ELDERTON

The Relative Strength of
Nurture and Nature

HQ
751
E4

The Relative Strength of

Nurture and Nature

x'

(SECOND, MUCH ENLARGED EDITION)
PART I

THE RELATIVE STRENGTH OF NURTURE
AND NATURE

(Second Edition, Revised)
BY

ETHEL M. ELDERTON,

GALTON RESEARCH FELLOW

PART II

SOME RECENT MISINTERPRETATIONS OF THE
PROBLEM OF NURTURE AND NATURE

(First Issue)
BY

KARL PEARSON, F.R.S.,

GALTON PROFESSOR

CAMBRIDGE UNIVERSITY PRESS

C. F. CLAY, MANAGER
LONDON: FETTER LANE, E.G.
EDINBURGH: 100, PRINCES STREET

ALSO

H. K. LEWIS AND CO., LTD., 136, GOWER STREET, LONDON, W.C.
WILLIAM WESLEY AND SON, 28, ESSEX STREET, LONDON, W.C.

CHICAGO : UNIVERSITY OF CHICAGO PRESS

BOMBAY, CALCUTTA AND MADRAS : MACMILLAN AND CO., LIMITED
TORONTO I J. M. DENT AND SONS, LIMITED
TOKYO : THE MARUZEN-KABUSHIKI-KAISHA

PREFATORY NOTE

The statistical material used in these two lectures is partly drawn
from Dr David Heron's memoir The Influence of Unfavourable Home
Environment and Defective Physique on the Intelligence of School
Children (Galton Laboratory Memoirs, No. VIII) and from A First
Study of the Inheritance of Vision and of the Relative Influence of
Heredity and Environment on Sight by Amy Barrington and Karl
Pearson (Galton Laboratory Memoirs, No. V), but is also largely
based on hitherto unpublished material, which cannot be issued in
bulk until the laboratory staff has greater leisure and more ample
funds for publication than are at present available.

CAMBRIDGE: PRINTED BY JOHN CLAY. M.A. AT THE UNIVERSITY PRESS

PART I1

THE RELATIVE STRENGTH OF NURTURE
AND NATURE

When we approach the fundamental problems of human life, and
endeavour to study the causes which make for the physical and mental
fitness of a nation, we are at once struck by the vast network of
associated factors, which render so complex any attempt at a solution
of the simplest problem in the relationship of nurture and environment
to human progress.

We might for example discover that the mentally and physically
fitter child comes from the cleaner and less crowded home, and
therefore argue that the spread of such homes is a condition for
national progress. But further investigation may show that the
essential condition for such homes is the existence of mentally and
physically fit parents, and that the fitness of the children flows after
all from the parentage and not from the home.

Or again if we take the problem of employment of mothers, we
might discover that employed mothers have the healthier children,
and we might on this ground oppose any attempt to restrict the
factory work of women during childbearing years. But on further
investigation we might ascertain that such mothers are, on the
average older and have older children, from whom time has weeded
the weaklings, and that the problem cannot be solved by the simple
question of the physique of the children of employed and unemployed
mothers, without regard to the age distribution of the children.

Once more it is conceivable that we might find the deathrate
of innkeepers much in excess of the deathrate of the clergy, and

1 This paper was originally delivered as the seventh of a course of Eight Lectures
on National Eugenics at the Galton Laboratory, 1909.

^ THE RELATIVE STRENGTH OF

;.».k tlu- fact that the average age at starting the occupation of
innkeeper i- tar higher than that of entering the church, because
ultimately to keep an inn is the ambition of men, who begin life in
a variety of other callings. As a last illustration we may take the
problem of the effect of the drinking of the father on the health and
ability of the offspring. We might find that the father who drinks
had abler and physically better developed sons than the non-drinker.
Yet on closer investigation it might appear (i) that on the av
the drinker was the abler workman, and thus the apparently gr
ability of the offspring was really an hereditary matter, (ii) that the
abler workman got higher wages, and thus, notwithstanding the
drink, the food at home, especially perhaps in the earlier days of the
family life, had been ampler and of better quality. There is hardly
a single problem of parental occupation and habit, of home environ-
ment and school influence which is not of the greatest complexity, and
full of pitfalls for even the most cautious statistician. And if this be
true, what must be said for the philanthropist and social reformer who
without hesitation preach that social salvation lies in this or that line
of conduct? What must be our judgment of the "practical man"
and the statesman who legislate in the direction indicated by the
falling scale of popular opinion, without any real examination of the
intense complexity, the subtle biological effects of even slight changes
in the factors on which our national weal depends?

A factory act may be carried owing to a wave of popular emotion,
which paints the horrors of child-labour in the mills. Twenty years
afterwards it may become apparent that children were taken care of
because their labour was of value, but that their value depending on
their labour, they are, since the act, an unmarketable commodity and
have ceased to be born on this very account1.

What guide can we take to indicate the path of true social reform
through such a tangle of cause and effect as we find involving the
relative influence of nature and nurture on human life? It is not
enough to show that results are associated with this or that factor;
we have a vast complex of associated factors, and out of this complex
we have in some way to pick out the more important and in a certain
sense the fundamental factors. The only effective method by which
it seems possible at present to approach such a problem is that of

1 For a fuller discussion of this point see Pearson : The Problem of Practical
Eugenics. Galton Laboratory Lecture Series V ; and Elderton : The English Birthrate,
Part I, North of the Humber, Galton Laboratory Memoirs. Nos. XIX and XX.

NURTURE AND NATURE 5

correlation1. Taking the social conditions we wish to modify, we
must study their correlation with as many factors as we can possibly
measure. In the choice of these factors we must of course be guided
by the reasonable probability of association and by the limits of
human life and energy. The correlations of a multiplicity of factors
being known we may justifiably assume that the factors with the
highest correlations are, among those dealt with by us, the most
important, and then the process of "partial correlation" will guide
us still further towards a final judgment of what fundamentally are
social cause and effect.

We admit to the full that even then we may not have avoided all
danger of pitfalls, that we may have overlooked possible factors, and
that spurious correlation may have arisen from all sorts of disregarded
selective processes. The prudent statistician will always advance his
conclusions with a word of caution ; he will simply state that they are
those which reasonably follow from the data provided. But at the
same time he will not hesitate to proclaim that in the present state
of our knowledge the calculus of correlation is the sole rational and
effective method available for attacking these urgent social problems.
If that calculus throws no light, when properly applied, on social
dynamics, then the only solution is to develop a finer statistical
calculus ; no other instrument, least of all general reasoning with
appeal to social or moral pre judgments, can at present aid us in our
difficulties with regard to what makes for, and what mars national
fitness. Indeed we venture to go further and to assert that much of
the canon of social conduct and moral action as it has existed in the
past and as it widely exists at the present will be found inadequate
or even antisocial, when we understand more fully the manner in
which legislation, social custom, and philanthropy (in the old sense)
tend to modify the biological factors on which human progress so
largely depends.

There is no one with the least biological knowledge who would

1 The coefficient of correlation measures the amount of resemblance or association
between characteristics of individuals or of things; it is represented by a decimal
which lies between o and i. As the correlation coefficient rises to i we approach a
condition of absolute dependence. As it falls to o we approach a condition of absolute
independence. Thus the correlation between right and left femur in man is -96 which
is practically unity, i.e. almost perfect dependence as we should expect. The inheritance
between stature of father and son is -51, half-way between absolute dependence and
absolute independence.

6 THE REI.A11VI -1KI N(.TH OF

6 to admit the enormous part played by environment in modifying
living i. .mis. This modification, however, is of two kinds:

A change in the somatic characters of the individual following
upon his transfer to different surroundings or his treatment to differ* -nt
conditions of nurture ; this environmental change in the individual
appears to be more marked in plants than in animals. If it is in any
case persistent after the offspring of the individual have returned to
the original environment, most biologists would assert that a germinal
change of some kind, an unrecognized selection of germ plasm, has
taken place in the original individual.

(b) A change in the germ characters of the race, owing to the
environment selecting for survival a certain differential class of
individuals, and their somatic characters thus becoming more frequent
and possibly dominant in the population owing to the strength of
heredity.

Now we are not in any way concerned with a consideration of this
second modification due to environment. We have solely to deal with
the problem of the extent to which the offspring are modified by an
indirect environmental factor, namely the occupations and habits of
their parents and the condition of their homes. We are not considering
(b), for we are dealing with surviving children, and not with the
selective infantile deathrate. Now the influence of the parental
environmental factor on the welfare of children is of fundamental
importance, quite apart from the selective infantile deathrate and the
possibility of the inheritance of acquired characters. It is at present
and has been in the past the chief direction of legislative and philan-
thropic attack on social evils. Degeneracy of every form has been
attributed to poverty, bad housing, unhealthy trades, drinking,
industrial occupation of women, and other direct or indirect environ-
mental influences on offspring. If we could by education, by
legislation, or by social effort change the environmental conditions,
would the race at once rise to a markedly higher standard of physique
and mentality? Much, if not the whole of the battle for social reform
has been based on the assumption that this question was obviously to
be answered in the affirmative. No direct investigation has really ever
been made of the intensity of the influence of environment on man.
To modify the obviously repellent was the immediate instinct of the
more gently nurtured and controlling social class. Was this direction
of social reform really capable of effecting any substantial change?
Nay by lessening the selective deathrate may it not have contributed

NURTURE AND NATURE 7

to emphasizing the very evils it was intended to lessen? These are
the problems which occur to the eugenist and call for investigation
and if possible settlement.

The outlines of our enquiry are not hard to sum up. We know
from a variety of investigations that the correlation between physical
and mental characters in parent and offspring is about -45 to -50.
The first question to be asked is : Are the physical and mental
characters of children correlated with their or their parents' environ-
ment to a higher, an equal, or a lesser extent? But even the answer
to this first question will not finally solve the relative intensity of the
environmental and heredity factors. We have still to ask how far the
parents' physical and mental characters are productive of the observed
environment. It is conceivable that the relation between children's
physique, for example, and parental occupation is an indirect result
of the inheritance of physique and a correlation between parents'
physique and their occupation. In other words, what we are
attributing to environment may be a secondary influence of heredity
itself. A weakling may have no option but to follow an unhealthy
trade, a man is a tailor or shoemaker, because he has not the physique
needful for smith or navvy. His offspring may be physically inferior
because he is a weakling and not because he follows an unhealthy
trade. Clearly, to solve our problem, we must know if there be any
correlation between the same character in the parent as we are
observing in the child and the environment we are correlating with
the child's character. Unfortunately data enabling us to determine
the relationship of any mental or physical character of the parent
with the environment which is supposed to influence the child is
rarely forthcoming. We have, however, some information which may
help us indirectly in our consideration of this part of the enquiry.

After this preliminary warning as to the difficulties of the problem
of heredity and environment we propose to consider what evidence is
available for determining the relative intensity of nature and nurture
in the case of man. The material at our disposal may be summed
up as follows :

I. The Report on the Physical Condition of fourteen hundred
school children in the City of Edinburgh, with some account of their
homes and surroundings, issued by the Charity Organization Society.
Information is given as to the number of children, living and dead,
the number of rooms, rent, father's work and wages, mother's work,
work and wages of other members of the family, and age, height,

8 1HK RELATIVE STRENGTH OF

,t. int. I;. . oiulition "i treth. eyes, ears, n'»e and throat

of the school childn n, and an account of tin- home in « ach case;
information i> pven a> to tin- drinking of the parents, cleanliness of
the home, morality "1 the parents, &C.

For thi> splrmlid piece of research school teachers, doctors, and
workers belonging to the C.O.S. have combined; information has
been sought from many sources and infinite pains must have been
taken. It is a pleasure to work from such a report, but one longs at
the same time that the numbers wen ten times as great, and that
the C.O.S. might obtain a far larger band of workers who would
investigate in like manner the condition of fourteen thousand instead
of fourteen hundred school children well distributed over the whole
of Great Britain. The machinery now exists for the medical exami-
nation of school children, but from the point of view of the eugeni>t
it will be of comparatively little value for measuring the effects of
environment unless the children are followed into their homes, and
information is collected and recorded as has been done by the C.O.S.
in Edinburgh.

II. A Record of Measurements &c. of children attending in 1905
the public schools in Glasgow, lent to us by the Scottish Education
Department. We have from this source information about 72,857
school children; among other facts recorded are the age, height and
weight of the children and the occupation of the father and employ-
ment of the mother.

III. An account of the children in the special schools in Manchester,
kindly provided for us by Miss Dendy. Among other information the
health and intelligence of the brothers and sisters of the defective
children under consideration are given, and the extent of drinking and
the health of the parents.

The main points to which we have turned our attention up to the
present time are:

(1) The influence of the employment of mothers on the physique
of their children.

(2) The influence of the occupation of the father on the physique
of the children.

(3) The influence of drink in the parents on the height, weight,
general health and intelligence of the children.

(4) The influence of overcrowding, bad economic condition of the
home, moral and physical condition of the parents on the intelligence,
eyesight, glands, and hearing of the children.

NURTURE AND NATURE 9

In our investigation of the first and third points we are fortunate
enough to possess in each case two distinct sets of statistics, so I feel
I can speak with greater confidence of the results.

In the other two cases we are dependent on one set of statistics
in each case ; for the influence of the occupation of the father we have
the Glasgow School Board statistics, and for the effect of environment
on intelligence, eyesight, &c. we have the Report of the Edinburgh
C.O.S.

We will first examine the results that have been found for the
correlation between the employment of mothers .and the height and
weight of their sons ; and we must ask the reader to keep in mind
our problem, namely: Are the physical and mental characters of
children correlated with their or their parents' environment to a
higher, an equal, or a lesser extent than they are correlated with their
parents' physical and mental characters?

The larger series of statistics from Glasgow is not really so
satisfactory as one might at first sight suppose, as it appears that
comparatively few married women in Glasgow have other than the
customary domestic work of their homes.

Dr Leslie Mackenzie, in his report on these statistics, divides the
schools into four groups according to the districts in which the schools
are situated, and we have kept to these groups in working out the
correlation coefficients — but have also worked the coefficients for all
schools taken together.

Group A comprises schools in the poorest districts of the city.

Group B comprises schools in poor districts.

Group C comprises schools in districts of a better class.

Group D comprises schools in districts of a still higher class, and
includes four out of five Higher Grade Schools.

What then is the effect of the employment of the mother on the
height and weight of her sons in Glasgow and in Edinburgh?

In the Introductory Note to the C.O.S. Report of the Edinburgh
children it is stated that the children come from a school in the poorest
part of the city, but that this school also has an admixture of the
children of the comfortable and respectable working class.

Table I gives the correlation coefficients for the effect of the
employment of mothers on the weight and height of their sons, and
we notice :

(i) That the employment of mothers is correlated to a small
extent with lesser weight of their sons, the coefficient being -12 for

10

THE RELATIVE STRENGTH OF

.ill x-houls in Glasgow and -14 for Edinburgh, and we may say that
th-- tv-ult^ obtained from the two distinct sets of statistics are well
in agreement.

TABLE I.

Partial Correlation Coefficients of Employment of Mother and
Physique of Children for Constant Age of Child.

Influence of the Employment of Mothers on the
Physique of her children

Sons

Daughters

Influence of
Heredity

Weight

Height

Weight

Height

Glasgow

\

8 |B

•° 2 c

C/>OD

All Glasgow

•06

•12

•ii
•ii

•ii

•15
•19
•17

•03
•ii
•ii
•09

•08
16
•19

•12

Stature: Mother and")
Son / '49

,, Mother and)
Daughter/

Eye-colour: Parental -49
,, Grand )
Parental/

,, Great ~|
Grand - -19
Parental J

•12

•20

•10

•O2

Edinburgh

•14

•09

•02

•oo

(2) That the employment of mothers is also correlated with lesser
height of sons, and that this correlation is somewhat larger than that
found for weight in Glasgow. The effect of the employment of mothers
on the height of sons in Edinburgh is measured by the coefficient -09,
which is distinctly lower than that found for all Glasgow. The
Edinburgh schools are, however, probably closest in grade to the
Glasgow Group A.

(3) That there is a distinct difference in the correlation coefficients
in the different school groups in Glasgow. This is not what we should
expect if the population were homogeneous, and leads one to hold
that some special influence must be at work which has lowered the
correlation in the worst school group or raised it in the better school
groups. Taking into consideration the value found for the Edinburgh
schools we should be inclined to think that some racial factor we have
not yet measured has artificially raised the coefficient in the better
school groups.

