718 
CHEMISTRY: BAXTER AND STARKWEATHER 
the so-called ^Indiera/ Indian types have persisted to a greater extent 
than elsewhere. I have not been able to find any definite indication of 
a difference in type; but I have measured only a few individuals from 
these districts. The material that I have been able to study comes 
from all parts of the island, but principally from the western-central 
part. The phenomena here described occur with equal intensity in 
all parts of the island. 
The question of the degree of instability of human types seems to 
my mind an exceedingly important one for a clear understanding of the 
problems of physical anthropology. It would be particularly desirable 
to study the problem among immigrants living in different rural com- 
munities of the United States, and it would be even more desirable to 
have information in regard to the types that develop among the East 
Europeans and South Europeans who return to Europe and settle 
in their old geographical environment. 
A REVISION OF THE ATOMIC WEIGHT OF TIN 
By Gregory Paul Baxter and Howard Warner Starkweather 
COOLIDGE MEMORIAL LABORATORY, HARVARD UNIVERSITY 
Read before the Academy, November 14, 1916. Received, November 23, 1916 
A recent investigation upon the atomic weight of tin by Briscoe,' 
in which stannic chloride was compared with pure silver, yielded a 
very concordant series of results, with an average value 118.698 (CI = 
35.457). This value has been adopted by the International Committee 
on Atomic Weights in preference to that found by Bongartz and Classen,* 
119.0, which has been in general use for some time. Since the electro- 
deposition of cadmium and zinc in a weighed mercury cathode has 
been found to be a process capable of great accuracy,-^ and since tin 
amalgam promised to be unusually well adapted for quantitative han- 
dling, this electrolytic method has been applied to the analysis of stannic 
chloride also. 
In brief the method of operation was as follows: Pure tin was con- 
verted to tetrachloride by treatment with pure chlorine, and the tetra- 
chloride was purified by fractional distillation. After weighed portions 
of the chloride had been dissolved in dilute hydrochloric acid, the 
metal was deposited electrolytically in a mercury cathode contained 
in a weighed glass cell similar to that previously described,^ 
Before converting the tin to tetrachloride it was freed as far as pos- 
sible from acid-forming elements by twice transporting the metal elec- 
trolytically through an acid solution of stannous chloride, the anode 
