The racial history of France is so integral a part of the racial history of western Europe as a whole that there is little need to review its earlier phases in detail. The Neolithic foodproducers who first settled this country came largely from the south, from Spain and also from Italy; the Danubian invasions affected France little, if at all, in a direct racial sense. Megalithic invaders paid considerable attention to the whole western shore of France, and penetrated up the river valleys of the north, while Brittany was their especial stronghold. They were not, however, the first food-producers to arrive, as in Ireland, Scotland, and Denmark; hence their influence upon the subsequent population was relatively slight.

France was a cultural backwash during the Bronze Age; the farmers of Neolithic tradition tilled the valleys and plains, while hunters and gatherers of Mesolithic inspiration still wandered about the infertile uplands. Only in the northeastern part of France, adjacent to southern Germany, was there a Bronze Age civilization of any importance. The Iron Age brought with it invasions from the north of considerable magnitude; first the waves of Keltic peoples, and then of Germanic, culminating in the establishment of Charlemagne's Frankish empire. These invasions gave to the whole north of France a Kelto-Germanic racial, cast, which has penetrated many other parts of the country. The Nordic infusion so produced has had a lasting effect upon the French racial composition.

Other movements of importance were the penetration of the Basques northward, as recorded in the preceding chapter; the arrival of the Northmen from Norway in what became, under their regime, Normandy; the earlier arrival of Saxons along the coast; and the settlement of Cornishmen in Brittany. In more recent times the infiltration of Italians into the Riviera is a racial movement of some consequence.

The Romans established themselves more firmly and with greater success in Gaul than in most of their colonies; the Romanized Kelts gave up their language for a popular variety of Latin, as did the Aquitanians in the southwestern portion of the country, and the Ligurians in the southeast. Greeks, Armenians, Jews, and other subjects of the Roman empire established themselves in Gaul in considerable numbers. The Parisian spirit of internationalism dates back to the Roman occupation. The survival of Romance speech through the blanket of Frankish German and of Norse in Normandy is a tribute to the strength of the Roman imprint.

Throughout her history, France has absorbed more than she has expanded; except for French Canada, she has never had a colony to which Frenchmen have gone in numbers to settle. In the same sense the territory of France is greater than her linguistic boundaries; on the corners of her domain are border provinces in which new foreign tongues have crept in, or in which older ones have long resisted absorption. Italian, in the southeast, is new; Basque and Breton date to the fifth century of our era-of the two the former is increasing, the latter slowly decreasing; Catalan in the Roussillon, so closely related to Langue d'Oc, is apparently static; in the north, Flemish, reaching westward from Belgium, is gradually on the decrease, as is German in Alsace. Although French is spoken by thousands of educated persons outside French territory as a second language, it is not an aggressive language within France itself. The total number of persons of native French citizenship within France whose mother language is not French is three and a half out of forty-two millions. At the same time four other millions out of the forty-two are naturalized or unnaturalized foreigners. The emigration of Frenchmen is negligible.

At the turn of the twentieth century, France was probably the best documented of the larger European countries in an anthropometric sense. Since that. time, however, almost no further statistical information has been collected; our sources are the same as those with which Deniker and Ripley worked. The material consists almost entirely of detailed studies of the distribution of a few characters, notably stature, the cephalic index, and pigmentation. The only new contribution that one can make lies in the field of interpretation.²

If we pass rapidly through the geographical distribution of stature, the cephalic index, and pigmentation, we shall have covered most of the existing information of an accurate nature. The mean stature of the French is about 166 cm.,³ which is neither tall nor short, but intermediate in relationship to other European peoples. France is divided into two principle stature zones by a slightly convex line which passes diagonally from Cherbourg to Marseilles, with mean statures of 166 cm. to 168 cm. lying to the northeast, and those ranging between 161 and 165 cm. on the southwest. Aside from this general scheme, taller people are found along the larger river valleys than in the hills, with one principal exception-the inhabitants of the northern slopes of the Pyrenees, from the Basques to the Catalan-speakers of the Roussillon, are taller than the people immediately north of them. In the northeast, in the taller region, there are stature modes of 164 and 168 cm.⁴ The centers of relatively short stature in France are: the Maritime Alps, to the east of the valley of the Rhône, which acts as a wedge of newer population between the mountain nuclei on either side; the Massif Central, the classic Alpine country; the Perigord-Limoges region, including the Dordogne, which is the strongest outpost of dolichocephals in France; and Brittany.

