabitant of the vicinity, his corresponding presence in the New World is denied. Thus, if we accept the theory that man first reached America after the opening of the ice in Alaska, we are under the necessity of assuming a sudden unprecedented growth of culture, unless it turns out that the lapse of time has been greatly underestimated. It seems far more consistent with the facts to assume that the peopling of the New World was contemporaneous with that of Western Europe, and that the subsequent return of the ice practically isolated the two hemispheres, leaving each to develop as it might.

If we leave out of account everything below the Tropic of Cancer, a certain parallelism appears between western Europe and eastern North America, though far more strongly accentuated in the former. The Crô-Magnon man, who appears rather abruptly in western Europe, has in his disharmonic face, one of the most prominent New World characters, and

Fig. 99. General Lines of Dispersion for Mankind. The broken lines represent the more recent movements.

it is not at all improbable that future research in Asia will give grounds for assuming the Crô-Magnon and contemporaneous New World peoples to be collateral branches from a central Asiatic type. More than once, attention has been called to certain vague similarities between certain Paleolithic races and the Eskimo, and in the New World certain older skulls from the remoter parts of South America are not far removed from this same Eskimo type. Incidentally, we may note that the Chancelade skeleton in western Europe belonging to Magdalenian time, is quite similar to the modern Eskimo. The earlier races appearing in Europe tend to be long-headed, and we have noted a less marked, but still noticeable tendency for the long heads in the New World to cluster on the extreme margins. That this is rather fundamental appears when we regard mammals as a whole, for we read that "when the parallel series in Europe and North America are sufficiently complete, they are seen to be not parallel phyla of independent local evolution, but periodically recruited by more progressive new stages, apparently from a common center of dispersal. The relations are like those of one side and the other of a branching tree whose trunk region is unknown to us."[33] The human phenomena we have been considering appear, therefore, as but an integral part of mammalian expansion, and, for that reason, become more evident.

In conclusion, then, we seem to have a large mass of hominidæ occupying the greater part of Asia and America, the tendency of which was toward round-headedness and straight hair; hanging on the outskirts of these were many variants. This interpretation is not antagonistic to those who regard the modern Indo-European as distinct from the Asiatic-American branch, for the suggestive parallels between earlier types of western Europe and America arise in a much earlier period of man's history. That the New World native is a direct descendant of the Asiatic Mongolian is not to be inferred, for the differentiation is evidently remote; what is implied, is that somewhere in the distant past the Asiatic wing of the generalized type diverged into strains, one of which we now know as Mongolian, and another as American.

GROUPING BY SOMATIC CHARACTERS

After these necessary digressions we may turn to our initial task, the classification of New World peoples according to somatic characters. In the first place, our previous discussions have revealed the primary somatic units to be the in-breeding local social groups. The resemblances of individuals within these are comparable to family resemblances among our own people. If, for example, we take the Navajo, Apache, Pawnee, Teton-Dakota, Patagonian, and Eskimo, there is often no difficulty in placing an individual by his portrait alone. On the other hand, to reduce the characters of identity to exact statement, is extremely difficult, just as is the case when one attempts to write a facial description of an acquaintance that would clearly identify him. Our problem, then, is quite the same as in the preceding chapters, viz., to find some empirical grouping for these small somatic groups.

Those who are reasonably familiar with anthropological literature can understand the sexual conditions that readily contribute to the leveling down of family differences to the group type. When we take up the question of group resemblances, we may, upon a priori grounds, assume that contiguous groups will often mix to an extent sufficient to bring them very near to each other. The data for tribal social practices give every justification for this assumption. Then when one plots out a series of anatomical measurements on a map he finds that high values often tend to cluster in one geographical area and low in another. Thus, in case of stature, we rarely find a single tall social group surrounded by groups of low stature. For example, the Cheyenne are rated as tall (1,748 cm.), near them were the Crow (1,732), Arapaho (1,728), and the Dakota (1,726). In fact, all the tribes of the bison area which cluster around the above group show high stature, but receding to the south, west, and north. This is comparable to what we find to be true of culture traits, viz., a tendency to radiate around a geographical center. Another good example is around the mouth of the Colorado River, where the Mohave are rated at 1,740, the Maricopa at 1,722, the Yuma at 1,700, and the Pima at 1,703, while surrounding them are groups of lower statures. A similar grouping appears in head measurements, a fine illustration of which is seen in Hrdlicka's[34] data on the Indians of northeastern United States, where we find certain agreements between tribal groups that are contiguous. In short, it appears that as we pass from one social group to another, there is a gradation of somatic characters and that these gradations radiate from centers in much the same way as we noted for culture characters.