NURTURE AND NATURE II

We must next compare the influence of the employment of the
mother on the physique of her son with the hereditary influence of
the actual height of the mother on the height of her son. This
hereditary influence is given in Table I for the sake of comparison.
The effect of inheritance of stature of mother and son is measured by
the coefficient -49. The coefficients for eye-colour for the parental,
grandparental, and great grandparental relationships are also given,
and it will be noticed that the only value at all comparable with the
environmental values is that found for great grandparents, which is
about equal to that found for the effect of mothers' employment on
sons' height in Glasgow, but distinctly higher than the other correlation
coefficients. We cannot help being struck by the comparative unim-
portance of, at any rate, one environmental factor as compared with
the heredity factor.

We will add one further remark on the differences we have found
in the correlations between employment of the mother and height of
her sons in Glasgow. In Glasgow the values found range from •!
to -2. At present we have no final explanation beyond heterogeneity
for this result. But we would mention a fact noted by Dr Tocher in
a paper on the pigmentation survey of Scotland, published in Biometrika,
vol. vi, pp. 129 — 235. In that paper Dr Tocher calls attention to the
large foreign element in Glasgow as judged by the names of the children
and confirmed by their hair and eye-colour. There are a large number
of Jews, especially in Groups B and C, and a fairly large proportion
of Irish in the schools in Group D. Can mixture of races account for
the higher correlations in some of the Glasgow groups1?

Future investigations will show whether the influence of the
mother's employment on the height of her sons is greater than (as
apparently in Glasgow) or less than (as apparently in Edinburgh) the
influence of her employment on the weight, and whether the influence
of employment on height is represented by the higher or lower
correlation coefficient, by -2 or by -i. At present we must content
ourselves with saying that the connection between the employment
of the mother and the weight of her sons is about -12, and that the
connection between the employment of the mother and the height of
her sons is between •! and -2. Taking into consideration the result

1 If we compare the mean height and weight of the Edinburgh boys with those
of boys in the different school groups at Glasgow, we find that they lie between the
Glasgow groups A and B, just where the correlation coefficients fall. Thus the
Edinburgh and Glasgow values may not be really contradictory.

12 '1HI. KKI.V1IYK MKICN'.TH OF

found in the worst school group we ;ire inclined to think that the

foe i- the more eon

n it tli. higher value |>roves to be the true one, -20 is still a low
!. it ion, and when we remember that the value for the correlation
it-iit of heredity between height of mother and son is -49 vv<
that employment of the mother has far less influence than heredity
on the physique of the son.

We now turn to the correlation between the employment of the
mother and the height and weight of her daughters. In the Edinburgh
schools we see that there is practically no connection between employ-
ment of the mother and the physique of her daughters ; the correlation
coefficients are -02 and -oo. When we examine the results for Glasgow
we see that the effect of the mother's employment on the weight of
her daughters is very uniform in the three school Groups B, C, D,
namely -n, -n and -09, and is about the same as the effect on the
weight of the sons ; it is insignificant in the worst class, Group D, of
schools. The height results are again more irregular, varying from -08
to --19, but when we take the data for all Glasgow the correlation is
insensible; it would seem that the height of girls is less affected by
the mother's employment than the height of boys, and the result for
all Glasgow agrees with the result found for Edinburgh.

The results we have reached show on the whole that the employment
of the mother has even less influence on the physique of her daughters
than on the physique of her sons.

Employment of the mothers then is associated to a small extent
with lighter weight and shorter stature of their sons, but we are not
sure that even the small association we find is necessarily a true
measure of cause and effect. There are secondary obscuring effects
of heredity. The weaker and more delicate mothers may not seek
external employment, and naturally they would have weaker children.
This lowers the association. Furthermore, inferior physique in the
fathers may be a reason for the necessity of the mother's work. This
would raise the association. We know from the C.O.S. report that
the mother's work is associated with a lower rate of wage of the
father — the average wage of the husband of an unemployed woman
is 265. 2d. and of an employed woman is 22s. nd. It seems reasonable
to suppose that a man of inferior physique may have to take a lower
wage, and thus it is his wife who will be obliged to work. Now if the
husbands of these working women are physically weaker than other
husbands we should expect their sons to be shorter and lighter than

NURTURE AND NATURE 13

other persons' sons, whether the mother worked or not. We want
to know the height and weight of the husbands of employed and
unemployed mothers before we can be sure what proportion of the
correlation we see between the mother's employment and son's height
and weight is due actually to the mother's work and what is really
due to the less weight and height of the fathers. We quite realize
that we may find low wages do not depend on the physique of a
workman at all, but, if they do, then some of the apparent result on
the child of the mother's work will probably be due to the inferior
physique of her husband.

There is another point in this connection we should like to mention.
If employment of mothers has a bad effect on their sons, either directly
before birth or indirectly owing to less home care, there ought to be
less bad effect when the mother only works out part of the day than
when she works out the whole of the day ; but this is not 'the case,
at least it is not the case in Edinburgh in those families investigated
by the C.O.S. From this report we are able to divide the mothers
into four classes : (i) those who do no work beyond their domestic
duties, (2) those who work in their homes, (3) those who work out
part of their time, (4) those who work out the whole of their time.

Weight of Boys at a Constant Age.

Db

\

O J

\

54

\

x

•<:

L> i

CO

None At home Part out

Employment of Mothers.

Whole out

The diagram shows the average weight of the boys for a constant age,
i.e. the mean age 9-6 years. The mean weight of the boys is practically
identical whether the mother works out all or only part of the time.

When the mother works "in" the weight of the boys is Iqwest,
but there are only 31 cases, so we can lay no stress on this result ;

I UK KK1.A1IYK >TRENGTH OF

the mnthiT works out part of UK- time in 129, and the whole time in
79 cases, which are more reasonable numbers on which to base a
conclusion.

\\. must now turn to the second problem under consideration,
the influence of the occupation of the father on the physique of
hi> sons. Unhealthy trades have been considered largely responsible
f<>r physical unfitness in tin- children. After some consideration we
decided to measure the unhealthiness of a trade by the annual
dt uthrate per 1000 living in the different trades between the a.g<
and 45. These numbers were taken from the Registrar General's
returns for Scotland, and range from 3 per thousand for mini>i«-i>
and gamekeepers to 18 per thousand among barmen, cellarmen and
general labourers. The classification is not by any means perfect but,
though rough, it gives a fairly satisfactory estimate of the healthiness
or unhealthiness of a trade.

TABLE II.

Low Mortality Rate of Fathers and Greater Height
and Weight of Sons.

Influence of Heredity

Schools

Weight

Height

Stature

A
B
C
D

•09
•04
•03

•01

•12

•07

•02

Father and Son -51
Father and Daughter -51

All

•09

•12

Here the results are very regular and practically the same for
height and weight. There is a certain very slight association between
the unhealthy trade of a father, as measured by the mortality rate,
and inferior physique in his children in the worst school group, but it
becomes practically nil in the better school groups. We must again
notice the low value of the correlation coefficient when we compare it
with the value found for the correlation between height of father and
son. . And in this case also we must point out that we do not know
how much of this apparent association is due indirectly to heredity,

NURTURE AND NATURE 15

and to the probable fact that on the whole weaklings are driven into
less desirable occupations. A man who is physically unfit has to work
for any employer who will take him, and we shall probably frequently
find him doing casual labour or working in a factory where little bodily
strength is required. Here again we need the average height and
weight of the fathers of all these sons to see whether fathers in a trade
with a low mortality rate are taller and heavier than fathers in a
trade with a high mortality rate. Is the casual labourer with a
mortality rate of 18 per thousand lighter and shorter than the carter
with a mortality rate of 8 per thousand? the factory hand with a
mortality rate of 13 per thousand than the blacksmith with a mortality
rate of 7 per thousand? If so, children of the casual labourer or the
factory hand will be lighter and shorter than the children of the carter
or blacksmith, quite apart from any direct influence of unhealthy
employment of the parent.

In the Appendix to the Report of the Royal Commission on
Physical Training a table is given showing the average stature of men
at all ages, divided into four classes according to their occupations.
These do not correspond exactly with our mortality rates, but they
do so roughly and are therefore of interest.

Class I includes farmers, clergy, lawyers, doctors, teachers, archi-
tects, &c. ; all these have a mortality rate of between 3 and 6 per 1000
and come in our mortality groups 3, 4, 5 and 6, except doctors, who
are in our group 9.

Class II includes law clerks, shop-keepers, dealers in drugs, books,
food, drinks, &c., and the majority come in our mortality groups 7
and 8.

Class III includes gardeners, railway and quarry workers, sailors,
miners, &c. ; these also are mostly included in the groups with mortality
rates 7 and 8 but they vary more than Class II. The mortality rates
of Classes II and III are, however, very similar.

Class IV includes workers in metal and stone, engravers, printers,
&c. who are mostly included in the groups with mortality rates 9
and 10.

The mortality rates then roughly correspond with these classes —
that is to say —

Class I includes occupations with the lowest mortality rates 3
and 6.

Classes II and III include occupations with higher mortality rates
7 and 8.

i6

THE RELATIVE STRENGTH OF

IV unhides occupation- with still hi^lu-r rates 9 and 10.
The next table gives the mean height and weight for each class.

TABLE III.

Class I

Class II

Class 1 1 1

Class IV

Height
Weight

69-4 ins.
168-4 Ibs.

67-9
157-0

67-5
157-5

66-7
M7'9

We see then that men occupied in a trade with a lower mortality
rate are heavier and taller than men in a trade with a higher mortality
rate, which suggests that the very small correlation we have found
bet ween unhealthiness of the father's trade and the height and weight
of his sons is not all due to the unhealthiness of the trade, but is due
at least, in part, to the inferior physique of the father. In fact \ve
have very little hesitation in saying that unhealthiness of a trade of
a father has small influence on the physique of his sons, and that it
is because men of inferior physique are forced into worse trades that
we see this apparent slight connection.

\Ye now turn to the question of drink and shall measure its effect
on the height, weight, general health and intelligence of the children.
More disastrous results have been attributed to drink than to any
other social evil — we might almost say than to all other social evils
taken together. Since we have been working at this subject we have
read carefully many speeches, as reported in the papers, on the subject
of drink, and confess we have been somewhat astonished at the
statements occasionally made.

For example: a well-known man is reported to have said in the
course of a speech enumerating the various evils resulting from drink
that 10 % of the children of drunkards are tuberculous, but he made
no mention of the fact that practically 10 % of the general population
suffer at one or another period from tuberculosis.

We do not say that alcoholism and tuberculosis are not associated,
it is quite possible they may be, but a statement such as that we have
quoted proves nothing of the sort, and is entirely misleading unless
we are told at the same time the percentage of the tuberculous in the
same class of the general population, which in this case was not stated.

We have two series of statistics for studying the question of the

NURTURE AND NATURE

effect of drink, the Edinburgh Charity Organization Society Report
and the statistics provided for us by Miss Dendy.

We will consider first the results obtained from the Edinburgh
statistics, which are given in the next table.

TABLE IV.

Drink and

Drink and

Drink and

Weight

Height

Intelligence

Boys

•0.5

•04

-•03

Girls

•08

•09

•oo

Here we see that drink in the parents has no effect on the intel-
ligence of boys or girls, and practically none on the height and weight
of boys. In the case of girls there is a very slight correlation between
drinking of the parents and lower weight and- height in the daughters,
but considering that the probable error of these results is about -03
they are only just significant. Is this slight difference between the
effect of drink on boys and girls due to the possibility that when the
mother drinks the girls have to look after the home, whereas the
responsibilities of the boys are not increased? But, whatever the
reason, we would emphasize the fact that a correlation of -09 is of
very little importance as compared with a correlation of -50 due to
heredity.

Before considering these results further we may turn to the
statistics obtained from the special schools in Manchester. Here we

TABLE V.

Father

Mother

Intemperance
and

Intemperance
and

Intemperance
and

Intemperance
and

Health

Intelligence

Health

Intelligence

Sons
Daughters

-•06
-•04

-•ii
- -02

-•07
-•03

•01

-•08

cannot find the correlation coefficients between drink and actual
height and weight of the children, but we can find the coefficient of
correlation between drink and general health, and between drink and

l8 1HI KMAIIVK STRENGTH OF

Iii this cast- \\v lia\v \v«.rkcd out the influence of the
drinking of (i) tin- tathn, u) tin- mother, <>n tin- health and intelligence
of the chilcli

Urn- a?Min we see that drink has practically no influence on the
general health and intelligence of boys and girls, and the little influence
it has is in favour of the children of drinking parents, they are healthier
and more intelligent. These results are certainly startling and rather
upset one's preconceived ideas, but it is perhaps a consolation that to
the obvious and visible miseries of the children arising from drink,
lowered intelligence and physique are not added.

But before asserting that intemperance of the parents has prac-
tically no effect on the physique and intelligence of the children, there
is a point to be considered which we mentioned in the first part of this
paper. What is the status of the drinking workman? Is the drinker
on an average the abler man and of finer physique ? If so, his children
should show greater ability and better physique than the children of
the non-drinking parent, and further, if the abler workman get higher
wages, and thus notwithstanding drink, the food at home has been of
better quality, we should expect his children to be better developed
physically than those of the non-drinker.

As before the only estimate we have at present of the intelligence
and physique of a workman is the wage he earns. A high wage on an
average will mean a stronger and more efficient workman. We want
to discover therefore whether drink and good wage go together to any
large extent. If they do, then drinking fathers should have stronger
and more intelligent children than non-drinking fathers. But unless
there is a fairly well marked correlation it will not be sufficient to
affect greatly our results.

We divided parents into three classes: (i) both parents drink,
(2) one parent drinks, (3) neither parent drinks, and the wages into
four groups: (i) under i8s. a week, (2) i8s. — 245. inclusive, (3) 265.
to 345., (4) 365. and over 365. The correlation coefficient found by
the fourfold method between the drinking of one or both parents and
a high wage is -03, which means that there is practically no connection
between drinking of the parents and a high wage. The means show
the same thing — the mean wage when both parents drink is 245. 8d.,
when one parent drinks is 255. 6d., and when neither parent drinks
is 255. $d. — there is a slightly lower wage when both parents drink
and a very slightly higher when one parent drinks, but we cannot
attach any importance to a difference of id. We tried to find from

NURTURE AND NATURE 19

the Manchester special schools what was the connection between
wages and drink, but the wages are very seldom given when the
father drinks. From the few cases where they are given I found the
average wage of a drinking father to be 235. yd., and of a non-drinking
father to be 235. 4^., i.e. a slightly higher wage for the drinking father
in Manchester, but the numbers are too few for the results to be
considered of any importance. If wage then is an estimate of ability
and physique we may say that the ability and physique of the
drinking workman is about the average, and we can state with greater
confidence that the "well-known fact" that drinking has a bad effect
on the physique and intelligence of the children has yet to be proved.

The whole subject of the influence of environment, owing to its
complexity, is a fascinating one, partly because we are only just
beginning to apply modern statistical methods to this side of eugenics,
and the results we obtain are often very unexpected, perhaps we may
say wholly contrary to current belief.

The fact that our conclusions as to the comparative unimportance of
the environment factors of employment of mothers, occupation of fathers
and drink of the parents are contrary to preconceived opinions, makes
more work and more statistics on this side of eugenics most necessary.

We now come to the last point under consideration — the effect of
the home environment on the intelligence, eyesight, glands and hearing
of the children. The statistics being taken from the C.O.S. Report.
The home environment factors we have chosen are:

(a) The number of people per room of the home. Unfortunately
we do not know the, size of the rooms, but in a general way we have
a measure of the space in the home by considering the number of
people per room.

(b) Economic condition of the home. We divided the homes into
"good" and "bad" from a consideration of the information given in
the C.O.S. "Summary of Facts," and took into account also the
cleanliness of the children, regularity of attendance, &c.

(c) Physical condition of the parents. The Report states if the
parents are broken down in health or suffer from tuberculosis, &c.,
persistent alcoholism was included in bad physique.

(d) Moral condition of the parents. The chief difficulty was to
decide how far alcoholism is to be treated as a moral complaint.

Generally speaking we placed the heavy drinker in the category of
moral failure when the drinking was accompanied by "conviction for
brutal assault on wife," "house a regular brothel," frequent convictions

20

THE RELATIVE STRENGTH OF

drunk and disorderly. Drinking is so prevalent that it is
impossible to take it as determining in itself bad moral conditions.
"Man a good workman, but goes on spree from time to time, is in
two thrift clubs and attends church," or "old soldier and widower
who takes a nip now and then, but is good to his girls — very nice,
tidy, clean people," can hardly suffice for placing the described in the
category of moral failure. On the other hand "Very dirty, untidy
home... Man teetotal, keeps well at his work... China and clothes lying
piled about room, thick with dust; air very bad. Children sickly
(eldest imbecile) ; wife a slattern," seems to be a case where there is
a moral deficiency likely to affect the condition of the children.

