716 
ANTHROPOLOGY: F. BOAS 
these data with the information given in the Italian census in regard 
to the immigration into cities. I found throughout that the variabiHty 
of head-form in each city is smaller than would be found in a popula- 
tion in which all the constituent genetic types were present without 
physiological modification. This result has been criticised by Corrado 
Gini, on the basis that in former times migration was less than what 
it is now. I grant this point; but nevertheless it is quite obvious, that, 
although no exact data are available, the mixture of population in a 
city like Rome or like Florence must be very great, since the political 
conditions for the conflux of Italians, and even of individuals from 
outside of Italy, have been favorable for a very long period. If this is 
true, we should expect a very high degree of variability in Rome, which, 
however, is not found. 
Turning to new data, I wish to mention the observation made by 
Dr. Hrdlicka, who, in a paper read before the Pan-American Scientific 
Congress, has stated that he found the width of face of Americans of 
the fourth generation — that is to say, of descendants of Europeans 
who had no foreign-born ancestor after the fourth generation back — 
was materially decreased as compared to the width of face found among 
European tj^es. This conforms strictly with what I found among 
the descendants of immigrants of all nationalities. 
A year ago I had the opportunity to make an anthropometric investi- 
gation of a considerable number of natives of Porto Rico. This work 
was carried on in connection with the Natural History Survey of 
Porto Rico organized by the New York Academy of Sciences. The pop- 
ulation of Porto Rico is derived from three distinct sources — from peo- 
ple belonging to the Mediterranean type of Europe, from West Indian 
aborigines, and from Negroes. The Mediterranean ancestry of the 
Porto Ricans leads back to all parts of Spain; but among the more 
recent immigrants, Catalans, people from the Balear Islands and from 
the Canary Islands prevail. There are also a fair number of Corsicans. 
The Spanish immigration has been quite strong even up to the present 
time. Among the individuals whom I measured, 14% had Spanish- 
born fathers, some even Spanish-born mothers. From all we know 
about the history of the people of Porto Rico, we must consider them 
essentially as descendants of male immigrants who intermarried with 
native women. It is evident that in early times this must have led to 
the development of a Mestizo population, in which, however, the amount 
of Indian blood must have decreased very rapidly owing to the con- 
tinued influx of Spanish blood, and the elimination from the reproductive 
series of the male Mestizo element. The Negro population is settled 
