42 BUREAU OF AMERICAN ETHNOLOGY [Bull. 184 
with two exceptions. Of the population of Sia, 62.5 percent was 
under 25 years of age (1957) as compared with 51.7 percent for New 
Mexico (1950). 
The data show that Sia closely resembles its Indian neighbors with 
regard to age distribution of population. They show also that Sia 
has a young population; and, at least for the age group under 20, it 
has become even younger since 1930. This may be due to decrease in 
infant mortality in recent years, coupled with considerable increase 
in population. 
SEX RATIO 
The percentage of males at Sia has exceeded that of females, some- 
times very greatly, for as long as data are available. But Sia is not 
exceptional in this respect by any means; the same situation has pre- 
vailed in pueblos in the vicinity of Sia (table 5). I am unable to 
explain this preponderance of males over females. In 1930 Superin- 
tendent Towers of the Indian Agency stated that the preponderance 
of males was due “largely to the death of mothers during childbirth, 
as we have been unable to prevail upon the Indian mothers to put 
themselves under the care of Government physicians or attend Gov- 
ernment hospitals during confinement” (U.S. Sen., 1932, pt. 19, p. 
9875). But his own figures do not support this explanation: in 1930 
the percentage of females between the ages of 20 and 39, inclusive— 
the childbearing period—was slightly larger than that for men: 26.3 
percent as compared with 26.0 percent for males. My census of 
1957 would lend some support to Towers’ theory, for it shows that 
51.5 percent of males and 53.9 percent of females were under 20 years 
of age, whereas 29.5 percent of males and 26.9 percent of females were 
between the ages of 20 and 39, inclusive (table 2). But, according to 
our information, from Government doctors and nurses as well as from 
the Sia, few women die in childbirth. I believe we have no adequate 
explanation of the preponderance of males. 
I believe we see a tendency of the predominance of males to di- 
minish between 1890 and 1957 among the five pueblos included in table 
5, although the trend in some cases is erratic. But in each case, 
except Santo Domingo, the number of males per 100 females is small- 
est for 1957. The data strongly suggest that some factor has been 
operating since 1890 toward an equalization of the numbers of the 
sexes, but what this factor may be I have no idea. 
MARITAL STATUS 
We have two marriage censuses of Sia; and marital status could 
be inferred from all census rolls of the Agency, but not with a high 
degree of accuracy. In 1923, Halseth (1924 b, p. 68) counted the 
married, single, and widowed. He did not specify ages; he merely 