In each case we have had to trust to personal judgment, and
though we think we should put the bulk of the cases in the same
classes if we went through the book a second time, it is possible that
in some doubtful cases our judgment would not be the same. The
negative sign when it occurs means that the worse environment is
associated with the better condition in the children.

Let us consider the results collected in Table VI.

TABLE VI.

Effect of Home Environment on Sight, Intelligence,
Glands, Hearing.

Refraction

Keenness of
Vision

Intelligence

Boys

Girls

Boys

Girls

Boys

Girls

No. of people per room

-•00

+ •03
- -oo

-•14

-•15

-•12
+ 'OO
-•05

-•06
-•oo

-•00

-•10

-•14
-•OI

+ •00
+ •06

•02
•10

-•04

-•07

•04
•16
•06
•03

Good Economic conditions
Good Physical condition of Parents
Good Moral condition of Parents

Glands

Keenness of
Hearing

Diseases of
Ear

Boys

Girls

Boys

Girls

Boys

Girls

No. of people per room

•02
•06
•°5

•02

-•01

•03
•06

•02

—07
•05
•13
•10

•00

•08
•oo

•06

•02
•02
-•01

•08

•07
•14

•01

•08

Good Economic conditions
Good Physical condition of Parents
Good Moral condition of Parents

NURTURE AND NATURE 21

The eyesight results are already published in a memoir by
Amy Barrington and Karl Pearson. Normal vision is on the whole
slightly associated with overcrowding, bad economic conditions and
morally defective parentage. Can it be that these home conditions
keep the children in the streets, and so relatively away from bad
environment and in relatively fresher air? Whatever may be the
cause the Edinburgh statistics show that the effect of home influence
is not one-tenth that of heredity, and what exists, if it be appreciable
at all, is in the opposite direction from what we should have antici-
pated. Intelligence is slightly associated with few people per room
and with a good economic condition in the home for boys and girls,
though for boys it is scarcely appreciable; it is also associated for
girls very slightly with good physical and moral condition in the
parents, but for boys we get better intelligence in the children
associated with bad physical and moral condition in the parents.

Glands and hearing tell the same tale. There is only a slight
connection between the presence of swollen glands or bad hearing and
bad environmental conditions. All the coefficients are small and
irregular, some scarcely appreciable, but in most cases they are
positive, i.e. better physique goes with better environment, though
it is but in a weak degree1.

Such a table as that on p. 20, indicating in many directions the
relative insignificance of nurture in influencing man's welfare, gives
us reason to pause when we consider the methods of modern social
reform.

The question of the respective influences of heredity and environ-
ment is becoming one of vital importance. It seems only too true at
the present time that the physically and mentally weaker stocks are
reproducing themselves at a greater rate than those of sounder
physique and intelligence.

This, unchanged, must mean that the average physique and ability
of our nation as a whole will decline and must decline unless we can
prove that by a better environment we can raise the level of the
community. So far as our investigations have gone at present they
show clearly the small influence of environment ; work of the mother,

1 Heredity as before plays far the larger part. The measure of the inheritance
of eyesight is equal to that found for other physical characters and the same is true
of intelligence (-49)- Recent work of the Laboratory (British Medical Journal, July 20,
1909 : see Appendix to this Lecture) shows that environmental conditions can hardly
be the chief source of eye-disease. The correlations are low and more often negative
than in the case of ear-disease.

22 THE RELATIVE STRENGTH OF NURTURE AND NATURE

an unhealthy trade of the father and the drinking of the pan-nts
seem to have very little influence on the physique of the children.
r. -wding, bad economic conditions, bad physical and moral
conditions of the parents have practically no effect on the intelligence,
eyesight, glands and hearing of the children. It is possible that better
measurements of environmental characters than we yet possess may
show more correlation, and it is also possible that other characters
may prove more influential. But so far as our researches reach we
think we have shown that it is quite easy to demonstrate a large
hereditary factor, and it is not at all an easy thing to show that any
of the environmental factors we have measured up to the present time
have any important effect on the children. Not only are the correla-
tions low but we have reason to think that such small correlations as
exist may be secondary results of racial or hereditary influences.

Practically all social legislation has been based on the assumption
that better environment meant race progress, whereas the link between
the two is probably that a genuine race progress will result in a better
environment. The views of philanthropists and of those who insist
that the race can be substantially bettered by changed environment
appeal to our sympathies, but these reformers have yet to prove their
creed. As far as our investigations have yet gone they show that
improvement in social conditions will not compensate for a bad
hereditary influence ; the problem of physical and mental degeneration
cannot be solved by preventing mothers from working, by closing
public-houses, or by erecting model dwellings. The only way to keep
a nation strong mentally and physically is to see to it that each new
generation is derived chiefly from the fitter members of the generation
before.

APPENDIX A1

HOME CONDITIONS AND EYESIGHT
By KARL PEARSON, F.R.S.

It may not be without interest, in view of recent criticisms of a memoir
by Miss Barrington and myself2, to publish some further determinations of the
relation of home conditions to sight.

Before doing so it will be well to point out one or two important considera-
tions which I think have considerable bearing on the manner in which statistics
ought to be collected, having regard to the medical inspection of school children,
which is now becoming universal.

The first point I would insist upon is that careful distinction must be made
between (a) home environment and (b) school environment.

Our% paper dealt chiefly with the influence of home environment on both
refraction and acuity of vision. Now, as far as I am aware, the only material
hitherto available for testing the influence of the actual home conditions on
the presence of eye disease, on the goodness or badness of vision, or on
shortsightedness, is that provided by the report of the Edinburgh Charity
Organization Committee, where we have for the first time the sociological report
on the home conditions placed alongside the ophthalmological report on the
child's eyes, and accompanied in each case by the age of the child. I do not
see how it is possible without such information to draw conclusions as to
whether the home conditions do or do not affect sight. I am quite prepared
to be shewn that the conditions in Edinburgh are exceptional, but the proof
can only be given when the children who are reported on elsewhere are followed
up into their- homes, and the state of these is recorded, as in the Edinburgh
investigation.

Further, the Edinburgh material was of special value, because there is not
the same extraordinary mixture of racial types in that city which is to be
found in Glasgow or London. We are told, for example, that Russian Jews
have in London a very high percentage of eye defect. Any one who has studied
the copious statistics of Randal must be convinced that the degree and extent
of myopia is markedly a racial character3. Those who have investigated any

1 Reprinted by kind permission from the British Medical Journal, July 17, 1909.

* Eugenics Laboratory Memoirs, No. v, "A First Study of Vision and of the Relative
Influence of Heredity and Environment on Sight." Cambridge University Press.

* The Jews, like the Germans, are largely brachycephalic, and the increasing
brachycephaly of the town populations is a point not without suggestiveness for
changes in eyesight.

24 THE RELATIVE STRENGTH OF

local group in this country know that in anthropometric characters it is usually
significantly ditU-ri-ntiatrd from any other local f^roup. The population of our
suburbs is usually more sedentary than the population in the working class
districts of the town itself and less mixed.

I know, as a matter of fact, that the cephalic index of school children varies
very sensibly from one district of London to a second. On this ground alone
it is not possible, without control measurements, to assert that, because the
percentage of eye defect varies from Whitechapel to Hampstead, the result is
due to home environment. It may be so, but a mere statement of percentages
in different districts without (a) age distributions, (6) racial proportions, and
(c) percentages of defective parents in each district, will not convince those who
desire logical statistical proof before forming any conclusion. It is well known
that the defective parents also gravitate to the worst districts, and we may
expect the defective children there also. For these reasons the work at
Edinburgh was especially valuable. It gave age distribution, it dealt with a
racially fairly homogeneous material, and it followed up the children into their
homes and told us something, if not all we might desire to know, about their
parents. I see no other way in which a real solution could be obtained for
London. An individual school or two must have all — not only the defective —
children examined, and the eyesight report must be accompanied by a
sociological report.

A report such as that of Dr A. Hugh Thompson on the London school
children, in which the numbers at each age of both the normal and the defective
children are not given, cannot help the statistician in the least to arrive at
definite results. Nor, unless the children are followed into their own homes,
is it even possible to say how far bad food, bad air, or parental neglect may
account for the presence of any disease. There is always the diathesis as a
contributory source, and the fact that all the children in the same family may
not be attacked, shows that the variability within the family also plays its part.

Again my critics will, I hope, pardon me if I say that I am not convinced
when I am asked whether this or that "is not a fact." For example, whether
it is not a fact that in the better homes the children are more studious, or that
the more respectable parents keep their children at home. Either may be the
truth, but what we want. are actual numerical measures of the effect of these
supposed causes in invalidating the apparently small influence of the environ-
mental as compared with the hereditary factor.

What we want are more data before we conclude that results are paradoxes
because of such or such an explanation being "a fact" or "a matter of common
experience."

Let us attempt to get any information we can on the problem of whether
the better class parents keep their children at home, and so their offspring are
more studious and suffer more from myopia1. Now, I know of no statistics
which at all touch this point of the studious character of the offspring of better

1 The reader of our memoir will remember that our conclusion was, not that the
better homes produced more myopia, but that there was no marked relationship
between bad homes and defective sight.

NURTURE AND NATURE

class parents, but I suppose the argument to be that the children are kept more
at close work and that this damages their sight. Now, it would be possible to
compare the sight of children of the same age who came to school at different
ages originally, and thus find out whether those who began to read earlier have
markedly worse sight. It would be very desirable to collect statistics de novo
on this point. The only material that I know of bearing on the subject is that
used by Miss Harrington and myself, and provided by Cohn, who gives the
correlation tables of sight and years of school life, and sight and age. We have
shown that the relationship is more intimate between sight and age than between
sight and years of school life, and that, considering the high correlation of years
of school life and years of age, the latter is most probably only a derived result
of the former. Now, we have recently re-worked Cohn's data and discovered
the correlation between the ages at which the child began to read and the
degree of myopia at a constant age. The partial correlation coefficient is :

0-04, if age and length of school life have a correlation of 0-8
0-13, if age and length of school life have a correlation of 0-9.

Now it will, perhaps, be said that the latter is a quite sensible correlation, but
compare it with the heredity correlations of 0-40 to 0-60, and we see that
environment is quite overmastered by heredity. Furthermore, the correlation
has the wrong sense, the later the child went to school, the worse at a given age is the
myopia. I have little doubt that the correlation in this case is sensibly zero,
or the age at starting reading has little influence on the result. Cohn's statistics
are, of course, not final, but they are very suggestive, because they have hitherto
been supposed to prove the exact reverse of what is really extractable from
them.

I now turn to the question of the effect of good or bad home environment
on disease of the eye. I will consider first corneal nebula and illustrate by
some of the percentages :

Percentages of Corneal Nebula.

Feature of the Home Estimated

Boys

Girls

Good

Bad

Good

Bad

Economic condition

3-6
2-5
3-8
3'5

2-3
3'3
2-5

2'5

3'9
2-5
3'i
2-5

2-2

3'4

2-8

3'5

Overcrowding

Physical state of parents

1 Moral state of parents

Means

3'35

2-65

3-00

3-00

1 This category should £ priori be considered the most important one in this
matter. It marks essentially neglect of the children. Where the home is practically
a brothel or the parents habitual drunkards there can be no proper cleanliness or care
of offspring.

26

111! K! .1. AII\i: STRENGTH OF

Now it -. •-.!.. nlniiirily difficult to lay any itmi ->n n Milts whose

differences are of this order. A bad moral state of parents would be interpreted
to mean less comeal m-bula for the boys and more for tin- nirls. the percentages
being just reversed. Good economic conditions in both boys and girls would
mean 1-5 per cent, more of the disease, and so would the good physical
condition of the parmts. In such circumstances what stress can be laid on
the 0-8 per nut difference between crowded and non-crowded homes? The
only safe conclusion to be drawn is that in Edinburgh at least there is no
marked relationship between corneal nebula and bad home environment. The
curious point is that if we take all diseases of the eye and eyelid, we appear
to get a slightly more definite, but still a quite negligible relation between
disease and home environment. A difficulty is whether to include strabismus
or not; the correlations have been worked out by Ethel M. Klderron, of the
Eugenics Laboratory, including and excluding "squint." In -each a positive
corr.-laticn means that there is more eye disease with the worse home condition.

Percentages of all Eye Diseases.

Bo

ys

Gi

rls

Feature of the Home Estimated

With
Squint

With-
out

With
Squint

With-
out

Economic conditions

O-IO

O-I4

— o-o^

O-O2

Overcrowding

o-os

0-o8

0-04

O-I2

Physical state of parents

— 0-02

— O-O2

— O-IO

— O-IS

Moral state of parents

O-O4.

O'O3

— O'OI

O-OI

Now, it is quite impossible to lay any stress on these results. Six ont of
the 16 show less disease with worse home conditions, and if we have positive
correlations of o-io and 0-14, we have negatives of - o-io and - 0-15.
Further, no correlation of this kind lying under about 0-25 can be really
considered to have definite importance. How very different these values are
from the values running from 0-4 to 0-6 found in our memoir for the inheritance
factor in eyesight !

It is worth while putting on record here the quantitative contributions of
refraction and diseases of the eye in producing defective vision. The refraction
results are taken from our memoir; the disease results have been recently
worked out by Ethel M. Elderton.

Keenness of Vision
and Refraction
(Contingency)

Keenness of Vision and Disease

Squint included

Without squint

Boys
Girls

0-58
°'54

0-36
0-50

' 0-21

0-32

NURTURE AND NATURE 27

It will be seen that, if we exclude strabismus, the diseases of eye and eyelid
do not produce half the effect on vision that refraction does.

Now I do not intend to argue for a moment that the Edinburgh data
provide the final word in these matters, but they form the first solid attempt
to provide information as to the influence of home environment. Neither the
school medical officer nor, as a rule, the teacher can follow the children into
their homes. All that has hitherto been done has been more or less plausible
guesswork as to the relation of home environment to the health and intelligence
of the child. I shall not be surprised to find, when further data are available,
that the nation has for years been putting its money on "Environment" when
"Heredity" wins in a canter. To say this is not to discourage all attempts
to better defective sight in the schools or to check the intensification of myopia.
But it would be foolish — in face of our increasing degeneracy — to neglect the
possibility that only a complete change of our methods for social improvement
can in the long run hinder the spread of defects. Better qualities, such as
keen and strong vision, are no longer an absolute requisite for survival, and
defectiveness, instead of meeting the stern judgment of Nature, is by govern-
mental and charitable agencies supported to multiply its kind. The curative
art in its tenderness for the individual may be disastrous for the race, unless it
realizes fully the relative biological importance of heredity and environment.

APPENDIX B
CLEANLINESS AND VISION

Through the courtesy of Dr F. E. Rock, School Medical Officer of the
Edmonton Education Committee, I have received data connecting the age,
acuity of vision, and cleanliness of body and clothing of 953 Edmonton
children. Statistically the data are somewhat erratic as there has been an
age selection in testing the eyesight and clearly home environment, if we are
to measure that by cleanliness of body and clothing, also produces an age
selection. Dr Rock, as many another medical officer, not unnaturally, takes
this cleanliness as the only measure available of home environment. Remem-
bering what children are and will do, I must submit that it can never replace
actually study of the homes and the parents. Taking it, however, as a measure
of home environment, Dr Alice Lee provides me with the following correlations :

Age and acuity of vision - -037

(i.e. the vision is slightly worse for the elder children).

Age and home environment + -029

(i.e. the cleanliness is slightly better with the elder children).

Acuity of vision and home environment + -072.

28 THE RELATIVE STRENGTH OF NURTURE AND NATURE

Hence the relation between goodness of vision and home environment for
a constant age = -073, or, the relationship between home environment, as
measured by cleanliness of body and clothing, and goodness of vision is just
about one-seventh of the intensity of the relationship between goodness of vision
in parent and child. Nor are we sure that the home environment is the source
of this small correlation ; for not only do the statistics indicate a marked
selection of the ages and the homes, but degeneracy in the parents, which is
usually characterised by defective vision among other factors, may be the source
of the want of cleanliness ; and the association of this with bad vision be thus
(inly a secondary hereditary effect. Until we know far better than we do at
present the closeness of correlation between various physical degeneracies and
the relation of these in the parent to the home environment, it seems hardly
possible to determine whether these small correlations are really the expression
of nurture, or only a secondary effect of nature. What is certain is that they
are practically negligible compared with the direct effect of nature.