It is curious that the Keltic-speaking Bretons are among the shortest people in France, and are, in fact, seven centimeters shorter than their kinsmen the Cornish who live directly across the Channel. A detailed stature map of Brittany by cantons shows that the jump from Cornwall is not as abrupt as it appears;⁵ around the coast extends a thin band of maritime cantons with stature in the 164-165 cm. class, which gives way rapidly through a zone of transition to an inner nucleus in which the mean stature is 162 cm. This evidence, as well as that of the cephalic index, indicates that Cornish speech has survived in Brittany among a people to whom it is an adopted tongue, while it has died out in southwestern England whence it came.

Stature has increased to a certain extent in France during the last century, as it has in other parts of western Europe; one of the most striking examples of this change is seen in the mountainous region of Savoie, especially in the canton of Mt. Blanc.⁶ In the five year period from 1807-12, the mean stature of some 12,000 men was 158 cm. Within this period, the stature seems to have been static. Between 1828 and 1837, the recruits from this same region had attained the mean of 162 cm., and in the 1872-79 interval they had reached 165 cm. Unfortunately there is no more recent data to trace the further history of this regional group. In the rest of France, the changes have been much less marked; the case of the Savoyards is apparently an example of diminishing isolation.

One of the most widely discussed subjects in French anthropology is that of the so-called taches noires, the black spots upon the stature map of France. These are regions in which the people appear stunted, and whole villages and whole cantons are characterized by stature means well under 160 cm. These dwarfed areas seem definitely linked with poor living conditions and general retrogression. Broca, who studied such an area in Basse Bretagne, attributed this stunting to mineral deficiency, since it occurred mostly in regions of granitic soil.⁷ Collignon,⁸ who studied a second such spot in the Limousin Hills, on the corner of the four departments of Creuze, Corrèze, Charente, and Dordogne, invoked general poverty and misery. His proof that this stature reduction was environmental is seen in a comparison of means between sub-samples of 83 recruits from the canton of St. Pierre de Chignac. Of these 83, 53 who were born there and had always lived there had a mean of 159.5 cm.; 24 who were born in better country but raised in St. Pierre, 159.9 cm.; 15 who were born in St. Pierre and raised elsewhere, 163.7 cm.

Bodily proportions of the French are known to us only through two general series by Collignon.⁹ The French as a group are not notably different from a general European mean; a relative span of 104 is greater than a Mediterranean condition, and resembles that of the western Norwegians, the East Baltics, and the Irish. The relative sitting height mean of 52.4 is not excessive, nor are absolute shoulder and hip breadths. On the whole, the resemblance is with northern Upper Palaeolithic survivors rather than Mediterraneans, which is to be expected.

The data on the cephalic index of the French, while covering smaller series than those for stature, are numerically fully adequate, and have been frequently discussed. France is a brachycephalic country, one of the most fully and intensely brachycephalic in the world. The mean cephalic index for the nation is 83.6, according to Collignon, which would be between 81 and 82 on the skull-in other words, it is about the same as it was during the Neolithic, judging by the relatively abundant cranial material reviewed in Chapter IV. Since most of the post-Neolithic invaders of France, who came in considerable numbers, have been dolichocephalic or mesocephalic, the present condition is evidence in itseff of a prodigious absorption and reëmergence.

