JULY 30, 1897.] 
the phenomena of life are coming into so much 
favor among students of evolution, the present 
volume possesses a distinct value as a biological 
treatise, in addition to its obvious practical 
utility. It may serve at once as an encour- 
agement and a warning to ardent statisticians, 
because, while many of the facts it sets forth 
are important and interesting, the closer it is 
studied the stronger grows the conviction that 
vital statistics need to be examined with ex- 
traordinary caution before being accepted at 
their ‘face-value.’ Homo sapiens is an animal 
whose habits are better known to us than those 
of any other, and we have exceptional facilities 
for estimating the factors which control his de- 
velopment and evolution. Accordingly, we 
readily perceive the more important probable 
sources of confusion in the statistics, as, for ex- 
ample : ; 
(1) The 28 cities of the size indicated in the 
title had an aggregate population of 9,697,960, 
of which 4,850,653 were males and 4,847,307 
females. Thus it might be inferred that the 
males of the human species were more numer- 
ous than the females, but for the recollection 
that large cities attract great numbers of men 
who either have no families or leave them else- 
where. Taking the cities separately, the ma- 
jority had more females than males, but the 
following eleven had an excess of males: Buf- 
falo, Chicago, Cleveland, Denver, Jersey City, 
Kansas City, Minneapolis, Omaha, Pittsburg, 
St. Louis (white) and St. Paul. It is signifi- 
cant that while Omaha had a population of 
80,108 males and only 60,344 females, those 
under one year of age consisted of 1,411 males 
and 1,422 females, showing an excess of female 
births, in accordance with current expectations. 
It is, however, very remarkable to find that 
in a number of cities there was an excess of 
male babies under one year of age, showing ap- 
parently an excess of male births. Thus Jer- 
sey City had 2,007 males and 1,966 females un- 
der one year; Chicago had 15,281 males and 
14,823 females under one year. But this re- 
sult is further complicated by the fact that 
there was a considerable excess of male mortal- 
ity under one year ; thus in Chicago the deaths 
under one year were 233.95 males and 190.72 
females per 1,000 population of corresponding 
SCIENCE. 
175 
age. In every city the male mortality under 
one year exceeded the female, often greatly ex- 
ceeding it. Thus including still-births, we get 
the following striking results: 
Total deaths wnder one year. 
Males. Females. 
Buffalo, 1,069 817 
Cleveland, 1,155 835 
Philadelphia, 3,707 3,055 
New Orleans, 1,084 844 
and so forth. 
Now how are we to account for an excess of 
male births, combined with a remarkable excess 
of male mortality, during the past year? I 
have heard it stated that the slightly larger 
average size of the male head leads to an in- 
crease of the number of still-births, but this ob- 
viously will only account for a small proportion 
of the facts here before us. It seems to the 
present writer that some light may be thrown 
on the problem by the theory, adopted by some 
biologists, that the sex is determined at an 
early age by the conditions of the nutrition of 
the germ, favorable conditions producing a pre- 
ponderance of females, unfavorable of males. 
If this is so, immediately we see the meaning 
of the above statistics. Omaha, which was the 
most healthy of the towns discussed, according 
to the data given, had an excess of female 
babies ; other cities, notoriously containing un- 
healthy and crowded wards, show an excess of 
male babies. Further, if disadvantageous con- 
ditions lead to an excess of males, it is easy to 
understand why those males, on the average, 
should be less able to survive the first year. 
Hence the excess of male deaths, even when 
there was an excess of female births. It is to 
be supposed that those families and wards in 
which occurred the greater part of the male 
deaths would show an excess also of male 
births, though the city as a whole might not 
show it. 
(2) Comparing 1880 with 1890, there is shown 
a great increase of infant mortality. Thus, for 
example: 
Proportion of deaths under 1 year per 1000 births. 
Newark, under 150 in 1880, over 250 in 1890 
Vasey Chhny, CO alizisy OF OGG OS UG 
Denver, seal DOU een aes 75 20005) =< 
Rochester ‘‘ 100 ‘* ‘* pom berh tree 
and so forth. 
