204 THE TREND OF THE RACE 
rate is allowed to take its natural course large families offer some 
evidence of physical vigor whatever they may indicate as to 
mentality. A general reduction of the birth rate has, therefore, 
its dangers, at least for the physical vigor of the population, since 
it would probably involve a greater proportionate reduction of 
healthy and vigorous stocks. 
It would indeed be unfortunate if a reduction of the birth 
rate in the larger families would lead to the reduction of the 
best members of the stocks in addition to the loss of physical 
vigor otherwise involved. Whether ordinal position in the family 
except in the matter of weight, size and infantile death rate, is 
per se a handicap is a question which most of our data do not 
enable us to decide. The fact that there is a greater percentage 
of deaths among the first born than there is among the second or 
third born does not prove that the second or third born member 
of any particular family is less likely to die than the first born. 
The large percentage of deaths among the first born may be due 
to the fact that a large proportion of early deaths occur in families 
containing only one or two children. The data do not prove that 
in families in which three or four children are born the later 
children have any greater expectation of life than the first. As 
we have already pointed out fecundity is correlated with longev- 
ity. Families limited by the early death of one or both parents 
would naturally show a high death rate on account of the prob- 
ability that the offspring would inherit a diminished vitality. On 
the other hand, large size of family very commonly has a very 
undesirable relation to infant mortality, despite the vitality of 
the stock from which large families come. This is due in part at 
least to economic causes and in part to the correlation between 
mental subnormality (this does not imply reduced physiological 
vigor) with a high birth rate. Where large families occur among 
intelligent and thrifty people as they did a century ago, there is 
much less correlation between size of family and a high early 
death rate. The following table from data collected by Dr. A. G. 
Bell is instructive in this connection: 