Two large zones in France are characterized by hyperbrachycephaly; indices of 86 and over are found in Auvergne and in Burgundy. The first center starts in Upper Gascony with the department of Gers, and extends eastward and slightly northward through Tarn-et-Garonne and Lot in the Guyenne, to Aveyron, Cantal, Lozère, and Haute Loire. This is the famous Massif Central, the granite country, the refuge area of Alpines in their least mixed form. This area of maximum brachycephaly does not, however, correspond exactly with the center of short stature which lies farther to the north and west; nor does it entirely merit the name "Auvergnat," because Auvergne forms merely the northwestern. most extremity of its distribution. Furthermore, it cannot be exactly correlated with any single geological or orographical phenomenon. The second zone of hyperbrachycephaly lies to the east and north of the first one; it is found in Savoie, eastern Burgundy, the Franche Comté, and Lorraine. The inhabitants of these regions differ profoundly from those in the first area, however; the Burgundians and Savoyards are much taller, and frequently blond.

Long-headed regional populations are scarce in France; true dolichocephals, with indices of 77 or under, are numerous only in the immediate region of Perigeux, in the Dordogne. Low mesocephals, with indices of 78 and 79, cover a wider zone around Perigeux, between the rivers Vézère and Dronne. Elsewhere relative long headedness, comprising indices between 78 and 81, is found in a number of regions: (a) the Channel departments, where Norman and Saxon blood is present, and here and there on the coast of Brittany. The Norwegian invaders, with a mean cephalic index of presumably 77, have pulled the regional mean down to 80 and 81 in most of Normandy; in Brittany, however, the Cornish invaders gave the inhabitants little beside their language. (b) the corridor reaching from Orleans to Bordeaux, through Marche, Poitou, and Berry; this has been a highway for invasions from the north since early times. (c) the Catalan-speaking region of Pyrenees Orientales. (d) the lower Rhône Valley, from Lyon to the Mediterranean, another much frequented corridor.

The rest of France, consisting of about half of the country, represents an intermediate condition in head form, with normal brachycephaly, the mean indices being between 82 and 85. In view of the skeletal history of France, and of the racial character of the living French, it is evident that a moderate brachycephaly is not, in this country, a normal racial condition, but an intermediate or mixed one, between end types which are genetically capable of reëmergence.

In France as in Norway, Denmark, and many other countries, there is a tendency for the cities to contain longer-headed populations than the surrounding country districts; in eight cities¹⁰ the mean difference between the two is 1.86 index points. Since the birth-rate in the cities is low, and the cities drain the human surplus of the surrounding country districts, there must be a process of selection at play, here as elsewhere, which tends in the long run to raise the cephalic index mean not only in the country districts, but also in the cities as well. This process is particularly important in France where there has been since the beginning of the Neolithic a highly brachycephalic hinterland population to draw from. In Brittany the change seems to have been particularly profound, since the Iron Age crania from this country in no wise give promise of the present-day round headedness.¹¹

Measurements of the head and face, and indices other than the cephalic, are extremely scanty.¹² Fortunately, however, they refer for the most part to the more brachycephalic element in the French population, which is of especial interest here. In the Alpine region par excellence, where the cephalic indices run to means of 85 and over, the head length means average about 184 mm., and the head breadth about 158 mm. The vault height mean is about 126 mm. These heads, with a cephalic module of 156 mm. (HL + HB + Aur. Ht. ¸ 3) are of moderate size for white people; they are much smaller than the heads of the Borreby brachycephals in Scandinavia and northern Germany, and a little smaller than one finds among brachycephals of equal index position in southern Germany. They are, however, comparable in size to those of Dinarics in the Balkans, and of Armenoids and Tajiks in Asia. Heads among all non-Borreby brachycephals, from France to Turkestan, are approximately equivalent in basic vault dimensions, whatever the differences in contours.

The French Alpine face, however, fails to maintain this level of similarity. The foreheads and jaws are both moderately broad, with minimum frontal and bigonial means of about 108 mm., as is the bizygomatic mean of 140 mm. These lateral dimensions exceed those of any Mediterranean group studied, and approach but do not equal the Borreby position. The French Alpine face breadth is equal to that of Tajiks, but less than that of some Dinarics in the Balkans, and of Armenoids.