K. P.

PART IP

SOME RECENT MISINTERPRETATIONS OF THE
PROBLEM OF NURTURE AND NATURE

BY KARL PEARSON, F.R.S.

One of the strange sequels to real human service is the not
infrequent spectacle of a man of transcendent power being made a
hero after his death by those who disregard the chief purpose of his
life, often in fact by those who have not the faintest conception of
what he aimed at, or indeed of what he actually achieved. Our intel-
lectual leaders even to-day are treated as was Karl der Grosse, who
within a few years of his death became the fabulous being who played
with serpents, rather than the master, who had controlled and inspired
men. There was this excuse in the case of Karl der Grosse — he left
no writings of his own to indicate what his aims, his methods and his
actions had been. The present day is the day of the printed word,
when every man of mark leaves his message to be read by those who
will, and yet — and possibly because of the superabundance of the
written word — that message is not studied and we continue to fable
about our heroes, as our forebears fabulated one thousand to two
thousand years ago.

It would be an interesting psychological study in the birth of
fable, if to-day we could first ask Francis Galton what he considered
the main feature of his life-work, and we could then independently
ascertain from a few of those who now celebrate his name what they
suppose his chief aim and principal achievement to have been !

1 This lecture was delivered at the Galton Laboratory, March 17, 1914. A portion
of it amplified from the analytical side was issued in Biometrika, vol. x, pp. 181 — 187,
in a paper entitled: "On certain errors with regard to Multiple Correlation occasionally
made by those who have not adequately studied the Subject." The correlation
coefficients cited are chiefly due to Ethel M. Elderton and will be published with
a detailed account as opportunity offers in the Eugenics Laboratory Memoir Series.

30 SOME RECENT MISINTERPRETATIONS OF THE

that from a knowledge of Gallon's writings alone — apart
from many intimate talks and much correspondence — I may safely
assert that the very basis of his lift.-- work was the conception that
nature for man is more important than nurture, that this principle
can be established quantitatively, and that only when it is fully
K alix-d will humanity be able to raise itself in the scale of nn ntal
and physical fitness. Did I not feel certain that this was his funda-
mental doctrine, and that this was the new pathway he hewed for
human progress through the jungle of undemonstrated human opinion,
I could not hold the post I do.

Gallon's "transvaluation of all values" was not a rapid or easy
one; he trod his own path as a solitary adventurer. He threw off
some of the fetters of ancient opinion when he came in his journeyings
of 1845-6 through Egypt, the Soudan and Syria into contact with
another faith, which he saw, or believed he saw, was more efficient
than his own in the control of conduct among oriental races. He
then grasped for the first time the relativity of religious belief, its
geographical, racial and environmental associations. But it was not
till the publication in 1859 °f Charles Darwin's Origin of Species that
Gallon attained to a positive creed, and to that book he attributed
his own actual freedom1. Henceforth he was a pioneer, if he led but
a corps of guides and scouts.

But Gallon differed essentially from Darwin, for his first idea was
the application of the doctrine of evolution to the conscious uplifting
of man by man himself. He desired above all things to accelerate by
his own work the ascent of man, and the first problems which fascinated
him were problems of human modifications due to environment, the
possibility of the inheritance of acquired characters, the continuity of
the germ plasm, the stirp theory, and generally all those problems
which are fundamental in their bearing on the relative influence of
nature and nurture. I well remember a conversation I had with him
shortly before his death on this very subject. We were discussing
the very bitter feeling that the papers by the staff of the Eugenics

1 "I always think of you in the same way as converts from barbarism think of
the teacher who first relieved them from the intolerable burden of their superstition.
I used to be wretched under the weight of the old fashioned 'arguments from design/
of which I felt, though I was unable to prove to myself, the worthlessness. Conse-
quently the appearance of your Origin of Species formed a real crisis in my life ; your
book drove away the constraint of my old superstition, as if it had been a nightmare,
and was the first to give me freedom of thought." Letter of Francis Galton to
Charles Darwin, Dec. 24, 1869.

PROBLEM OF NURTURE AND NATURE 31

Laboratory on the influence of environment had called forth, and he
interjected: "I wish your critics would study the subject of twins and
read my paper of 1875." How many of our critics have even now
read that paper? How many have really studied Galton's Hereditary
Genius ?

Here are the actual words1 he uses in the section of that work
entitled "Nature and Nurture":

"When nature and nurture compete for supremacy on equal terms
in the sense to be explained, the former proves the stronger. It is
needless to insist that neither is self-sufficient ; the highest natural
endowments may be starved by defective nurture, while no carefulness
of nurture can overcome the evil tendencies of an intrinsically bad
physique, weak brain, or brutal disposition. Differences of nurture
stamp unmistakable marks on the disposition of the soldier, clergyman,
or scholar, but are wholly insufficient to efface the deeper marks of
individual character" (p. 12).

How did Galton try to solve the relative strengths of "nature and
nurture" — this "convenient jingle of words," as he terms it, which
"separates under two distinct heads the innumerable elements of
which personality is composed"? He noted that twins are of two
kinds — those born physically and mentally alike, and those born as
unlike as ordinary brothers and sisters. He proceeded to determine
how far like twins were differentiated by unlike environments, and
how far unlike twins were rendered like by their common nurture.
He discovered that whatever the environment like twins remained
alike and unlike twins remained unlike, even as they were born. Thus
he sums up his History of Twins, as a Criterion of the Relative Powers
of Nature and Nurture :

"There is no escape from the conclusion that nature prevails
enormously over nurture when the differences of nurture do not
exceed what is commonly to be found among persons of the same
rank of society in the same country. My only fear is that my evidence
seems to prove too much, and may be discredited on that account,
as it seems contrary to all expectation that nurture should go for so
little. But experience is often fallacious in ascribing great effects to
trifling circumstances. Many a person has amused himself by throwing
bits of stick into a tiny brook and watching their progress ; how they
are arrested, first by one chance obstacle, then by another; and
again, how their onward course is facilitated by a combination of
circumstances. He might ascribe much importance to each of these

1 Portions of this lecture have been used in Chapter I of my Life of Francis Galton,
Cambridge University Press, 1914.

32 SOME RECENT MISINTERPRETATIONS OF THE

events, and think how largely the destiny of the stick has been
governed by a series of trifling accidents. Nevertheless all the sticks
succeed in passing down the current, and they travel, in the long run,
at nearly the same rate. So it is with life, in respect to the several
accidents which seem to have had a great effect upon our careers.
The one element, which varies in different individuals, but is constant
for each of them is the natural tendency ; it corresponds to the current
in the stream, and inevitably asserts itself." (Journal Anthropological
Institute, 1875, p. 391, &c.)

Such work as the Galton Laboratory has done was to give quanti-
tative definiteness to this conclusion of its founder.

It is to me one of the most astonishing illustrations of the worship
of the fabulous hero — the worship of the man whom you carve out as
idol to appear what your fancy desires him to have been, regardless
of what he was — when we find Francis Galton honoured by men who
suggest "that at present we should, as far as possible, avoid such
phrases as the relative influence of heredity and environment1," or
who deny that there is evidence to show that the mental characters
are inherited like the physical. To such men we can only cry in
amazement : What in the world do you suppose Francis Galton did
teach ? What contributions do you imagine he really made to scientific
knowledge? What principles do you consider he actually deduced for
human conduct from his discoveries? When one member of the
Eugenics Education Society preaches "the decadence of the Galtonian
or biometric study of heredity," when a second holds that it is still
an open problem whether mental characters are mainly due to heredity
or to training, and when the President of the Society wishes to avoid
the fundamental necessity of determining before action can be taken
the relative influence of nature and nurture, surely we may demand
of them why they honour Galton at all2? Is it only because a
heterogeneous conclave needs some well-known flag to follow, requires
the shadow of some great name to sanction its proceedings, regardless
of what the flag stands for, or what were the most cherished principles
of the man chosen as its fabulous hero?

But I shall get warmer than is fitting, if I stay to discuss this

1 Major Leonard Darwin, The Eugenics Review, vol. v, p. 154.

* Another recent or present member of the Eugenics Education Society who prates
much of "our master, the late Sir Francis Galton" states that "the fact remains that
he omitted to discuss factors of racial injury which are quite immeasurably more
important than any which he considered at all." As Galton once said of similar
utterances, they remind you of the slathering of the boa constrictor preparatory to
his making a meal of you.

PROBLEM OF NURTURE AND NATURE 33

matter further. I must turn to the main topic of my lecture to-night,
and bring up — as Francis Galton would himself have wished — such
vagaries of opinion as I have quoted against the wall of solid statistical
fact.

We have three fundamental questions to answer:

(i) What do we as Galtonians and Biometricians understand by
heredity, and how do we measure it?

(ii)' What do we of the same school understand by environment?

(iii) What do members of this Laboratory mean when they assert
that nature or heredity is far more important than nurture or
environment ?

First what do we mean by heredity?

There are to my mind two distinct factors involved under the idea
of heredity, which may not unfitly be described as interspecial (or
even interracial) heredity and intraspecial (or intraracial) heredity.
If every possible human being in a given race were equally likely to
be the product of every possible male and female within the race, we
as biometricians should assert that there was no heredity in the race.
Our measure of heredity is the measure of the extent to which a
deviation from average or type within the race is inherited by the
progeny. If it is not inherited then there would be no intraracial
heredity. We are perfectly aware that this is not the sense given to
the word heredity by some biologists, they would attribute to heredity
the fact that a goat produces a goat — and not a sheep — even if the
characters of individual goats had no relation to parental deviations
from type. The difference in meaning is not a subject for dispute,
but a question of definition, of a proper understanding of the use of
terms in a particular context. For a Galtonian heredity — or better
intraspecial heredity — is a measure within a definite group of the
inheritance of such deviations from type as actually occur within
the group.

Secondly what does the Biometric School understand by "environ-
ment"? Certainly their idea of it is not identical with the silly sort
of notion by which certain biologists consider they can demonstrate
that nurture is more important than nature. Without environment
they state an organism could not survive at all, hence environment
must be more important than heredity. We might cap such an
argument by asserting that as supermen would produce supermen, and
as no environment has yet produced a superman heredity must be
more effective than environment ! What the biometrician understands

34 SOME RECENT MISINTI- KPRETATIONS OF THE

by environment is tin- actual rani^r of environments in which he
finds the organism >ul»istini,'. and he asks whether when environ-
ment IN varied within this range the characters of the organism are
changed to the same extent as they are by such variations as occur
in the ancestry of existing members of the race. In other words he
measures and compares variations in the individual produced by
varying existing environment and by varying ancestry among possible
existing types. The biometrician does not stay to discuss what would
happen if you kept your human being under a pressure of fire
atmospheres or at a temperature of 100° Fahr. The biometrician
contents himself with a practical problem; he asks how changes
which are within" the range of practical politics — the replacement of
"back-to-back" houses by "through" houses, the abolition of cess-
pools, the diminution of overcrowding, the provision of better lighting
or ventilation and so forth — would improve the child as compared
with selecting parentage by sound health, efficient mentality and good
habits. Such is the problem of environment and heredity that the
biometrician has in mind, and such is the problem that he answers
when he says that the influence of nature on the average is five to
ten times as important as that of nurture. We assert that given the
existing range of variation in possible parents and given the existing
range of variation in possible environments, the association between
variation in the offspring and variation in the parent is far more
intense than the association between variation in the offspring and
variation in the environment. We obtain this result by proving that
the correlations in the two cases are of a wholly different order. In
other words the relationship between parent and offspring is far closer to
a causal bond than' the relationship between environment and offspring.
Biometricians have, I hold, been perfectly consistent in what they
have done ; they have not used environment or nurture in any vague
sense ; they have taken the range of environment which occurs within
practical experience, they have dealt with variations of nurture subject
to municipal or state control ; they have considered the time children
spend in the school, at home, on the streets; they have dealt with
occupations, wages and ages of parents; they have considered in
short all the social, industrial and economic conditions of the environ-
ment of the child of which they could procure data, or which admitted
of measurement and record, and they have shown that the variations
in these factors of nurture are of relatively small importance compared
with the results of variation in the physique, the mentality and the

PROBLEM OF NURTURE AND NATURE 35

habits of parents. Surely before we can better the race effectually we
must learn where our endeavours will have most effect. But because
biometricians have ventured to consider what is after all the funda-
mental problem of Eugenics, nay indeed of all valid attempts to better
social conditions, and because they have been forced to a conclusion
which is opposed to much unfounded charitable and philanthropic
opinion they have been calumniated as defenders of slums, as the
servants of brewers and publicans, as the suborned agents of capitalists,
and as modern Herods, whose object is to destroy infancy1.

If we turn from critics of this ill-balanced type of mind to those
who read and think before they proclaim in the market-place, we still
find almost as much obscurity in their criticism of what we have said
and in their appreciation of the evidence on which it is based. What
we complain of is that even our most friendly critics are liable to
ascend the cathedra with a huge bag of phrases but rarely with a
waistcoat-pocketful of established facts. I must stay to illustrate
these points by quotations from one or two of our critics. Here are
some remarks from a recent lecture by Professor J. Arthur Thomson :

"Character in its widest sense depends fundamentally on the
innate organisation, but its expression depends largely on nurture.

The adult mind is as much made as born Accepting Professor Pearson's

valuable work, we are not, however, led to any depreciation of nurture,
an art in which we are still in fact tyros.... There is little reason to
suppose that we make anything like the most of our hereditary nature
along the lines of psychical development. There are great fallow areas
in the brain still to be cultivated. Instead of attaching too much
importance to nurture, as Professor Pearson contended, we have not
yet begun to attach enough."

I would draw the reader's attention to the passages I have
italicised. Now these phrases are absolutely typical, for not a single
measurable fact is produced to substantiate them. "The adult mind
is as much made as born." What evidence is there for this? The
direct evidence that it is not the fact produced by Galton is tacitly
put on one side. Galton showed — and he has been confirmed by
Thorndyke — that like twins remained absolutely alike under different
environments, and that unlike twins did not become mentally or
physically alike under the same environment.

1 A good illustration of this type of writing occurs in the recent utterances of a
man of the market-place who throws overboard "all the deplorable and notorious
nonsense published in the interest of obsolete statistical 'methods' or of class or
political prejudice."

3—2

36 SOME RECENT MISINTERPRETATIONS OF THE

If .till "tyn>s in the art of nurture" is it not because we

have m-vrr hitherto procercK-d to nu-iiMire the effect of various types
viionnu-nt. but assumed that what made life easy for man mu>t
also In hot for man? If mankind has remained a tyro aft«
thousand yrars < xjxriinenting, is it not rather idle for Professor
Thomson to trll us that somehow or somewhere, he or somebodx
is going to makr more out of nurture? What evidence has he that
tlu-re are "large fallow areas in the brain still to be cultivat
Nature does not long preserve what is not used; she is nothing if not
economical! Have those great fallow areas in the brain been lying
idle since the date of our apelike ancestors, or only since neolithic
times? \Yhat necessity of evolution called them into existence and
why have they remained quiescent ever since? Which section of the
brain do they lie in? How has Professor Thomson discovered that
present day man is superman asleep, and what is the sort of nurture —
phosphorus or the pedagogue — which will rouse these fallow areas into
activity? Such wordy dogmatism would be outbalanced by a single
grain of the investigating statistical spirit of Francis Galton. Professor
Thomson is a member of the Council of the Eugenics Education
Society. Let us now turn to the words of its President.

Major Leonard Darwin writes as follows in the Journal of the
Eugenics Education Society.

" It is impossible to compare heredity as a whole with environment
as a whole as far as their effects are concerned; for no living being
can exist for a moment without either of them1. Moreover, in order
to compare two things so as to be able to use the words more or less
in connection with such a comparison, we must have a common unit
of measurement applicable to them both. But what is the unit by
which both heredity and environment may be measured? I myself
have no idea. May we not be discussing questions as illogical as
enquiring what portion of the area of a rectangle is due to its width
and what to its length? Is it ever wise to use words in scientific
literature without endeavouring to attach a definite meaning to them2?"

