FLUORINE IN WATER SUPPLIES—VAN BURKALOW 213 
supplies in each county, in terms of the same four degrees of concen- 
tration. 
Several defects are inherent in this method of mapping. First of 
all, because of the incompleteness of the data, the maximum fluorine 
concentration on record for a county may not be the maximum value 
that really exists there. In a county represented on the map as hav- 
ing a maximum fluorine concentration in the range from 0.0 to 0.4 
p. p.m. there may actually be water supplies in use that contain more 
fluorine. On the other hand, all values less than the maximum are 
omitted, though they may be more representative of general condi- 
tions in the county. It is essential to keep this point in mind; for the 
technical device here used of depicting the maximum fluorine value for 
a county by shading the entire area of the county may be wrongly 
interpreted as meaning that that value applies to all places within the 
county. That this is not so will easily be realized by comparing 
figure 1, where all known fluorine values for the Dakotas are shown, 
with figure 2, where only the maximum value for each county is 
mapped. 
For practical use in epidemiological studies this method is defective 
also; for it does not reveal the nature of individual water supplies and 
the number of people affected by them. It does, however, show in a 
general way the sections of the country that suffer from either excesses 
or deficiencies of fluorine. For studies of conditions in specific areas 
detailed large-scale maps will be necessary, and in many cases these 
cannot be made until more water supplies have been tested for fluorine. 
THE SOURCE OF THE FLUORINE 
In rock analyses, as in water analyses, fluorine has seldom been 
mentioned until recently because of the lack of accurate methods of 
determining it in small amounts. Clarke ™ estimated that it makes 
up about 0.027 percent of the earth’s crust. However, more recent 
investigations by Shepherd,” based on new methods of analysis, in- 
dicate that fluorine is more abundant than this, making up at least 0.04 
percent of the earth’s crust. 
Information on the amount and distribution of fluorine in rocks is 
still far from complete, but enough is known to indicate that it is 
present in markedly differing amounts in different kinds of rocks. 
Shepherd found the highest average concentration (0.07 percent) in 
obsidian; the lowest average (0.01 percent) was found in more fluid 
lava flows, from which the fluorine evidently escaped with ease. Of 
course, fluorine contents much greater than these averages have been 
4 Clarke, F. W., The data of geochemistry, U. 8. Geol. Surv. Bull. 770, 5th ed., 1924. 
12 Shepherd, E. S., Note on the fluorine content of rocks and ocean-bottom samples, Amer. 
Journ. Sci., vol. 238, pp. 117-128, 1940. 