The total face height mean seems to be about 121 or 122 mm.;¹³ the upper face height mean about 73 mm. These figures agree closely with those of the Tajiks of Turkestan, who are also for the most part Alpines; but fall far short of those for Dinarics and Armenoids. The Borreby brachycephals in the north do not have much longer faces. The French Alpines are mesoprosopic and mesene. Their nose height mean is about 53 mm., and breadth about 34 mm.; the nasal index approximately 64. Thus the noses are absolutely of moderate size, and moderately leptorrhine. They are, again, close to those of Tajiks, and much shorter than those of Dinarics or Armenoids.

To sum up this material, the Alpines of France, in the measurements and proportions of the head and face, seem to be smaller replicas of the Borreby people of northern Europe. They closely resemble the sedentary Iranian-speaking Tajiks of Turkestan, with whom we shall deal at some length later, and thus have possible relationships with a similar people far to the east. They furthermore differ greatly in facial dimensions and proportions from Dinarics and Armenoids in southeastern Europe and in western Asia. They differ profoundly from any group of Mediterraneans studied, and show a manifest affiliation to the general Upper Palaeolithic European group.

The one region of complete dolichocephaly in France, that of the Dordogne country, is characterized by unusually large head diameters. The mean head lengths of several cantons run as high as 196 and 197 mm., with the breadths at 150 mm. and greater.¹⁴ The vaults are relatively low, being about 3 mm. lower than those of neighboring brachycephals. The bizygomatic means of the long-headed cantons are about 137 mm., as compared to 140 mm. for the brachycephalic cantons. This unusual head size, coupled with short stature, is unquestionably indicative of an isolated local type; but it is too great to refer wholly to a normal, small Mediterranean, Early Neolithic racial group. These dimensions remind one of the Mesolithic people of Teviec; and Ripley may not have been wholly wrong when he saw in the Dordogne dolichocephals a survival from pre-Neolithic times. The Mesolithic is still a period of uncertainty to the student of race, but the one thing that we do know is that it was, like all others before or since, a period of complexity. The Dordogne dolichocephals present a problem similar to that of the more primitive of the brunet dolichocephals of Wales.

The pigmentation of living Frenchmen, like their stature and cephalic indices, was subjected to extensive investigation during the last century, and there is no modern scale material for use in determining absolute standards. The most recent work, that of MacAuliffe and Marie on 6625 men from France as a whole,¹⁵ finds but 4 per cent of black and near-black hair color, 23 per cent of dark brown, 43 per cent of medium brown,. 14 per cent of light brown, 12 per cent of various degrees of blond, and some 4 per cent of reddish-brown and red. The virtual absence of truly black hair is notable, as well as the high degree of rufosity. The characteristic French hair color is a dark to medium brown, which often has a reddish glint; this color is typical of the Alpine race in its French manifestation.

The regional distribution of hair color in France follows closely that of stature. Although the position of the French in regard to hair pigmentation is intermediate between blond and black, the diagonal line from Mont St. Michel to Orleans, Lyons, and the Italian border divides the country into a northeastern quadrant, in which the hair is somewhat lighter than medium, and a southwestern, in which it is somewhat darker. High ratios of black and very dark brown hair are found not in the typically Alpine country, but along the slope of the Pyrenees, in Catalan-speaking country, and on the Mediterranean seacoast. Blond hair is commonest along the Channel, in regions settled by Saxons and Normans, in Burgundy and the country bordering Switzerland, and down the course of the Rhône. In northern France it seems to follow upstream the rivers which empty into the Channel. The hair color of the departments occupied by Flemish speakers, and of others directly across the Channel from England in Normandy, seems to be nearly as light as that in the southern English counties; the coastal cantons of Brittany are lighter than the inland ones, and approximate a Cornish condition. In the same way, the northeastern French departments are probably as light-haired as some of the provinces of southern Germany. Truly light hair is uncommon enough, and so placed geographically that it may be in large part attributed to the Keltic and Germanic migrations. But the hair of the pre-Keltic inhabitants of France can by no means have been wholly or even largely black; the intermediate brown hair shade of the Alpines, with its rufous and incipiently blond tendencies, must be ancient in France; it is comparable to the slightly blonder hair color range of the Borreby type, with its tendency to rufosity.