1 There would in our sense be no heredity if the average child born to noteworthy
parents was equal to the average child of the whole community. Yet it is perfectly
easy to understand how living beings could exist under such a law of reproduction.
Major Darwin seems to be confusing two things, interspecial and intraspecial heredity,
the fact that a man is born true to his species, and the fact that he resembles his
immediate ancestry. It is the latter fact only which concerns us when we compare
heredity and environment, i.e. how far variation of immediate ancestry affects the
individual's physical or mental characters. But without such heredity individuals
might quite well exist.

: The Eugenics Review, vol. v, p. 152. The italics are mine.

PROBLEM OF NURTURE AND NATURE 37

It is hard to conceive a paragraph of the same length more full
of evidence of complete ignorance of the methods used in modern
science for comparing correlated variates! Yet it goes out as the
opinion of the President of a Society which is endeavouring to spread
the scientific doctrines of Eugenics among the people ! Major Darwin
begins by stating that it is needful to have a common unit of measure-
ment in order to compare two variates. To begin with we are not
comparing two things, but we are comparing the influence of two
things on a third, i.e. the intensity of a certain environmental influence
and the intensity of a certain somatic character in the parent, say, on
the intensity of the somatic character in the offspring. Yet Major
Darwin tells us we cannot do this because we cannot measure these
things in the same unit ! — How suavely yet forcibly Sir Francis Galton
himself would have ridiculed such ignorance in high places as is passed
by the Editor of this Eugenics Journal ! — We can see him now telling
us how the intensity of each character could be measured by its grade,
and how the problem turned on whether the same change in grade in
the environment and in the parental somatic character produced
greater or less change in the grade of the filial somatic character.
When we inquire whether inter-racially stature is more closely related
to cephalic index or to eye colour, are we to be met by the statement
that these characters cannot be compared because they cannot be
measured in a "common unit," and are we then to be told that it is not
"wise to use words in scientific literature without endeavouring to attach
a definite meaning to them?" Every trained statistician knows that
each character is measured in the unit of its own variability — in what
he tenns its standard-deviation1, and that this standard- deviation
provides him with a measure of the frequency of each value of the
variate in question. It seems to me that the only correct sentence
in this paragraph, is the author's statement that he himself has no
idea what unit is "common" to heredity and environment.

But our author continues:

"Take any quality, and we find that the human beings composing
any community differ more or less considerably as regards that quality.
Now we can measure the correlation between the differences shown in
this quality and the differences of environment to which the members
of the community in question had previously been exposed2. This is

1 Of course he may or does need other constants to help in the description of the
frequency.

* loc. cit. p. 153.

38 SOME Rl-t I'M MI -INTERPRETATIONS OF THE

one correlation. Then we can also measure the correlation coeffin. m
t.itluT and son, as regards the quality in question. I Im-
is a second corn lation; and if we are told that the relative influence
of environment and heredity is measured by the ratio between these
two correlation coefficients, we certainly do thus get a clear conception
of what is meant1."

But has the writer really obtained a clear conception of what such
coefficients of correlation mean, when in the next paragraph he
continues :

"Imagine an ideal republic, in some respects similar to that
designed by Plato, where not only were all the children removed
from their parents, but where they were all treated exactly alike.
In these circumstances none of the differences between the adults
could have anything to do with the differences of environments, and
all must be due to some differences in inherent factors. In fact the
environment correlation coefficient would be nil, whilst the hereditary
correlation coefficient might be high2."

Could any better evidence be adduced that the President of the
Eugenics Education Society did not know what a coefficient of
correlation meant at that date? The coefficient of correlation for the
environment might be anything from — i to + i ; the only obvious
fact would be that you could not find its value, except in the form
o/o, from an environment, which entirely precluded any occurrence of
variation. How again Sir Francis would have smiled at the notion
that the coefficient of correlation for a constant environment must be
nil. Why should we follow such advice as that given by the President
of the Society to avoid as far as possible "such phrases as the relative
influence of heredity and environment," when on his own showing he
does not in the least appreciate the methods by which this relative
influence is measured?

Then Major Darwin continues:

"Surely what we want to know is how we can do most good —
whether by attending to reforms intended to affect human surroundings,
or to reforms intended to influence mankind through the agency of
heredity. But does this ratio [that of the environmental and hereditary
correlation] give us any sure indication of the relative amount of
attention which should be paid to these two methods of procedure ? "

1 This seems to contradict the writer's previous assertion that two things are
incomparable, if they have not a "common unit"'

1 I wrote at once to Major Darwin pointing out the error of such a statement and
he withdrew it in the next number of his journal. But the harm done by an article
of this kind cannot be reversed by correcting a single misstatement out of many.

PROBLEM OF NURTURE AND NATURE 39

Our only reply can be that these correlations certainly do, and
that as long as the President of the Eugenics Education Society fails
to grasp their meaning, he is doing grave harm to the science of
eugenics, and to the studies which Francis Galton had at heart.

One of the chief achievements of Francis Galton was his introduc-
tion of the conception of correlation to measure the relative intensity
of those very factors of social progress, which the President of the
Eugenics Education Society holds to be beyond comparison. Corre-
lation measures for us the extent to which one variate is determined
by a second; when correlation is high then we approach causation.
When we say that the correlation of the character of the offspring
with ancestral characters is ten times as intense as it is with environ-
mental factors, then we mean that the degree of dependence of the
child on the characters of its parentage is ten times as intense as its
degree of dependence on the character of its home or uprearing.

Now if the reader will examine the accompanying tables which
were published in a lecture of igio1 he will be at once struck by the
wholly different order of the correlations due to Nature and those due
to Nurture; we have an average Nurture correlation of about -03 as
against a Nature correlation in the closest relationships of about -50.
The number of these correlations have been very largely extended
since that date and the diagram on p. 42 shows the distribution
of frequency for a long series of Nurture and Nature values. It will
be seen at once that the two frequency distributions have no common
range. There is no real comparison possible between them, and
whereas ancestry always draws the offspring in its own direction, a
good environment may actually be associated with inferior physique
or mentality in the child.

At this point, however, fresh criticisms of our work have been
advanced, also by members of the same Eugenics Education Society.
We admit, they say, that the environmental correlations may be of
the order -03 or -05 and the inheritance correlations of the order -50.
But this is the correlation of one character with environment. You
ought to take ten or twenty, and then you will have multiplied up
environment to be more effective than heredity for -03 x 20= -60. In
the first place we may suggest that it would be just as reasonable,
if the argument were a valid one to multiply up the favourable
hereditary characters, to take weight, height, muscular activity,

1 Nature and Nurture, The Problem of the Future, Eugenics Laboratory Lecture
Series, No. VI, Cambridge University Press.

4O SOME RECENT MISINTERPRETATIONS OF THE

TABLE I. Strength of Nature.
A. PARENTAL RESEMBLANCES.

(i) Physical characters.
Pair Organ Correlation

Father and son Stature -51

Span -45

..,,,, Forearm -42

,, ., Eye colour -55

Father and daughter Stature -51

Span -45

,, » Forearm -42

Eye colour -44

Mother and son Stature -49

,, >. ,, Span -46

Forearm -41

,. » » Eye colour -48

Mother and daughter Stature -51

,, » Span -45

,, ,, ,, Forearm -42

,, >, ,, Eye colour -51

(ii) Pathological characters.

Parent and offspring Pulmonary tuberculosis (Pearson) -40 to -60

>, (Goring) 43 to -62

,, ,, Insanity (Heron) -53

(Goring) -47

,, ,, „ Deaf-mutism (Schuster) -54

„ ,, ,, Corneal refraction (Barrington) -60

(iii) Mental characters.

Father and son Ability (Oxford Class Lists, Schuster) -49

„ „ „ Intelligence (Family Records, Pearson) -58

Mean Parental Correlation '49

B. FRATERNAL RESEMBLANCES.

(i) Physical characters.

Brother and brother Stature -51

» Span -55

,, Forearm -49

,, Eye colour -52

,, Cephalic index -49

,, Hair colour -59

,, General health 49

Sister and sister Stature -54

Span -56

Forearm -51

Eye colour -45

Cephalic index -54

Hair colour -51

General health -56

PROBLEM OF NURTURE AND NATURE

Pair
Brother and sister

Pairs of siblings

Brother and brother
Sister and sister
Brother and sister
Pairs of siblings

TABLE I (continued)

Organ
Stature
Span
Forearm

Eye colour .

Cephalic index
Hair colour
General health

(ii) Pathological characters.

Phthisis (Pearson)
Insanity (Heron)
Deaf-mutism (Schuster)

(iii) Mental characters.
Ability

Temper
Handwriting

Mean .Fraternal Correlation
Mean "Nature" value

Correlation

•55
•53
'44
•46

•43
•56
•62

•4«
•73

•52
•45
•49

•52
'52

TABLE II. Strength of Nurture.

Pair of characters dealt with

Correlation

Keenness of vision and home environment as measured by cleanliness

of body and clothing ... ... ... ... ... ... ... +-07

Eye disease and overcrowding ... ... ... ... ... ... +-05

,, ,, ,, economic condition of home ... ... ... ... +-03

,, ,, ,, physical state of parents ... ... ... ... ... --06

,, ,, ,, moral state of parents ... ... ... ... ... +-02

Myopia and age at which child begins to read ... ... ... ... - -08

Liability to phthisis and destitution of home ... ... ... ... + -02

Keenness of vision and overcrowding (no. of persons per room) ... - -10

Myopia and overcrowding (no. of persons per room) ... ... ... --07

Refraction of offspring and moral state of parents ... ... ... --09

,, ,, ,, ,, physical state of parents ... ... ... -oo

,, ,, ,, ,, economic condition of home ... ... --05

Keenness of vision and moral state of parents ... ... ... ... --02

Keenness of vision and physical state of parents ... ... ..'. ... --01

Mental capacity and weight of child ... ... ... ... ... + -04

stature of child ... ... ... ... ... +-08

condition of teeth ... ... ... ... ... +-09

condition of clothing (boys) ... ... ... ... +-04

condition of clothing (girls) ... ... ... ... +-24

state of nutrition (boys) ... ... ... ... +-oi

state of nutrition (girls) ... ... ... ... + -08

cleanliness (boys) ... ... ... ... ... +'14

cleanliness (girls) ... ... ... ... ... +-07

state of glands in children ... ... ... ... +-08

state of tonsils (boys) --or

state of tonsils (girls) ... ... ... ... +-n

SOME RECENT MISINTERPRETATIONS OF THE

TABLE II (continued)

Pair of characters dralt with Correlation

Keenness of vision and time spent out of doors ... ... ... ... -oo

Mycipi. i and time spent out of doors ... ... ... ... ... -oo

-ht of child and alcoholism of parent ... ... ... ... ... +-06

Stature of child , ,» .. . ••• ••• ••• ••• ••• +-06

Hc.iltli (.1 child , ,. --05

Intelligence of ch Id and alcoholism of father ... ... ... ... --06

,, , , mother ... ... ... ... --04

Myopia ,, , ,, ,, , parent ... ... ... ... --12

Eye disease „ , „ father ... ... ... ... --08

,, ,, , , mother ... ... ... ... +-06

Weight of child and " unhealthy " trade of father ... ... ... +-04

Stature of ,, ,, ,, ,, ,, ,, ... ... ... +-07

Weight of son and employment of mother ... ... ... ... +-n

Weight of daughter and employment of mother ... ... ... ... + -07

Stature of son and employment of mother ... ... ... ... +'14

Stature of daughter and employment of mother ... ... ... ... +-n

Intelligence of son and employment of mother ... ... ... ... --16

Intelligence of daughter and employment of mother ... ... ... + -12

Health of child and employment of mother ... ... ... ... + -08

Weight of child and wages of father ... ... ... ... ... +-io

Stature of ,, ,, ,, ,, ,, .... ... ... ... ... +-09

Weight of child and number of rooms occupied ... ... ... ... +-ii

Statureof „ „ „• „ ,, ,, ... ... +-n

Mean Nurture Value +'03

A negative sign in the above table indicates that an unfavourable parental or
environmental condition appears on the basis of the data available to show improved
welfare in the child.

Comparative Values of Nature and Nurture Correlation Coefficients.

4CH

-1-0 -9 -8 --7 --6 --5 -4 -3 -1 -1 o -M *'2 +-3 +-4 +'5 +-6 +-7 +-8 *'9 +1-0

Scale of Correlation
Values available when diagram was drawn and tabled without selection

to sub-ranges of -05.

PROBLEM OF NURTURE AND NATURE 43

health, intelligence, caution, and many other desirable factors, and
these not only in one parent but in brothers, sisters, aunts, uncles and
grandparents and treat the cross-correlation of these with the character
under discussion. But although every improvement in stock would
reflect itself in improvement in offspring, correlations cannot be added
together — any more than forces by simple arithmetical addition. You
do not combine two hereditary correlations any more than two
environmental correlations by mere addition. You must proceed by
a combination process which is of course familiar enough to the
trained statistician.

Yet here is a statement which the Editor of the Eugenics Review
admits without contradiction to its pages1:

"The point that we wish to make is this. In the face of so much
ignorance concerning, not only heredity itself, but also its complement,
the influence of environment, how can any one be justified in making
sweeping generalisations with reference to these subjects?

Such generalisations, however, are made. It is said that we have
a definite proof that inheritance is of far greater strength than
environment. This argument takes the following shape. The corre-
lations between parent and offspring for a number of features have
been calculated, and the mean is found to be somewhere about -5.
Correlations between individuals and various aspects of their environ-
ment have also been worked out — as, for instance, mental ability and
conditions of clothing, or between myopia and the age of learning to
read2 — and the mean value is found to be about -03. It is then said
that the mean 'nature value' is at least five to ten times as great as
the mean 'nurture value,' and upon this is founded the generalisation
that 'nature' is of far greater importance than 'nurture'3. It may
be questioned, however, whether such a comparison does not involve
a serious mistake. For if we consider the two mean values that are
compared, we find that, whereas the 'mean nature value' is the mean
value of a number of observations, all of which provide a full measure
of the strength of heredity, the ' mean nurture value ' is the mean value
of a number of observations, each of which measures only the strength
of some one isolated aspect of environment. It would appear then that
the full strength of inheritance has been compared, not with the full
strength of environment, but with the average of a number of small

1 Vol. v, p. 219, in an article by A. M. Carr-Saunders.

2 As the writer phrases this correlation, it is very liable to be misinterpreted.
What the Galton Laboratory did was to show that myopia was very markedly
inherited, and that the theory that it was largely due to school environment was
incorrect, because children who began to read late, i.e. went late to school, were not
less myopic than those who went early.

8 Karl Pearson, Nature and Nurture, Eugenics Laboratory, Lectures VI, p. 25.

44 SOME RECENT MISINTERPRETATIONS OF THE

isolated aspects of the latter. As a matter of fact it is quite beyond
our power at present to sum up the full effect of environment upon
the individual and compare it with the full effect of heredity. \\V
are, therefore, justified in saying that we neither know in particular
cases how far the environment can produce any effect, nor can we
make any definite statement as to the comparative strength of ' nature '
and 'nurture.'"

Now this is the doctrine passed by the Editors of the Eugenics
Review, a society, which masquerades under the name of Francis
Galton1, and it is passed, because the editorial committee of that
society does not grasp the meaning of multiple correlation! In the
first place, of course, a single correlation coefficient does not provide
a full measure of the strength of heredity. In the table cited the
coefficients are those for one parent or for one brother or sister. Each
relative — and those for independent stocks are either non-correlated
or inter-correlated very slightly — provides such a coefficient, and
further each character in such relatives may be correlated with the
character under discussion in the person in question. In the next
place the environment factors do not consist of "some one isolated
aspect of environment." All these factors or aspects are closely
interlinked, and this was a fact well-known to the workers in the
Galton Laboratory. The real interpretation of such a difference as -50
and -03 in the average values of single coefficients can only be appre-
ciated by those who are conversant with the theory of multiple
correlation, and it is quite clear that those who profess to guide the
public in this very difficult problem — which is essentially a scientific
problem — lack any adequate knowledge of the sole instrument by
which any legitimate conclusion can be drawn.

The writer quoted appears to be wholly ignorant of the nature of
multiple correlation in the first place and in the second entirely to
overlook the very high correlations which exist between environmental
factors. Bad wages, bad habits, bad housing, uncleanliness, insanitary
surroundings, crowded rooms, danger of infection, &c., &c. are all
closely associated together, and while the order of correlation between
environmental and physical characters is low, that between individual
environmental factors is in our experience very high. Thus the
problem of heredity and environment is essentially the same as the
problem of multiple correlation.

1 If there was one point on which Francis Galton felt strongly and wrote it was
on this point of the relatively great intensity of "nature" as compared with "nurture."

