704. REPORT—1904. 
the chief cause of death at this period of life is pulmonary tuberculosis, which both 
in case-incidence and death-rate draws most heavily from those of brunet traits, 
so that as the time passes on the relative frequency of darker individuals is dimi- 
nished, and the parents of children born later in life are less likely to present the 
same relative preponderance of brunet traits. However, it is also true that fewer 
children are born of late than early marriages, so that this would not quite 
equalise matters. We have seen that the most overcrowded areas are, on the 
whole, the most brunet areas, and also those in which the infant mortality is 
greatest. That they do not become still more brunet is, perhaps, to be explained 
by the earlier rise of mortality from pulmonary tuberculosis, the greatest scourge 
during the child-bearing periods of lite. 
The remaining brunet-selection diseases, those of the nervous systems and 
cancer, presenting their maximum later in life, only slightly affect the population 
during this important period, and their influence must be slight. 
4. An Anthropometric Survey. its Utility to Science and to the State.' 
By Joun Gray, B.Sc. 
The principal object of an Anthropometric Survey is to make maps showing 
the distribution of physical and other measurable characters of the population of 
acountry. Topographical and geological surveys have already been carried out 
in great detail by most civilised States, but only a few countries have made 
more or less feeble attempts to map out the characteristics of their populations with 
the same precision as they have mapped out their topographical features and 
their geological strata. 
It may be objected that an anthropometric survey would be impracticable and 
useless because there is not the same permanence in the physique of a people that 
we find in the topography and in the geological strata. But we know enough 
of the law of ancestral heredity to be practically certain that the average bodily 
dimensions of a stationary population will be transmitted with little or no change 
from one generation to another for vast periods of time, provided the environment. 
or conditions of life remain constant. For example, recent investigations have 
shown that the physique of the present population of Egypt is practically identical 
with that of the population 9,000 or 10,000 years ago. 
There is, therefore, no necessary lack of the permanency necessary to make a 
survey of the national physique possible. If the environment should not be con- 
stant, the stability of the physique would still be sufficient to enable surveys to be 
taken at intervals of five or ten years. 
The Ideal Anthropometric Survey. 
In an ideal anthropometric survey statistics would be collected of the com- 
plete bodily and mental features and activities of the population in every part of 
the country. The environment peculiar to each section of the population would 
also be noted. All characters observed should be capable of more or less precise 
measurement. 
The dimensions of the body can now be measured with the greatest precision. 
Measurements of physiological activities, such as the acuteness, &c., of the 
senses, say of sight, hearing and smell, can also be measured with considerable 
accuracy. Psychological characters are more difficult to measure, but still a fair 
estimate of the mental characters of a local population may be formed from its 
occupations and amusements, and from the number of distinguished men it has 
produced. 
The Practical Anthropometric Survey. 
In a practical survey we must be content with the measurement of a few 
characters in order to keep the cost within moderate limits. 
Asa practical scheme that might be carried out by the State, I give, in out- 
1 To be published in full by the Anthropological Institute. 