Eye color observations on the French are equally abundant and equally difficult to equate to standard shades and degrees of pigment.¹⁶ Pure dark eyes are apparently found among roughly 25 per cent of Frenchmen; 17 the departmental range runs from 14 per cent in Morbihan (Brittany) to Basse Pyrenees and Gers, with 41 per cent and 42 per cent, and thence to the very dark-eyed departments of Bouches du Rhône (57 per cent), and Alpes Maritimes (59 per cent). Out of 87 departments, 49 have between 20 per cent and 30 per cent of "dark" eyes.

The distribution of eye color in France follows roughly that of stature and hair color, but is less regular than either. Light eyes are especially numerous in the northeast, in the region of Keltic and Germanic influence, and in northwestern France, along the Channel from Flanders to Brittany. Topinard finds 25 per cent of blue eyes in these northern and eastern departments. In the Pyrenean departments, and along the Riviera, these blue eyes, which probably include light-mixed shades, sink below 15 per cent, but never below 10 per cent. Even in the departments where there is little historical or skeletal evidence of Nordic influence, there is always a large minority element of eye blondism. On the whole, the distribution of eye color differs from that of hair color in one particular: light eyes are relatively common in western France, especially in Brittany, in regions of dark hair color; while light hair is commoner in eastern France than the ratio of light eyes would warrant were the two strictly correlated. France repeats on a lesser scale the hair and eye color disharmony of northern Europe. The reason is the same in both areas; the eye blondism is partly Nordic, partly of Palaeolithic or Mesolithic derivation, while the really light hair is largely Nordic.

The foregoing summary of the detailed regional distributions of somatic characters among Frenchmen has made it clear that France, while more than anything else an Alpine country, is differentiated into a number of racial sub-areas. At the same time it is evident that in France as a whole, a number of distinct racial types may be easily distinguished among individuals. Starting on the regional basis, we have observed that the northern part of France, including the Channel departments and those stretching eastward as far as Burgundy, contains a population characterized by moderately tall stature, a variable but slightly fairer than intermediate degree of blondism, and a variable, sub-brachycephalic or brachycephalic head form. This population obviously contains strong vestiges of the Nordic invasions of Kelts and Germans, but in it fully qualified Nordics of Keltic or Germanic aspect are rare. They are much commoner, however, in French Flanders, and in Normandy. Portrait material indicates that the Nordic element was especially strong among the old French nobility.

In northeastern and eastern France, in the region where relatively tall stature, relatively light hair and eye color, and extreme brachycephaly coincide, this partial Nordicism passes into a Dinaric or Dinaric-like condition. Here the cephalic index is as high as in the central Alpine country; the heads, furthermore, are no larger, and a Borreby element cannot be induced to explain the difference in stature and pigmentation. We must remember, however, that in the Neolithic period the stature of extreme brachycephals in this region was moderately tall, and that the accompanying Mediterranean crania were associated with much shorter stature.

It would seem that the infusion of Nordic blood produced by the Keltic and Germanic invasions helped to maintain this original stature level, or to reënforce it, while at the same time adding considerably to the local blond increment. A study of Savoyards on the basis of head form, head size, stature, and pigmentation¹⁸ demonstrates that in a local group with a mean stature of 170 cm., there is no evidence of Borreby head size, and that two related elements seem to account for almost all of the sample; a Dinaric and a Noric, the latter being a blond brachycephal of general Dinaric morphology. The unavoidable inference is that the original Alpine type has absorbed not only Neolithic Mediterranean factors, but also Iron Age Nordics, in such proportions that the Alpine cephalic index level has been preserved, but that the facial characters have to a certain extent been taken over from the Nordics. In other parts of northern France, in the Seine and Marne valleys, for example, the Alpine element has not been strong enough to produce such a phenomenon consistently, although it has done so with individuals.