PROBLEM OF NURTURE AND NATURE 45

But does any one trained in statistics believe that you add
correlation coefficients together to get their combined effect?

You might as well suggest that when you combine a force of 10 Ib.
with a force of 10 Ib. the resultant is 20 Ib., whereas it might be 2 Ib.
or possibly zero. The whole relationship depends on the angle between
the forces. In the same way the value of the coefficient of multiple
correlation depends on the correlations between the combined factors.

It is absolutely needful to impress this on the reader, and I will
illustrate it for a special case. Let the problem be: How far does
professional occupation, and prosperity influence the size of the
family ?

I take for the London districts1 the number of births per 100 wives
from 15 — 54 in a district. I take the number of professional men
per 1000 occupied males and the number of female domestic servants
per 100 females. We have:

Correlation of births and professional men = — -78 ;
„ „ and domestic servants = — -80.

Now suppose we wanted to find the total effect of professional
occupation and middle class prosperity as measured by the prevalence
of domestic servants on the birthrate. If you were to add — -78 to

- -80 you would get — 1-58, which is senseless, for correlation cannot
be greater than the perfect value unity! Actually the multiple
correlation coefficient is — -82, only -02 more than the value as found
for domestic servants alone and this result is as wholly reasonable as
the additive result was wholly fallacious. For to provide a measure
of the professional men in a district is almost equivalent to providing
a measure of the domestic servants in the same district. The correlation
between these two measures is in fact + -85. A knowledge of the two
factors does not give double information, and enable us to determine
double as accurately the birthrate ; it only raises the correlation from

- -80 to - -82.

Let us take a second illustration of this same most important
principle. Suppose we desire to predict the probable character of an
individual from (i) his two parents or from (ii) a large number of his
brothers and sisters. In the first case and in the second the correlation
between the individual and one only of his relatives is about the same,

1 See Heron, On the Relation of Fertility in Man to Social Status, &c., Cambridge
University Press, Drapers' Company Research Memoirs.

46 SOME RECENT MISINTERPRETATIONS OF THE

namrly -50. But there is an essential difference between the two
cases; pairs of brothers or sisters have a relatively high correlation
with each other, namely -50, but parents have a low correlation with
each other, on the average in the middle classes, perhaps, -15, due to
assortative mating; in the lower classes this sinks almost to zero.
The following table gives the results:

TABLE III.

Showing the Effect of Intercorrelations between Relatives, when a Judg-
ment has to be formed of the Character in an Individual (A) from
a knowledge of it in his Relatives.

Multiple Correlation Coefficients

A and

No assortative
mating

Assortative
mating

A and

Two parents

•707

•659

Two brothers

•577

Three ,

•612

Two parents "l

Four

•633

and four

.707

•664

Five

•646

grandparents;

Six ,

•655

Ten ,

•674

Infinite no. of\

brothers /

•707

Thus we see that unless there be assortative mating, we do not
better matters by introducing for the purposes of prediction the four
grandparents. And even with assortative mating two parents are
worth more than six brothers or sisters. Two parents who have not
selected their likes would form as good a basis for prediction as an
infinity of brethren, were such indeed available. Now the correlation
between brethren is only -50, and that between environmental factors
is often higher, thus this illustration will indicate how very little
additional intensity of correlation is gained by combining inter-
correlated factors. We do not add the correlations due to separate
factors as Mr Carr-Saunders' paper appears to suggest. If the factors
are absolutely independent, as when persons marry without assortative
mating, then the multiple correlation coefficient is the square root of
the sum of the squares of the correlations combined.

Thus if we took 100 independent environmental factors with a
strength of about -05, the resulting combined or multiple correlation

PROBLEM OF NURTURE AND NATURE

47

would only be -50, the value of the correlation of an individual with
a single parent.

If we suppose a system in which the environmental correlations
are roughly of the same order, and the inter-environmental correlations
approximately of value e, we shall be able to deduce approximately
the multiple correlation of the individual with n factors of environment
by multiplying the mean environmental correlation by the quantity

— i. The following table provides the value of this

i -\- €. \n — i)

expression for values of n and of the mean inter-environmental
correlation e.

TABLE IV.

Table of multipliers for estimating approximately the effect
of combining n environmental factors.

Number of Environmental Factors

2

3

4

5

10

16

IOO

QO

•oo

1-41

i-73

2-OO

2-24

3-16

4-00

10-00

_

c
o

•10

1-35

1-58

I-75

1-89

2-24

2-48

3-03

3-16

•20

1-29

1-46

1-58

1-67

1-89

2-OO

2-19

2-24

-^-H

.30

1-24

i-37

1-45

1-51

1-64

1-71

•80

1-83

a

•40

i -20

1-29

1-35

i-39

1-47

•51

•57

1-58

~

•50

1-15

1-22

1-26

1-29

•35

•37

1-41

3

•60

I-I2

I'I7

I -2O

I-2I

•25

•26

•29

1-29

•70

I -08

I-I2

I-I4

I'I5

•17

•18

•19

I -20

•80

I-05

I-07

I -08

1-09

•ii

•ii

•12

I-I2

•90

1-03

1-04

I-O4

1-04

•05

•05

•05

1-05

I-OO

I-OO

I-OO

I-OO

I'OO

•00

•00

•oo

I-OO

Now this table brings out some exceedingly interesting points.
In the first place if the inter-environmental correlations were as low
as -30, we should not double the multiple correlation of the individual
and his environment by taking into account an infinity of environ-
mental factors. If the inter-environmental correlations were as high
as -60, we should only add about 30 % to the mean value of the
correlations of the individual's character with the environmental
factors. It will be seen how soon we reach a limit to the sensible
increase of association obtained by combining numerous environmental
factors even if these be only moderately correlated together.

1 See Biometrika, vol. x, p. 181.

48 SOME RECENT MISINTERPRETATIONS OF THE

For example suppose we took 100 environmental factors — rather
more I fancy than we could ever hope within a reasonable time to
modify by social or political effort — and considered their intercorre-
lations to be of order -40. Then we should have to multiply by the
factor i -57. Thus with an average environmental correlation of -05,
the result would not be Vioo x -05, i.e. the '50 of independent environ-
mental factors, still less the unintelligible 100 x '05 of Mr Carr-Saunders'
notion, but only 1-57 x -05 = -078, or not a sixth part of the intensity
of the hereditary influence of a single parent !

I propose that we should now turn to a more detailed study of
environmental correlations in order to measure the applicability of
the above results. Five years ago in the lecture on Nature and Nurture,
the Problem of the Future, I put before my audience the Tables I and
II just reproduced (p. 40). The average value of the Nature factor
was -51 and of the Nurture factor -03. I never anticipated that any
one would be so misguided as to crudely multiply -03 by 17 and then
compare the result with the hereditary correlation for a single character
between a single pair of relatives. But I have learnt much in these
five years, and not the least important element of my knowledge is
the confirmation of my opinion that the great battle-cries of social
parties will in the present century turn on this problem of Nature
versus Nurture1. In the consciousness of this development the Galton
Laboratory has devoted much of its energies during this period to the
accumulation of data bearing on the influence of environment on child
welfare. Owing to the unremitting labours of my colleague Ethel M.
Elderton, we now know vastly more than we did five years ago. Where
we then knew one environmental correlation we now know a dozen;
further we have learnt the necessity for the classification of these
environmental factors, not only because certain classes are capable of
direct municipal control and others are not, but because certain classes
must be met by educative rather than by legislative action if they
present features which call for remedy2. If we term "environmental,"

1 Since this lecture was delivered the present war has come to test on a gigantic
scale how far free political institutions are compatible with the national organisation
requisite for national survival. However the war may ultimately end it cannot fail
to emphasise the movement towards national organisation as a necessity for existence.
This national organisation cannot stop short at organising the existing, it must face
the wider problem of how to obtain the highest type of citizen as material for
organisation.

1 It is characteristic of this period of change that greater stress is being laid on
educative remedies, especially on the instruction of parents. This is in accord with
the result reached by the Galton' Laboratory that parental habit is far more influential

PROBLEM OF NURTURE AND NATURE

49

all the influences that follow the birth of an individual — climate,
housing, food, parental conduct, — and do not distinguish between these
various classes, we may be able to show that environmental are less
influential than hereditary forces, but we shall not discover which
categories of environmental forces are most worth dealing with out
of the wide field provided by nurture. We shall distinguish first
accordingly between the physical environment — housing, sanitation
and ventilation — which within certain limits can be modified by
legislation or municipal regulations — and parental environment —
cleanliness, wage-earning power, feeding, parental habits and parental
health. But even this classification is not clear cut. For example
relative to other factors we find ventilation an element of some import-
ance in child nurture. But there are two phases of ventilation — the
one an absence of the means of ventilation is a feature of true physical
environment, but the other phase the neglect of existing means of
ventilation is a feature of parental environment that is of parental
habit. It may thus be quite as much a nature as a nurture factor.
As a matter of fact the second phase of ventilation is more highly
correlated than the first with health in the child, and the reader will
find this result almost universal: whenever a nurture factor can be
interpreted as possibly an indirect nature factor then it is likely to
be more influential.

Take the case of suitable feeding again, this may depend upon
father's wage, and that on his physique, or it may depend on the
mother's efficiency as a housekeeper, which in turn is closely related
to her habits and health. Or again it may be the outcome of municipal
arrangements as in the provision and inspection of milk supply ; further
geographical conditions may produce effect when we contrast urban
with rural districts, or the great seaports1 with manufacturing towns.

than physical environment. But just as popular opinion has overlooked the necessity
for estimating the relative intensity of Nature and Nurture in past measures of social
reform so it appears to escape our present pioneers of educative methods that a
fundamental d, priori problem is to determine how far parental habits in the mass are
the product of tradition or of inheritance — the heredity of an inert nature ; indeed
there is further some evidence that the tendency to maintain traditions is itself largely
a racial character. We not only observe that individuals brought up under the same
environment, physical and educational, have wholly different standards of order and
cleanliness, but a very small experience even of European travel shows that order and
cleanliness are racial characters, which are as conspicuous in certain European lands
as the predominating negligence and filth in others.

1 The relation of infantile diseases to food supply deserves careful study. The
prevalence of diphtheria in seaports may be connected with this subject.

4

50 RECENT MISINTERPRETATIONS OF NATURE AND NURTURE

In these aspects feeding is associated with physical rather than
parental environment. It will thus be evident that our division into
categories can only be very general. Even such a physical feature
as "housing" may be really a result of nature. An alcoholic father
may mean irregular wage and this enforces low rent and so bad
housing on the children. We have therefore in some way to get rid
of parental habits and parental health from the physical environment
correlations before we can say, this is the measure of pure nurture on
child welfare. We achieve this by correcting for these nature factors,
we take the partial correlation of the physical environmental factor
with the child's health or other welfare character for constant parental
habits and health. We must allow for the fact that the physically
and mentally inferior parents tend to occupy the cheaper houses.
It is impossible to attribute to the house what is really due to physical
and mental wreckage drifting where there is least to pay. It is largely
this fact, not the physical state of the home which is reflected in the
nurture of the children. With the assistance of Miss Elderton I have
endeavoured to classify "environmental factors" as follows:

(i) Physical Factors, as in housing and sanitation.

(ii) Parental Factors.

(a) Direct Parental Factors as in Habits, Cleanliness, Use of

Ventilation, &c.

(6) Indirect Parental Factors as in Wage, Occupation of Father,
Employment of Mother, &c.

Now I will ask you to look at the accompanying Table V for three
manufacturing towns, Rochdale, Bradford and Blackburn. It does
not cover any material previously dealt with in this paper and is a
fair sample of the environmental correlations we have reached. On
the left-hand side of the table we give the crude correlation values;
on the right-hand side we correct these correlations in part for health
and habits of parents, so as to remove as far as possible the nature
factor. But we have only been able to do this in part. Thus for
Rochdale we could correct for habits of parents only; for Blackburn
we could correct for health of parents only. In Bradford we have
corrected for health of mother only. Thus it will be clear that the
values reached are not true corrected values, but maximum values,
the actual associations between welfare of child and pure environ-
mental influences being still lower than those recorded in our
"corrected" columns. Now the reader will observe that of the three
main divisions Physical Environment, Indirect Parental Environment

-JOO SITE3UI P3I3U9Q

0
IO

o

i

10
TH

Corrected correlations

Corrected for health
of both parents

Blackburn

j-, CO

~ "£

CS H C

E* 1

O O HI

M MM

CO

CJ
TH

M IO

O M

o
o

M

co

TH

CO

0
CO

eo

+

+

Infantile
mortality

IO I O M

9 1 9 ?

CO
0

OO Tl-

M 9

M

M

9

00

9

?

0
IO

Corrected for
health of mother

Bradford

"3 M "c

0) -I-" O

oo in o -<J-M
9 9 T1 9 ?
i

0
0

t^<

9 "

>< CO t^

-* M O

00

~

o

?* f T f

o
eo

Ci

Infantile
mortality

to HI ro to O

9 9 M o M
i i

«*

CM

0

9999

TH

9999

0

TH

O

Corrected for
habits of parents

Rochdale

Health
at six
months

OO O rf- N «O
O O O O O

co

0

9

t^-O

9 T1

CO
CO

o

<s 0

N N

0

TH

Infantile
mortality

HI M M O ^ CO

O M O O O O

c-

0
0

o
9

O O 1

CO
CO

9
*l

cT'o
1

to

0
0

*l

1 1

Crude correlations

Uncorrected

Blackburn

frj

H HI M

c-

CO

TH

9 N 9 9

?

00

• ~.

TH

co

0

r.
co

Infantile
mortality

to o •<*•

9 9 T1

CO
CO

0

M O 9 W

?

0

9

00

H

0

co

TH

Uncorrected

Bradford

Health
at 12

months

oo '*• ioO M

M O M O N

o

TH

t^oo r

M f^ H

^ t^
•i 9

i

CO

O "

^t- *

- O 00
t CO CO

00

co

CO

Infantile
mortality

9 9 ? 9 T1

o

00

0

to t^oo TJ-

HI 9 9 9

10
00

9

'-

00 00 00 CO

M M M HI

CO

TH

Uncorrected

Rochdale

Health
at six
months

co^SSco

s

CO

CO

M T}-

CO

M N 00

co

0

10

'+

+

+

Infantile
mortality

to o ci cooo t^»o

HI 0< HI HI O O O

CD

TH
TH

HI O
HI HI

M 9

03

0

i~

C

_•—

HI N

t-
c-

TH

+

Post-natal factors

^^

Mean values

g 0) O
.5 § U3 £~rt

QnJOJ>H^i^PH

fa
a

a
B

B. Indirect parental environmen
Father's occupation
Regularity of work
Wage of father
Employment of mother
Income of family
Insufficiency of food

Mean values

C. Direct parental factors
Habits of father
Habits of mother
Cleanliness of home
Use of ventilation *

'-> '-S

B

_c s.

52 SOME RECENT MISINTERPRETATIONS OF THE

and Direct Parental Environment, only the correlation coefficients in
the last case rise to any significance. The Physical Environment is
of small importance compared to the Parental Environment, and we
are really thrown back on the Health and Habits of Parents as the
only really important factors in infantile mortality and the health of
surviving children. We have shown elsewhere that general health is
a markedly hereditary character1. With "Habits" the case is slightly
but not essentially different. There is evidence that the habits of the
mother are not only very sensibly correlated with her health (-52 in
Bradford), but that the father's habits are also associated with the
mother's health (-49 in Bradford). These matters will be discussed at
length when our full data are published, but it is clear that the father
of bad habits is far less frequently able to procure a wife of good health
than a father of good habits, and that apart from the heredity of the
orderly habit, the orderliness of the mother is sensibly a matter of
health — itself an hereditary character. As far as the evidence of
Rochdale, Bradford and Blackburn extends, we seem safe in asserting
that the Physical Environment correlation will give a mean value
below -05, and that probably if we could fully correct for health and
habits of both parents the mean value would scarcely exceed -03
or -04.

In regard to the Indirect Parental Environment the correlations
when corrected for parental habits and health are of much the same
order. Only in the case of Blackburn do they reach in the matter
of insufficiency of food (-31) to any importance. But in this case we
cannot correct for habits of parents, although we know that insufficiency
of food is veiy sensibly correlated with father's health ('4O2), father's
wage ('45 2) and his regularity of work ('6i2), indirect environmental
factors of the parent, all closely associated, as experience from
elsewhere shows, with the father's habits. Thus we are left with the
Direct Parental Environment as the only sensible influence on the
mortality and health of children up to twelve months of age. But
when we look at such factors as cleanliness of house and use of
ventilation, and note how the crude Rochdale values are reduced by
correcting for habits of parents, the crude Blackburn values by
correcting for health of parents and the crude Bradford values by

1 See Biometrika, " On the Hereditary Character of General Health," vol. ix,
pp. 320 — 329 and "Questions of the Day and of the Fray," No. VI, Eugenics and
Public Health, Cambridge University Press.