Now, if we generalize, it appears that the same leveling causes that unify the internal somatologies of the several social groups also operate to reduce group differences. On logical grounds this leveling will be most marked where the opportunities for sexual contact are greatest, and it follows that these opportunities will be greatest where cultural diffusion is accelerated. It appears, then, that our best lead in the development of a somatic classification is to seek for correlated distinguishing characters in each recognized culture area.

Before proceeding, it may be well to try grouping a few measurements according to these areas. The most accessible are stature and cephalic index, for which we present the

STATURES OF TRIBAL GROUPS ACCORDING TO CULTURE AREAS

(a) (b) (c) (d) (e) (f) (g) (h) (i) (j)
cm.
150
155 1 2 3
160 1 3 2 2 3 11
165 3 2 1 8 2 16
170 5 2 10 4 2 6 3 32
175 1 3 1 1 6
180  
—--
68

(a) Eskimo
(b) Eastern Woodlands
(c) Mackenzie
(d) Plains
(e) North Pacific Coast

(f) California
(g) Southwest
(h) Southeast
(i) Mexico
(j) Total

CEPHALIC INDEX ACCORDING TO CULTURE AREAS

(a) (b) (c) (d) (e) (f) (g) (h) (i) (j) (k) (l) (m) (n) (o)
71 1 1
72 1 1 2
73 3 2 5
74 1 1
75 1 1 1 1 1 5
76 1 1 1 3
77 2 2 1 2 1 1 9
78 1 4 6 1 1 1 14
79 1 1 3 1 1 3 10
80 1 3 1 1 1 1 8
81 1 1 2 1 2 1 8
82 1 1 1 3
83 1 1 1 1 1 1 6
84 1 2 1 1 2 3 1 11
85 1 1 2
86 1 1 2
87 1 1 2
----
92

(a) Cephalic Index
(b) Eskimo
(c) Mackenzie
(d) North Pacific
(e) Eastern Woodlands

(f) Plains
(g) California
(h) Southwest
(i) Mexico
(j) Southeast

(k) Guiana
(l) Andean Highland
(m) Patagonia
(n) Eastern Brazil
(o) Total

accompanying tables. If, for instance, there were no correspondences between the statures of social groups within the same area, we should expect the distributions in the successive columns to be similar, or to have a chance relation. This is not what we observe, but, on the contrary, in almost every case there is a tendency to cluster around an area average. Thus, our initial assumption seems justified and we may expect a somatic grouping of natives at least roughly coincident with the culture grouping.

If, however, we disregard all such assumptions and proceed empirically with the compilation of a distribution map we reach a result shown in Fig. 100. First, we may be reminded that the data and conditions under which this map must be made do not permit of such clear-cut distinctions as were possible in the preceding chapters. The determining characters used here were stature, pigmentation, head, and face form. Proceeding, then, with these limitations, we find, first of all, that the Eskimo are fairly distinct. Again, centering in the North Pacific area, we have another type, and stretching east and above the Great Lakes and out to the Atlantic, is a third. The centering of 4, 5, 6, and 7 on our map is approximately coincident with culture areas. Central America seems to be a separate area, but the data are meager.

South America is still more vague, but we have the suggestion of the following: First, an upper Andean region including the greater part of Ecuador and Colombia; second, the great Amazon Basin; third, eastern and southern Brazil, with a possible extension into Peru; fourth, the Patagonian-Araucan group; and finally, the Fuegians.

It should be understood that somatic identity is not claimed, for such is improbable outside of the social groups we have designated as the fundamental units, but that a general similarity exists. Additional data will surely modify the boundaries we have designated, but it is unlikely that the areas as a whole will be effaced.