If the tall, relatively light-pigmented hyperbrachycephals of northeastern France have absorbed some Nordic blood without change of cephalic index, then it is possible that the shorter, darker ones of south-central France have absorbed various quantities of Mediterranean, since Mediterraneans have been present in this region since the beginning of the Neolithic, if not earlier. Among the French Alpines convex noses are common, and an approach to the Dinaric facial appearance; one wonders if this is not partly due to the absorption of Mediterranean blood. Alpine facial types of the classic variety, with a straight or concave nasal profile, combined with the Alpine abundance of beard growth, and the stiff but wavy, unruly Alpine hair, are by no means found among all Frenchmen who are metrically Alpine.

Here and there one sees a Frenchman of general Alpine type whose facial features, due largely to peculiarities of nose form and to malar prominence, approach a Lappish or mongoloid condition. The same may be seen occasionally in North Africa among Berbers. This must be attributed not to mongoloid invasions, but to the relationship between Lapps and other incipient mongoloids and Upper Palaeolithic Europeans in the Pleistocene. Ainu-looking Alpines are commoner than incipiently mongoloid ones.

Montandon, a keen observer of the French racial scene, proposes the following racial proportions for the French nation: Nordic, 1 per cent; Sub-Nordic, 30 per cent; Dinaric-like, 15 per cent; relatively pure Alpine, 30 per cent; Small Mediterranean (Ibero-Insular), 10 per cent; Atlanto-Mediterranean (Litoral), 10 per cent; Basque type, l per cent; others, 3 per cent. Although the Alpine increment receives only 30 per cent, it must be remembered that in the Sub-Nordic as in the Dinaric-like category, there is a strong Alpine element; furthermore, the Atlanto-Mediterraneans of the Pyrenees and the Riviera are strongly tinged with Alpine. If Collignon's head diameters are correct, then the small Mediterraneans of the Dordogne are not pure Neolithic descendants, but have absorbed a much older non-Alpine racial entity.

The final conclusions derived from this survey are as follows. France, notwithstanding her brilliant contributions to civilization and the international character which she, as a great cultural center, has assumed, was a culturally retarded, marginal area from the end of Mesolithic times until the Iron Age. At the same time, it has remained, since the end of the Pleistocene, a marginal or refuge area from the racial standpoint also, since the invasions of brunet Mediterraneans and of Nordics have together been less important here than in most European countries. In France the Alpine race, a smaller-sized and less blond replica of the northern Borreby race, has reëmerged as the principal racial element and can be seen in a relatively pure form. France is essentially an Alpine nation.