* Values for Blackburn.

PROBLEM OF NURTURE AND NATURE 53

correcting for health of mother alone, we may well ask what would
be left of the mean value -154, if full corrections for both health and
habits had been made. The present movement to educate parents is
right in so far as it recognises that the essential factor in the environ-
ment of the young child is the direct parental rather than the direct
physical environment. The parental health and the parental habits
are everything. But again the educationalists like the environ-
mentalists have overlooked the necessity for a preliminary inquiry
in order to determine; (a) How far is health of parent an hereditary
character? (b) How far are habits the outcome of health? and
(c] How far are habits themselves — orderliness and inertness of mind —
the result of inheritance ? Just as the environmentalists have neglected
the dominance of nature over nurture in child welfare, so there is
danger that the far more reasonable movement of the educationalists
will fail if they do not give due weight to the hereditary factor in
dealing with both the health and habits of parents.

We have next to consider the interenvironmental correlations.
Our chief illustration will be from Rochdale, the data for which are
not only more complete but have up to the present been more fully
worked out. We shall consider in order : (a) the correlations between
factors of physical environment, (b) the correlations between factors
of indirect parental environment and (c) the correlation between the
physical and indirect parental environments1.

TABLE VI.
Inter correlations of Physical Environmental Factors. Rochdale.

Number of rooms and dampness ... ... '42

lighting -54

,, ,, possibility of ventilation ... ... -88

,, ,, type of house ... ... ... ... '65

Type of house and overcrowding ... -54

Mean correlation... ... ... ... ... *606

It will be seen at once that the intercorrelations of the Physical
Environmental factors are of a totally different order to the corre-
lations of those factors with infant welfare.

1 The exact measurement of the characters, the methods by which the correlations
were determined, the size of the populations dealt with, and other features of the work
will all be published at length later with the tables of data. They cannot be
reproduced in a mere lecture. Thus "sanitation" covers cesspool, pan, and water
closet; type of house, "back to back" and "through" houses; father's occupation
skilled and unskilled, &c., &c.

54 SOME RECENT MISINTERPRETATIONS OF THE

In the next table we find illustrations of the correlations between
physical environment and indirect parental environment. I have in
this table purposely left out the correlations of mother's employment
with certain other factors for our experience has shown that its effect
is of a mixed nature. It may arise from extreme poverty, i.e. be
associated with low wage of father and thus with insufficiency of food,
or in other cases, especially in the textile towns, the young mother
almost invariably goes to work and this work is a source of increased
family income and better food supply. The high correlation given
above for emplovment of mother and income of family show that in
Rochdale the work of the mother is a source of prosperity1.

TABLE VII.

Inter correlations of Indirect Parental Environmental Factors.
Rochdale and Blackburn.

Rochdale Blackburn

Income of family and father's occupation ... ... ... -56

,, „ ,, employment of mother ... ... -60

Insufficiency of food and father's wage ... ... ... — -45

,, ,, ,, father's regularity of work ... — -61

,, ,, rent ... ... ... ... — -48

Father's wage and class of father's labour ... ... — -75

,, ,, ,, rent ... ... ... ... ... — -34

Mean ... ... -541

We next turn to the intercorrelation of physical and indirect
parental factors in the following table (Table VIII).

It will be seen that these intercorrelations are brought below -50
owing to the last two values for overcrowding. I doubt very much
whether overcrowding as measured by persons per room is to be
considered as a fair measure of bad environment; it frequently
denotes only numerous children in a possibly prosperous home.
There are obviously mixed causes at. work, largely depending on the
age of both father and mother, when we consider the association of
the wage of the former and the employment of the latter in relation
to overcrowding.

1 In Blackburn the correlation of mother's employment with father's wage (not
total income of family) is -27, with insufficiency of food - -03 and with cleanliness
of home «O2. All these results in my opinion show the influence of mixed causes,
poverty compelling to employment in some cases and the increase of prosperity
following from employment in fairly prosperous families actually bettering the home
conditions in other cases.

PROBLEM OF NURTURE AND NATURE 55

TABLE VIII.

Inter correlation of Physical and Parental Environmental Factors.
Rochdale and Blackburn.

Indirect parental environmental factor and

physical environmental factor Rochdale Blackburn

Insufficiency of food and number of rooms ... ... — -49

„ ,, ,, rent ... ... .-.. ... — -48

,, ,, ,, overcrowding ... ... ... — -39

Occupation of father and number of rooms ... ... -50

Regularity of father's work and number of rooms . . . -48

Father's occupation and overcrowding ... ... ... -45

Means of family and overcrowding ... ... ... ... -58

Father's occupation and type of house ... ... ... -56

Means of family and type of house ... ... ... ... -58

Income of family and type of house ... ... ... -67

,, ,, overcrowding ... ... ... -52

,, „ lighting" ... ... ... ... -46

,, ,, possibility of ventilation ... ... -58

Employment of mother and type of house ... ... -58

„ ,, number of rooms ... ... -45

,, ,, possibility of ventilation ... -53

lighting -31

,, ,, overcrowding ... ... 'iy

Father's wage and overcrowding — -12

Mean correlation ... ... *468

We now turn to the Direct Parental Environmental Factors which
form a very closely correlated group.

TABLE IX.

Intercorrelations of Direct Parental Environmental Factors.
Rochdale and Bradford.

* Bradford Rochdale

Health of mother and habits of mother '52

habits of father '49

Habits of mother and cleanliness of home *79

„ -63

use of ventilation '59

;; „ „ — -69

Health of mother and cleanliness of home ... '45

„ use of ventilation ... ... ... '3°

Mean correlation '568

Thus the direct parental factors of environment, which we know
in the case of health and suspect in the case of habits to be largely
hereditary have an intercorrelation of the order -6.

56 SOME RECENT MISINTERPRETATIONS OF THE

Now although it seems proper to exclude these probably hereditary
factors from an examination of true environmental factors it is worth
while to notice the extent to which the indirect parental and physical
environmental factors depend on the direct parental factors. The
fact is that parents of bad habits and poor health by a process of
selection gravitate to the worse type of dwelling. When we remember
that infant mortality and health of mother have a correlation of -30 in
Blackburn, and delicacy of infant and health of mother a correlation
of -41 in Bradford, while health of offspring and health of either parent
have a correlation of about -50 in the professional classes, we realise
how largely health is a matter of nature, not of nurture. In confirma-
tion of this consider the following results :

TABLE X.

Association of Health of Mother with Physical Factors.
Bradford and Blackburn.

Bradford Blackburn
Health of mother and overcrowding ... ... ... -25* -01

,, ,, ,, type of house ... ... ... -19

,, „ „ number of rooms ... ... -13 -20

,, „ „ sanitation ... ... ... --05

„ „ ,, rent ... -26 -12

Mean correlation ... ... '14

* Measured for Bradford by pence per person spent on rent.

It would thus appear that the association of bad physical environ-
ment with mother's bad health is not a third as intense as the
association of mother's health with that of her offspring. Even then
it would be wrong to assert that the bad health of the mother flows
from the environment. The unhealthy woman fails to find a healthy
normal husband, she is less often employed and thus the family income
being lower, there is more overcrowding, and the home is made in a
cheaper and inferior type of house. The fact that in Blackburn the
correlation between mother's health and insufficiency of food in the
family is -47 and between father's health and insufficiency of food
is -40, indicates how greatly health in parents affects the economic
condition of the family. It may be doubted whether physical
environment would show any sensible relation to mother's health
were the values corrected for income of family.

On the other hand as the next table will show the direct parental
environmental factors summed up in habits of mother and cleanliness

PROBLEM OF NURTURE AND NATURE

57

of home are closely associated with the factors of physical environment.
The data in this table are for Rochdale and Blackburn :

TABLE XI.

Influence of Direct Parental Factors on Physical Environmental Factors
Rochdale and Blackburn.

Rochdale Blackburn

Habits of mother and dampness of house ...
lighting ......

overcrowding
possibility of ventilation
type of house
number of rooms
Cleanliness of home and overcrowding

number of rooms
rent

type of house
Mean of " habits " correlations

~ t, i i • »» i , •

" cleanliness correlations

General mean -4R1

VJCllCI dl IllCcill, TOO

•62

•54

•63

•65 —

•55

•49

•57 '42

•37 '24

•29

•42

•580
•385

We have found that on the whole the direct parental factors
appear to be more closely correlated with physical environmental
factors than with the indirect parental factors, although certain of
the latter correlations are high, e.g. between habits of mother and
father's occupation, -55 (Rochdale), between means of family and
habits of mother -67 (Rochdale), between habits of father and occu-
pation of father -41 (Bradford). It cannot therefore be asserted that
even the physical environmental factors are free from the nature
influence. The parents of the worse health and habits are found
in the lower type of physical environment, and thus even the slight
association of worse physical environment with lesser welfare of infant
and childlife may in part or wholly be due to the direct parental factors ;
in part if we believe health but not habits to be hereditary, wholly, if
we consider orderliness and briskness of mind, nay, even the power to
receive and carry on a tradition to be essential parts of our mental
inheritance.

In any case we are in a position to sum up the problem of the
relative intensity of Nature and Nurture as illustrated by the data
for mortality of infants and health of surviving children from the
towns of Rochdale, Bradford and Blackburn. If we pool all environ-
mental factors, we shall not reach an average correlation higher
than -io with the welfare of offspring, while the interenvironmental
correlations will be at least of the average order -40, and accordingly
by Table IV an infinity of such factors could only provide a multiple
correlation of -158, for nurture as against the nature value for two

•

58 SOME RECENT MISINTERPRETATIONS OF THE

parents of -707. Thus the nature value is at least 4-5 times the nurture
value. But this is a minimum value of the relative intensity. If we
pick out the physical environmental factors and the indirect parental
factors the mean correlation corrected for either health or habits only
is '055, and the mean interenvironmental correlation of the order -60,
which provide by Table IV in the case of an infinity of environmental
factors for a multiple correlation of -071. This is just one-tenth of
the nature factor -707, as represented by two parents. But the
value -055 is probably too large ; it is not only higher than the value
found for material from other districts (see Table II), but we have
seen that the direct parental factors are very sensibly correlated with
the physical environment. The mean physical environmental corre-
lation is hardly likely to exceed -03 or -04, which with an interenviron-
mental correlation of -60 leads to a multiple correlation for an infinity
of factors of only -04, or -05 — not more than ^th to ^th of the value
of the intensity of nature as represented by the inheritance from two
parents. In other words our present results fully confirm the earlier
statement that the relative intensity of nature was five to ten times
that of nurture. We have chosen the cases of infant mortality and
delicacy of childhood for consideration as these are usually considered
to be peculiarly the outcome of the physical environment. As a
matter of fact the most influential post-natal factors are parental
health and parental habits, matters least susceptible of modification
by act of parliament or municipal regulation, unless we forbid parentage
to the unhealthy or place in the stocks parents who refuse to open
their windows, or who spend money on other matters than food.
I feel confident that within the limits to which we have applied our
statements — to the range of infantile and child nurture — and to those
variations of environment which fall within the field of experience or
of practical politics our assertion that nature is five to ten times as
influential as nurture is free from any exaggeration and forms a solid
ground upon which to base the direction of reforms which shall
accelerate racial progress. It is five to ten times as profitable for a
child to be born of parents of sound physique and of brisk, orderly
mentality as for a child to be born and nurtured in a good physical
environment. To state this is not to discredit all improvement of en-
vironment ; such improvement adds greatly to the amenities and to the
possibilities of human life, but it is to parentage itself that the patriot
who would work for racial progress must turn in the first place, if he
would achieve a greater success than the environmentalists with a

PKOBLEM OF NURTURE AND NATURE 59

century of social reform have hitherto been able to claim. This is
the sole ground on which the eugenist is distinguished from the old
social reformer, and those who fail to realise this emphasis of nature
over nurture are entirely missing not only Galton's methods of
investigation, but the essential feature of what he understood by
eugenics.

Yet what do we find is the present state of affairs with regard to
any really scientific study of sociological or racial problems? Why
that such study is met by wholly uncritical or often unthinking
attacks — attacks which show that their writers neither appreciate the
facts already known to us, nor grasp in the slightest degree the methods
by which alone these facts can be analysed. They honour Francis
Galton without studying what he devoted his life to demonstrating;
they establish an anniversary festival to his memory, and at the same
time deny the actual validity of the calculus which he first introduced
to solve these very problems of nature and nurture. They talk idly
of using "words in scientific literature without endeavouring to attach
a definite meaning to them" although the man they proclaim as a
leader had shown how to obtain a definite quantitative measure of
these very words. They talk vaguely about "large fallow areas of
the brain" still uncultivated and state that "instead of attaching too
much importance to nurture we have not yet begun to attach
enough" ; they trust to verbal discussion where Francis Galton would
have told them to measure and learn the facts before they spoke.
It was Galton's fundamental principle that before we know the
meaning of anything we must measure it, and that ultimately
everything would be found capable of measurement, if we use its own
appropriate footrule. I have even a letter from him in which he
discusses how we could obtain a definite numerical measure of the
influence of Mrs Grundy upon social reform. The one thing that
wearied Galton's almost inexhaustibly enthusiastic nature was a
torrent of words without any admixture of measured facts. He
invariably judged the worth of a publication by the extent of its
appeal to statistical data in the form either of observation or experi-
ment. Yet all this appears wholly forgotten by those who trumpet
most loudly his name. Again I ask: What results do they suppose
Galton reached and by what methods do they suppose he reached
them? The reality is summed up in the words of Galton himself,
which form the motto of our Biometric Laboratory:

60 RECENT MISINTERPRETATIONS OF NATURE AND NURTURE

" Until the phenomena of any branch of knowledge have been
submitted to measurement and number, it cannot assume the status
and dignity of a science."

Those who do not realise that this was the essence of Galton's
teaching, those who believe that they can solve the extremely difficult
problems of race-progress and of eugenics by mere verbal discussion
will only succeed in bringing his name and his work into discredit.

CAMBRIDGE: PRINTED BY JOHN CLAY, M.A. AT THE UNIVERSITY PRESS

mbridge University Press, Fetter Lane, London, E.C., and their Agents,
z now the sole agents for the sale of the following publications of the
'Iton and Biometric Laboratories, University of London:

BIOMETRIC LABORATORY PUBLICATIONS
Drapers' Company Research Memoirs.

Biometric Series.

.ematical Contributions to the

rheoryof Evolution.— XIII. On the Theory
>f Contingency and its Relation to Associa-
.ion and Normal Correlation. By KARL
'KARSON, F.R.S. Price 4s. net.
athematical Contributions to the
Theoryof Evolution.— XIV. On the Theory
of Skew Correlation and Non-linear Regres-
sion. By KARL PEARSON, F.R.S. Price 5s.
'let.

athematical Contributions to the
Theory of Evolution.— XV. On the Mathe-
•natical Theory of Random Migration. By
.•KARL PEARSON, F.R.S., with the assistance
jf JOHN BLAKEMAN, M.Sc. Price 5s. net.
athematical Contributions to the
Theory of Evolution.— XVI. On Further
Methods of Measuring Correlation. By
KARL PEARSON, F.R.S. Price 4s. net.

V. Mathematical Contributions to the

Theory of Evolution.— XVII. On Homo-
typosis in the Animal Kingdom. A Co-
operative Study. [Shortly.

VI. Albinism in Man. By KARL PEARSON,

E. NETTLESHIP, and C. H. USHER. Text,
Part I, and Atlas, Part I. Price 35s. net.

VII. Mathematical Contributions to the

Theory of Evolution.— XVIII. On a Novel
Method of Regarding the Association of
two Variates classed solely in Alternative
Categories. By KARL PEARSON, F.R.S.
Price 4s. net.

VIII. Albinism in Man. By KARL PEARSON,

E. NETTLESHIP, and C. H. USHER. Text,
Part II, and Atlas, Part II. Price 30s. net.

IX. Albinism in Man. By KARL PEARSON,

E. NETTLESHIP, and C. H. USHER. Text,
Part IV, and Atlas, Part IV. Price 21 s. net.

Studies in National Deterioration.