CHRONOLOGICAL TYPES

In the preceding, we have assumed the same attitude as in the other initial classifications, viz., treating all as if contemporaneous in origin. This is, of course, the only attitude to assume toward observations upon the living, but skeletons from burials may belong to very widely separated periods. We must, therefore, give some consideration to the chronology of somatic types. In the first place, no skeletal remains have so far been found to which positive antiquity can be assigned, at least, not to the satisfaction of the critical. A considerable number of fragmentary skeletons have come to light, and achieved literary fame, among which are the Lansing Man, the Calaveras Skull, the Nebraska Loess Man, and the Homo pampæsus proposed by Ameghino. The claims of all these to

Fig. 100. Somatic Areas

a respectable age have been vigorously assailed by Hrdlicka,[35] and, consequently, placed in the doubtful column. Nevertheless, one must suspect that where so many cases arise which exercise the utmost ingenuity of scientists to disprove, the probability of some being authentic is very great. The one positive point resulting from these controversies is that whatever may be the age of the skeletons in question, they are of the same general type as those of the surviving New World native. This finding is consistent with our conclusions as to his homogeneity and probable single origin. On the other hand, it will not do to argue that unless we find skeletons that do differ from this general type, they cannot, therefore, be old. But as this question of the antiquity of man in the western world is not our present concern, we may accept the above conclusion in-so-far as it applies to the homogeneity of type. Under such conditions, the problem of antiquity shifts to the geological and faunal associations. For example, the contemporaneity of man and the mammoth has not yet been established for North America, whereas in Europe there cannot be the least doubt of it. There is just one case of association with extinct fauna that promises something. In 1914 a human skeleton was found in the famous La Brea asphalt bed of southern California, among the bones of both extinct and extant animals.[36] When this deposit is completely excavated and its faunal strata determined, we may have an early date for man's appearance here, though not earlier than late Pleistocene. Again, however, the somatic type is the same. We may anticipate, therefore, no future skeleton finds but what are of the general New World type, but if we consider the data in other parts of this work, we must expect some of them to have a respectable antiquity.

While this point of view applies to the New World as a whole, it still remains possible that there have been changes in the sub-types occupying some of the areas we have just designated. Something like this is suggested for South America,[37] where we find some reasons for assuming that the coast, at least, was first occupied by a fishing people with long heads, who later gave way to a round-headed population. There are, also, some indications that these long-headed South Americans are somatically similar to the Eskimo and so, possibly, connected; but for the present, such interpretations must be taken as problems stated rather than solved. The facts are that since a general unity appears to hold for the whole New World population, we cannot expect definite chronological grouping for somatic types until the time relations of their cultural associates are determined. Our previous discussion of the time relations for culture shows that little aid can come from that quarter until new evidence is produced. Without this help, it would serve no definite purpose to summarize the few random observations of local somatic sequences found in the special literature of the subject.


  1. Deniker, 1900. I, p. 292.
  2. Posnansky, 1916. I.
  3. Boas, 1895. I.
  4. Jenks, 1916. I.
  5. Jenks, 1916. I.
  6. Hrdlicka, 1916. I.
  7. Duckworth, 1904. I, p. 317.
  8. Deniker, 1900. I, p. 89.
  9. Posnansky, 1916. I.
  10. Ripley, 1899. I, p. 43.
  11. Martin, 1914. I.
  12. Boas, 1912. I, pp. 177–183.
  13. Martin, 1914. I.
  14. Berry and Robertson, 1914. I.
  15. Martin, 1914. I.
  16. Deniker, 1900. I.
  17. Martin, 1914. I.
  18. Martin, 1914. I.
  19. Deniker, 1900. I, pp. 63–64.
  20. Sapir, 1916. I.
  21. Boas, 1907. I.
  22. Spinden, 1913. I.
  23. Thomson, 1899 I, pp. 125–128
  24. Hrdlicka, 1915, I.
  25. Hrdlicka, 19195, I., p 91.
  26. Haddon (no date).
  27. Giddings, 1909. I.
  28. Chapin, 1913. I, p. 210.
  29. Duckworth, 1904. I.
  30. Osborn, 1910. I.
  31. Matthew, 1915. I, pp. 214, 215.
  32. Boas, 1912. I.
  33. Matthew, 1915. I, p. 270.
  34. Hrdlicka, 1916. I.
  35. Hrdlicka, 1907. I; 1912. I.
  36. Merriam, J. C., 1914. I, pp. 198–203.
  37. Joyce, 1912. I.