2. The old material has been ably summarized and interpreted by Professor Georges Montandon in L`Ethnie Francaise. His volume contains a complete bibliography of the older sources. Chief among those which have been used in the present section are:
Atgier E. A., BSAP, ser. 4, vol. 9, 1898, pp. 617-637; ser. 4, vol. 10, 1899, pp.171-199.
Aubert, RDAP, ser. 3, Vol. 3, 1888, pp. 456-468.
Bouchereau, A., Anth, vol.11, #6, 1900, pp. 691-706.
Boucherean, A., and Mayet, L., BMSA, ser. 5, vol. 6, 1905, pp. 426-448.
Carlier, G., BSAP, ser. 4, vol. 4, 1893, pp.470-476.
Carriere, G., Homme, vol. 2, 1885, pp. 334-337.
Carret, J., MDSS, vol. 21, 1883, pp. 1-108.
Chassagne, A., RDAP, ser. 2, vol. 4, 1881, pp.439-447. Collignon, R., Anth, vol. 1, 1890, pp. 201-224; vol. 4, 1893, pp.237-258. REAP, vol. 7, 1897, pp. 339-347. BSAP, ser. 6, vol. 3, 1883, pp. 463-526; ser. 3, vol. 10, 1887, pp. 306-312; ser. 4, vol. 1, 1890, pp. 736-805. MSAP, ser. 3, vol. 1, fasc. 3, 1894, ser. 3, vol. 1, fasc. 5, 1895.
Debièrre, C., BSAL, vol. 5, 1886-87, pp.129-149.
Durand de Gros, J. P., BSAP, ser. 2, vol. 4, 1869, pp.193-218.
Garnier, M., Anth, vol. 24, 1913, pp.25-50.
Grimere, BSAP, ser. 6, vol. 4, 1913, pp.392-400.
Hervé, G., REAP, vol. 11, 1901, pp.161-177.
Hovelacque, A., and Hervé, G., MSAP, ser. 3, vol. 1, fasc. 2, 1894, pp.1-256.
Lagneau, G., BSAP, vol. 6, 1865, pp.507-511.
Lapouge, G. V. de, BSSM, 1897, vol. 4, pp. 235-243.
MacAuliffe, L., and Marie, A., Ethnographie, No.5, 1922, pp.41-48.
MacAuliffe, L., Marie, A., and Thooris, A., BMSA, ser. 6, vol. 1, 1910, pp.307-311.
Manouvrier, L., BSAP, ser. 3, vol. 11, 1888, pp. 156-173.
Papillault, G. F., BMSA, ser. 5, vol. 3, 1902, pp.393-526.
Pommerol F., BSAP, ser. 3, vol. 10, 1887, pp.383-397.
Routil, R., ZFRK, vol. 5, 1937, pp.177-181.
Topinard, P., RDAP, ser. 3, vol. 4, 1889, pp.513-530; JRAI, vol. 27, 1897, pp.96-103; Anth, vol. 4, 1893, pp.579-591.
France more than almost any other European country stands in need of a new and complete anthropometric survey. The older surveys suffer in the technical sense as well as in the paucity of criteria studied.

4. Montandon, C., op. cit., p.64.

5. Chassagne, A., RDAP, 1881.

6. Carret, J., MDSS, 1883.

7. Thus anticipating Marett's work by half a century.
Broca, P., BSA, ser. 2, vol. 1, 1866, pp.700-708.

8. Collignon, R., MSAP, 1894.

9. Collignon, R., BSAP, 1883; Anth, 1893.

10. Bordeaux, La Rochelle, Pau, Bayonne, Tarbes, Rodez, Milhau, and Lyon. Calculated from:
Bouchereau and Mayet, 1905; Collingnon, R., MSAP, 1894; Durand de Gros, J. P., BSAP, 1869.

12. Three series are most useful:

(1) MacAuliffe, L., Marie, A., and Thooris, A., BMSA, 1910. A series of 100 French soldiers.
(2) Hawes, C. H., a series of 51 French soldiers, mostly from Lozère, measured in Crete in 1905. This series has not been published previously.
(3) Papillault, C., BMSA, 1902. A series of 100 cadavers measured in the Paris morgue. This series is especially complete and accurate, but unfortunately there had been some shrinking of soft parts, or else social selection was important here, for the cadavers are smaller in many dimensions than living groups.
Aside from these three series, we have partial data on 22 other series, 17 by Collignon, and the others by Carlier, Carrière, Grillière, and Debièrre.

13. There are no accurate total face heights available for France. I am basing this figure on French Canadian convicts in American jails, who scem to be of basic Alpine type. This material is taken from Hooton's extensive criminal survey.

15. MacAuliffe, L., and Marie, A., Ethnographie, 1922. Older surveys which cover France geographically are those of Topinard and of Collignon.

16. Most of the French observers use the terms "marron" and "chatain" to designate the commoner shades of brown eye color, presumably meaning dark brown and light brown, although Topinard pointed out that the only "chatain" that resembled a human eye color was one with a worm in it. Topinard and others observed eye color by standing at a distance and observing the total tone, although Bertillon advocated an accurate system which took into account the anatomy of the iris.

17. This figure is obtained by combining MacAuliffe and Marie's "chatain" and "marron pur"; Topinard's "dark" class gives the same figure.