We know, in fact, that man measures one or two centimetres more on rising in the morning than on going to bed at night, when the fibro-cartilaginous discs situated between the vertebræ are compressed, more closely packed, and the vertebral column is more bent. Unscrupulous conscripts whose stature is near the regulation limit know perfectly well that if the day before the official examination they carry heavy loads, they compress their intervertebral discs so that their height is sometimes diminished by three centimetres.
It is necessary then, in order to avoid error, not only to have measurements taken from adult subjects, but also from several series containing a great number of these subjects. Calculation and inference have shown us that it is necessary to have at least a series of one hundred individuals to guarantee the exact figure of the height of a population but slightly blended. Series of 50 to 100 individuals may still furnish occasionally good indications, and series of 25 to 50 individuals an approximation; but with series under 25 individuals doubt begins and the figures are often most deceptive.
I have brought together and grouped in the table at the end of this volume ([Appendix I].) average statures calculated in series of twenty-five individuals or more. These series have been based on the collation of hundreds of documents, of which limits of space prevent a full enumeration.
An examination of our table shows that the extreme averages of different populations fluctuate, in round figures, from 1 m. 38 (4 ft. 6 in.) with the Negrillo Akkas, to 1 m. 79 (5 ft. 10.5 in.) with the Scots of Galloway.[30] But if we set aside the pigmy tribe of the Akka, quite exceptional as regards stature, as well as the Scots of Galloway, and even the Scots of the north in general (1 m. 78), who likewise form a group entirely apart, we arrive at the extreme limits of stature, varying from 1465 mm. with the Aeta or Negritoes of the Philippines, and 1746 mm. with the Scots in general. In round figures, then, we can recognise statures of 1 m. 46 (4 feet 9.5 inches) and 1 m. 75 (5 feet 9 inches) as the extreme limits of averages in the different populations of the globe. The medium height between these extremes is 1 m. 61, but if we put on one side the exceptional group of Negritoes (Akka, Aeta, Andamanese, and Sakai), we shall note that the rest of mankind presents statures which ascend by degrees, almost uninterruptedly, from millimetre to millimetre between 1 m. 54 and 1 m. 75, which makes the average 1 m. 65 (5 feet 5 inches), as Topinard has discovered.[31] Topinard has likewise proposed the division of statures, since universally adopted, into four categories, viz.: short statures, under 1 m. 60; statures under the average, between 1 m. 60 and 1 m. 649; statures above the average, between 1 m. 65 and 1 m. 699; and lastly, high statures, 1 m. 70 and over.
Our table shows conclusively that there are many more populations (almost double the number) whose stature is above or under the average, than populations of a short or high stature.
Short stature is rare in Africa, being found only among the Negrillo pigmies and Bushmen; in South America a few tribes of low stature are also met with; but the true home of low stature populations is Indo-China, Japan, and the Malay Archipelago. In the remaining portion of Asia this low stature is only met with again in Western Siberia, and among the tribes called Kols and Dravidians in India.
Statures under the average predominate in the rest of Asia (with the exception of the populations to the north of India and anterior Asia) and in Eastern and Southern Europe, while statures above the average comprise Irano-Hindu populations, the Afrasian Semites, the inhabitants of Central Europe, as well as the Melanesians and Australians.
Thus high stature is plainly limited to Northern Europe, to North America, to Polynesia, and especially to Africa, where it is met with as well among Negroes as among Ethiopians.
What is the influence of environment on stature? This is one of the most controverted questions. Since the time of Villermé the statement has been repeated in a variety of ways that well-being was favourable to growth and increase in stature, and that hardship stunted growth. There are facts which seem to prove this. In a population supposed to be formed of a mixture of many races, the well-fed upper classes appear to possess a higher stature than the lower classes; thus, while the English of the liberal professions are 69.14 inches (1757 mm.) in height, the workmen of the same nation are only 65.7 inches (1705 mm.).[32] But can we not likewise adduce here the influence of race? That predominating in the aristocracy and well-to-do classes does not, perhaps, predominate in the working classes. Beddoe[33] and others have remarked that the stature of miners is lower than that of the population around them; in the same way, workmen in shops and factories are inferior in height to those who labour in the open air, and this in Belgium (Houzé) as well as in England (Beddoe, Roberts) or Russia (Erisman, Anuchin).[34] According to Collignon,[35] the populations of Normandy and Brittany living in the neighbourhood of railways and high-roads are superior in height to those living in out-of-the-way places. He concludes from this that the material conditions of life being improved since the formation of roads, the stature of the population has increased. According to Ammon and Lapouge, the population of the towns in France and Southern Germany are taller in stature than those of the country, because of the migration towards urban centres of the tall dolichocephalic fair race which they call Homo Europeus. However, Ranke observed just the opposite, and there are other objections to be raised against this theory, based on the data of recruiting. These town-dwellers of high stature are perhaps only conscripts too quickly developed; town life accelerates growth, and town-dwellers have nearly reached the limit of their height while dwellers in villages have not finished growing. This is so true that in countries where statistics have been taken of the civic population, as in England for example, the population of the towns is shorter in stature than that of the country. Beddoe explains this fact by the bad hygienic conditions in towns, the want of exercise and drinking habits of dwellers in cities.[36]
To conclude, the influence of environment cannot be denied in many cases: it may raise or lower stature, especially by stimulating or retarding and even arresting growth; but it is not demonstrated that such a change can be perpetuated by hereditary transmission and become permanent. The primordial characteristics of race seem always to get the upper hand, and the modifications produced by environment can alter the stature of the race only within very restricted limits. Thus miners of a high stature like the Scotch, for example, while shorter than the Scotch of the well-to-do classes, will be still taller than the individuals of the well-to-do classes in, for example, Spain or Italy, and much more so than those of Japan (1 m. 59). Stature is truly then a character of race, and a very persistent one.