(Chapter I, section 3)
Museums both public and private, in almost every European country as well as in America. contain thousands of crania and long bones which represent the osseous remains of individuals of every race. Many of these, without doubt the majority, are those of persons of white racial origin. For the purposes of the present study, these skeletal remains assume vastly different values, depending upon a number of circumstances. In the first place, only those which have been measured, described, and published were of any use to the present author, since it has not been possible for him to travel from museum to museum measuring and observing the unpublished material. The majority of collections are still unpublished, and hence the majority of data is as useless as if they were still in the ground. To make such a measuring trip would probably take the best years of one investigator's lifetime.
The first consideration is, then, whether or not the material has been published. The second is, whether or not it is properly documented as to sex, provenience, and cultural association. A number of older cranial series is has been published without regard to sex, which makes measures of variability of slight value, and jeopardizes the use of means. Others include skulls from different localities, vaguely labelled and catalogued, which should never have been put together. Still others, and these are many, were unearthed at a time when the archaeologists had not yet so perfected their techniques that the cultural and chronological associations of these remains could be determined. Still others were brought into museums by amateurs who paid no attention to archaeology.
In many cases it is possible to review the published documents as to archaeological settings, and to revise them in the light of present knowledge, especially when illustrations are given identifying the grave furniture and types of sepulchre. Therefore the number of crania and other bones which may be realigned so as to fit into geographical, cultural, and chronological pigeon-holes is not as small as it might be if this material were gathered without recourse to this salvaging process. The realignment mentioned above is the principle upon which the following six chapters have been constructed. It has involved abstracting single skulls and small series of crania, with or without accompanying long bones, and combining the data so abstracted into statistical series based on an identity of place, time, and cultural milieu. In some cases earlier investigators had already effected this process of compiling and combining in a suitable way, so that much of the labor involved could be omitted.
The materials upon which Chapters II to VII are based consist, therefore, of a number of series of crania, in some cases accompanied by other bones, each series representing a cultural, chronological, and geographical entity, the existence of which seems fully justified in the light of our present knowledge of archaeology and of history. Published materials which cannot be reasonably documented in all of the respects mentioned have been ignored, or used with caution.
The crania which meet these requirements and which represent ancestral strains of the white race are numerous enough to permit a reasonable reconstruction of the racial history of the white peoples; but they are not numerous enough to permit us to be sure that our reconstruction is the only possible one in every place and instance. We therefore present with some confidence the main thesis of our reconstruction, but we are not confident that it is correct in every period, in every region, and in every cultural unit.
The entire Palaeolithic period in Europe, for example, is represented by no more than one hundred published and documented skulls, while the Mesolithic is represented by a no greater number. Certain Neolithic samples, especially in Egypt, consist of several hundreds of crania, and the same is true in the Bronze and Iron Ages. No craniological series yet published exceeds one thousand adult specimens of a single sex, although several closely approach that figure.
Skeletal material of human and near-human primates, from the Lower and Middle Palaeolithic cultural levels, is derived from chance finds of unburied fossil bones. In Europe, Neanderthal man first buried his dead in such a way that entire skeletons would be preserved for anthropologists of the future. At various points in human history cremation appeared, to confuse and dismay the racial historian; the chief vogue of this science-inhibiting custom began during the late Bronze Age in Europe, and lasted well into the Iron Age.
In our era another force has arisen to prevent the use of skeletal material; this is the practice of burying bodies in Christian and Moslem cemeteries, both of which are inviolate on religious grounds. Even where they are not inviolate, the absence of grave furniture in the tombs of these followers of revealed religion makes looting by archaeologists unprofitable. The only skeletal collections of any abundance in post-Christian times are those derived from mediaeval charnel houses or crypts, especially in South Germany and Austria, and in certain English cathedrals.
From the statistical standpoint our skeletal materials stand in a borderline position. A few series are large enough to permit the exercise of all of the statistical constants of the modern biometric school; most, however, are so restricted in numbers that a simple calculation of means, a simple determination of variability and homogeneity, and an informal compari-son and discussion are the only techniques which seem justified.22 Too great a mechanization would render such series inflexible and destroy much of their interpretive value. To make up for their statistical weakness, their use as context material for cultural and chronological horizons provides a certain strengthening. A series, however small, tells us what is present, but does not tell us what is additionally present, or what is absent. The extent to which small series may be employed in an interpretative sense must depend upon the circumstances.
The number of criteria measured upon the crania used in this survey range from one - almost always the cranial index - to the five thousand of von Török. In combining and reseriating these series I have made no fast rule as to what criteria to admit and what to exclude, but have employed what seemed to be a reasonable number, with especial emphasis upon those which find parallels on the living. For example, I have usually ac-accepted the three principal dimensions of the cranial vault - glabello-occpital length, maximum biparietal breadth, and basion-bregma height; the usual circumferences and arcs of the cranial vault; the minimum and maximum frontal and bizygomatic diameters; the interorbital and biortital diameters, and the height and width of the orbits; the height and breadth of the osseous nose, the diameters of the palate, and of the foramen magnum; the heights of the face from nasion to menton, and nasion to alveon; the principal dimensions of the mandible, such as the mental height, the breadth of the ascending ramus, and the bicondylar and bigonial diameters. In the rest of the skeleton, I have used almost exclusively the maximum lengths of the principal long bones, such as the femur, tibia, fibula, humerus, radius, and ulna, and then almost entirely for the sole purpose of reckoning stature, by means of the Pearson formulae.23
In other words, I have used what I could find in such a way as to derive the maximum useful information from it; I have not concerned myself with techniques or routines which had little bearing on my problem. Or the whole I have worried little about technical discrepancies due to differences in measuring methodology; where possible I have followed the techniques approved by Morant, and where possible I have made allowances for such differences as I could readily detect. I do not feel, however, that technical discrepancies in the craniological materials are important enough to make any perceptible difference in my conclusions, either detailed or general. The treatment of the material has been done in such a broad manner that such minutiae are of little importance. Craniology is a more accurate science than is the anthropometry of the living; when we come to the later chapters we may concern ourselves with the question of technique, but for the moment it is relatively unimportant.
21 For an exhaustive study of this subject the reader is referred to the standard text in physical anthropology, Rudolf Martin's Lehrbuch der Anthropologie, 3 vols., second edition. The present section is intended merely as a brief statement concerning some of the fundamental uses of osteometric techniques, as well as of the sources and numbers of materials, employed in the present study.
22 For a more detailed discussion of the use of statistics in racial studies, see Chapter VIII, section 2.
23 See Martin, Lehrbuch der Anthropologie, second edition, vol. 2, pp. 1020-1021.