U the Relation of Fertility in Man

to Social Status, and on the changes in this

(Relation that have taken place in the last

50 years. By DAVID HERON, M.A., D.Sc.

Price 3s. net. Sold only with complete sets.
First Study of the Statistics of

IPulmonary Tuberculosis (Inheritance). By
KARL PEARSON, F.R.S. Price 3s. net.
Second Study of the Statistics of
Pulmonary Tuberculosis. Marital Infec-

rtiou. By ERNEST G. POPE, revised by KARL
PEARSON, F.R.S. With an Appendix on

Assortative Mating by ETHEL M. ELDERTON.
Price 3s. net.

Ihe Health of the School-Child in re-
lation to its Mental Characters. By KARL
PEARSON, F.R.S. [Shortly.

m the Inheritance of the Diathesis
of Phthisis and Insanity. A Statistical
Study based upon the Family History of
1,500 Criminals. By CHARLES GORING,
M.D., B.Sc. Price 3s. net.

VL A Third Study of the Statistics of

Pulmonary Tuberculosis. The Mortality
of the Tuberculous and Sanatorium Treat-
ment. By W. P. ELDERTON, F.I. A., and
S. J. PERRY, A.I.A. Price 3s. net.

VII. On the Intensity of Natural Selection

in Man. (On the Relation of Darwinism
to the Infantile Death-rate.) By E. C.
SNOW, D.Sc. Price 3s. net.

VIII. A Fourth Study of the Statistics of

Pulmonary Tuberculosis : the Mortality of
the Tuberculous: Sanatorium and Tuber-
culin Treatment. By W. PALIN ELDERTON,
F.I. A., and SIDNEY J. PERRY, A. I. A.
Price 3s. net.

IX. A Statistical Study of Oral Tem-
peratures in School Children with special
reference to Parental, Environmental and
Class Differences. By M. H. WILLIAMS,
M.B., JULIA BELL, M.A., and KARL
PEARSON, F.R.S. Price 6s. net.

Technical Series.

a Theory of the Stresses in Crane

and Coupling Hooks with Experimental
Comparison with Existing Theory. By
E. S. ANDREWS, B.Sc. Eng., assisted by
KARL PEARSON, F.R.S. Issued. Price3s.net.

•n some Disregarded Points in the
Stability of Masonry Dams. By L. W.
ATCHERLEY, assisted by KARL PEARSON,
F.R.S. Issued. Price 3s. 6d. net.

»n the Graphics of Metal Arches
with special reference to the Relative
Strength of Two-pivoted, Three-pivoted
and Built-in-Metal Arches. By L. W.
ATCHERLEY and KARL PEARSON, F.R.b.
Issued. Price 5s. net.

IV. On Torsional Vibrations in Axles
and Shafting. By KARL PEARSON, F.R.S.
Issued. Price 6s. net.

V. An Experimental Study of the
Stresses in Masonry Dams. By KARL
PEARSON, F.R.S., and A. F. CAMPBELL
POLLARD, assisted by C. W. WHEEN, B.Sc.
Eng., and L. F. RICHARDSON, B.A. Issued.
Price Is. net.

VI. On a Practical Theory of Elliptic and
Pseudo-elliptic Arches, with special refer-
ence to the ideal Masonry Arch. By KARL
PEARSON, F.R.S., W. D. REYNOLDS, B.Sc.
Eng., and W. F. STANTON, B.Sc. Eng.
Issued. Price 4s. net.

Complete sets may be purchased at 25s. net.

Drapers' Company Research Memoirs — (cont.).

icstions of the Day and of the /•'/<///.

I.

H.

III.

The Influence of Parental Alcoholism

on the Phy>i«]iie and Ability of the < )fl'-
sprinv,'. A Keply to the Cambridge r>»no.
mists. By KAKL PEAKS. »x, K. K.S. .

U H'-t.

Mental Defect, Mai-Nutrition, and
the Teacher's Appreciation of Intelligence.
A Reply to Criticisms of the Memoir on
'The Influence of Defective Physique and
riifavourable Home Environment on the
Intelligent- of School Children.' By DAVID
HKKON. D.Sc. /'/•/

An Attempt to correct some of the
Mis-statements made by Sir VICTOR HORS-
LEY,F.R.S., F.R,C.S.,an.l MARY D. Si
M.D., in their Criticisms of the Memoir :
' A First Study of the Influence of Parental
Alcoholism,' &c. By KARL PEARSON, F. R.S.
Price Is. net.

IV.

The Fight against Tuberculosis

Death-rate from 1'hthi-i-. l',y

I'K \K>oN, K. U.S. /'/•/'••• l.<.

Social Problems : Their Treatir i

l',i>t. I'nvM-nt and FnTur.-. 1 \\-
I'i .AU>')\, F.K.S. Price \t.

Eugenics and Public Health.
to the York ( 'on^re.xs of the
Institute. By KAKL PEARS<>
1«. net.

Mendelism and the Problem of Mei
Defer -T. i. A. Criticism of Recent Am<
Work. By DAVID HKRON, D.Sc, D
Number.) Price 2*. /

VIII. Mendelism and the Problem of Men

Defect. II. The Continuitv of M
Defect. By KARL PK

GOSTAVA.JAEDERHOLJI, Ph.D. 1

V.

VI.

VII.

IX.

Mendelism and the Problem of Mental Defect. III. On the Graduated (

Mental Defect, and on the need for standardizing Judgments as to the Grade of Social I
which shall involve Segregation. By KARL PEARSON, F.R.S. (Double Number.) Pr,

Eugenics Laboratory Publications

MEMOIR SERIES.

I. The Inheritance of Ability. By EDGAR

SCHUSTER, D.Sc., Formerly Galton Research
Fellow, and ETHEL M. ELDERTON, Galton
Scholar. Price 4#.

II. A First Study of the Statistics of

Insanity and the Inheritance of the Insane
Diathesis. By DAVID HERON, D.Sc., Form-
erly Galton Research Fellow. Price 3«. net.

III. The Promise of Youth and the

Performance of Manhood. By ED<;AK
SCHUSTKR. D.Sc., Formerly Galton Research
Fellow. Price 2s. 6d. net.

IV. On the Measure of the Resemblance

of First Cousins. By ETHEL M. ELDERTON,
Galton Research Fellow, assisted by KARL
PEARSON, F.K.S. Price 3s. 6d. i«.-t.

V. A First Study of the Inheritance of

Vision and of the Relative Influence of
Heredity and Environment on Sight. By
AMY BARRINGTON and KARL PEARSON,
F.R.S. Price 4*. net.

VI. Treasury of Human Inheritance

(Pedigrees of physical, psychical, and patho-
logical Characters in Man). Parts I and II
(double part). (Diabetes insipidus, Split-
Foot, Polydactylism, Brachydactylism,
Tuberculosis, Deaf -Mutism, and Legal
Ability.) Price 14*. net.

VII. The Influence of Parental Occupation
and Home Conditions on the Physique of
the Offspring. By ETHEL M. ELDERTON,
Galtou Research Fellow. Shortly.

VIII. The Influence of Unfavourable Home
Environment and Defective Physique on
the Intelligence of School Children. By
DAVID HERON, M. A., D.Sc., Formerly Galton
Research Fellow. Price 4*. net. Sold only
with complete sets.

IX. The Treasury of Human Inheritance
(Pedigrees of physical, psychical, and patho-
logical Characters in Man). Part III.
(Angioneurotic Oedema, Hermaphroditism,
Deaf-Mutism, Insanity, Commercial Abili-
ty.) Price 6s. net.

X. The Influence of Parental Alco.
on the Physique and Intelligence!
Offspring. By ETHEL M. E
.sisted by KAKL PEARSON. ,v
/'/•ice 4s. net.

XI. The Treasury of Human Inhei?

(Pedigrees of physical, psychi
logical Characters in Man). R
(Cleft Palate, Hare- Lip, Deaf-Muti
Congenital Cataract.) Price 10*. it

XII. The Treasury of Human Inherr

(Pedigrees of physical, psychical, an*
logical Characters in Man). Partfl
VI. (Haemophilia.) Price 15*. n*

XIII. A Second Study of the Influe?

Parental Alcoholism on the Phy.>i<
Intelligence of the Offspring. Bl
PEARSON, F.R.S., and ETHKL M. KL
Pria.' 4s. net.

XIV. A Preliminary Study of Ex

Alcoholism in Adults. By AMY B.
TON and KARL PEARSON. F.R.S.,
by DAVID HERON, D.Sc. Price 4*.

XV. The Treasury of Human Inheri

Dwarfism, with 49 Plates of Illuss
and 8 Plates of Pedigre-

XVI. The Treasury of Human Inheri
Prefatory matter and indices to
With Frontispiece Portraits of Sir
Galton and Ancestry. Price 3*. net

XVII. A Second Study of Extreme

holism in Adults. With special rr
to the Home-Office Inebriate Refoi
data. By DAVID HERON, D.Sc. Prici

XVIII. On the Correlation of Fertilitj

Social Value. A Cooperative
Price 6s. net.

XIX— XX. Report on the English Birt1
Part I. England, North of the H
By ETHEL M. ELDERTON, Galton Rl
Fellow. Price 9*. net.

Vol. I of The Treasury of Human Inheritance (VI + IX + XI + XII + XV + XVI of the
above memoirs) may now be obtained bound in buckram, price 52«. 6d. net. Buckram cases for

Eugenics Laboratory Publications— (cont.).

binding can be purchased at 2s. 3d. with impress of the bust of Sir Francis Galton. A photo-
graph (11 x 13") of Sir Francis Galton by the late Mr Dew Smith can be obtained by sending a
postal order for 10s. Gd. to the Secretary to the Laboratory, University College, London, W.C

TURE SERIES. Price Is. net each (Nos. Ill and X excepted).

The Scope and Importance to the

State of the Science of National Eugenics.

By KARL PEARSON, F.R.S. Third Edition.
The Groundwork of Eugenics. By KARL

PEARSON, F.R.S. Second Edition.
The Relative Strength of Nurture and

Nature. Part I. By ETHEL M. ELDERTON.

Part II. By KARL PEARSON. (New and

much enlarged Edition. First Edition" out of

print.) Price 2s. net.
On the Marriage of First Cousins. By

ETHEL M. ELDERTON.
The Problem of Practical Eugenics.

By KARL PEARSON, F.R.S. Second Edition.

VI. Nature and Nurture, the Problem of

the Future. By KARL PEARSON, F.R.S.
Second Edition.

VII. The Academic Aspect of the Science

of National Eugenics. By KARL PEARSON.
F.R.S.

VIII. Tuberculosis, Heredity and Environ-

ment. By KARL PEARSON, F.R.S.
IX. Darwinism, Medical Progress and Eu-
genics. The Cavendish Lecture, 1912. By
KARL PEARSON, F.R.S.

X. The Handicapping of the First-born.
By KARL PEARSON, F.R.S. Price 2«. net.

BIOMETRIKA

A JOURNAL FOR THE STATISTICAL STUDY OF
BIOLOGICAL PROBLEMS

Founded by W. F. R. WELDON,. FRANCIS GALTON and KARL PEARSON (Editor).

CONTENTS. VOLUME X. PART IV

(With Plate XX and

I. Association of Finger-Prints. By H. WAITE, M.A., B.Sc.
Thirty Diagrams in the text).

II. On the Problem of Sexing Osteometric Material. By KARL PEARSON, F.R.S.

(With One Diagram in the text).

III. Further Evidence of Natural Selection in Man. By ETHEL M. ELDERTON and

KARL PEARSON, F.R.S.

IV. Frequency Distribution of the Values of the Correlation Coefficient in samples

from an indefinitely Large Population. By R. A. FISHER.

V. On the Distribution of the Standard Deviations of Small Samples : Appendix I to

Papers by "Student" and R. A. FISHER. (Editorial.)

VI. Tuberculosis and Segregation. By ALICE LEE, D.Sc.

VII. The Influence of Isolation on the Diphtheria Attack- and Death-rates. By ETHEL

M. ELDERTON and KARL PEARSON, F.R.S. (With Two Diagrams in the text.)
Miscellanea : (i) On the Probable Error of a Coefficient of Mean Square Contingency. By
KARL PEARSON. (ii) Measurements of Medieval English Femora. A rejoinder
to Dr F. G. Parsons. By KARL PEARSON.

The subscription price, payable in advance, is 30s. net per volume (post free) ; single numbers
10s. net. Volumes I — X (1902 — 15) complete, 30s. net per volume. Bound in Buckram 34/6 net per
volume. Index to Volumes I to V, 2s. net. Subscriptions may be sent to C. F. Clay, Cambridge Uni-
versity Press, Fetter Lane, London, E.G., either direct or through any bookseller, and communications
respecting advertisements should also be addressed to C. F. Clay.

Till further notice, new subscribers to Biometrika may obtain Vols. I — X together for £11 net — or
ibound in Buckram for £13 net.

The Cambridge University Press has appointed the University of Chicago Press Agents for the sale
of Biometrika in the United States of America, and has authorised them to charge the following prices : —
'$7.50 net per volume ; single parts $2.50 net each.

The following work prepared in the Biometric Laboratory can be
<obtained as a Government Report from Messrs Wymari and Sons, Ltd.

The English Convict, A Statistical Study. By CHARLES GORING, M.D.
Text. Price 9s. Tables of Measurements (printed by Convict- Labour). Price 5*.

Recently iaMtn/. YV/Vv 2ls.

The Life, Letters, and Labours o

Francis Galton

Vol. I. Birth 1822 to Marriage 1853

WITH 5 PEDIGREE PLATES AND 72 PHOTOGRAPHIC
PLATES, FRONTISPIECE AND 2 TEXT-FIGURES

BY KARL PEARSON, F.R.S.

GALTON PROFESSOR, UNIVERSITY OF LONDON

" It is not too much to say of this book that it will never cease to be memor-
able. Never will man hold in his hands a biography more careful, more
complete." — The Times

"A monumental tribute to one of the most suggestive and inspiring men of
modern times." — Westminster Gazette

" It was certainly fitting that the life of the great exponent of heredity should
be written by his great disciple, and it is gratifying indeed to find that he has
made of it, what may without exaggeration be termed a great book." — Daily
Telegraph

Recently issued. Price 9s. net.

Tables for Statisticians & Biometricians

EDITED BY KARL PEARSON, F.R.S.

GALTON PROFESSOR, UNIVERSITY OF LONDON

ISSUED WITH ASSISTANCE FROM THE GRANT MADE BY THE WORSHIPFUL
COMPANY OF DRAPERS TO THE BIOMETRIC LABORATORY.

" To the workers in the difficult field of higher statistics such aids are invaluable. Their
calculation and publication was therefore as inevitable as the steady progress of a method which
brings within grip of mathematical analysis the highly variable data of biological observation.
The immediate cause for congratulation is, therefore, not that the tables have been done but

that they have been done so well The volume is indispensable to all who are engaged in

serious statistical work." — Science

" The whole work is an eloquent testimony to the self-effacing labour of a body of men and
women who desire to save their fellow scientists from a great deal of irksome arithmetic; and
the total time that will be saved in the future by the publication of this work is, of course,
incalculable To the statistician these tables will be indispensable." — Journal of Education

" The issue of these tables is a natural outcome of Professor Karl Pearson's work, and apart
from their value for those for whose use they have been prepared, their assemblage in one
volume marks an interesting stage in the progress of scientific method, as indicating the number
and importance of the calculations which they are designed to facilitate."— Post Magazine

CAMBRIDGE UNIVERSITY PRESS

C. F. CLAY, MANAGER
LONDON: FETTER LANE, E.G. EDINBURGH: 100, PRINCES STREET

H. K. LEWIS AND CO. LTD., 136, GOWEB 8TBEET, BOMBAY, CALCUTTA AND MADBAS :

LONDON, W.C. MACMILLAN AND CO., LIMITED

WILLIAM WESLEY AND SON, 28, ESSEX STBEET, TORONTO I J. M. DENT AND SONS, LIMITED

LOXDON, W.C. TOKYO : THE MARUZEN-KABUSHIKI-KAISHA
CHICAGO : UNIVERSITY OF CHICAGO PRESS

II I III III II 1 1 I III Ml ••! IU1 >•• II II • •• •' «

3 1205 00929 0741

THE LIBRARY
UNIVERSITY OF CALIFORNIA

Santa Barbara

THIS BOOK IS DUE ON THE LAST DATE
STAMPED BELOW.

MAR l 0 ,990

Series 9482

OCT ' A