FLUORINE IN WATER SUPPLIES—VAN BURKALOW 217 
phosphates be formed. He finds that the major phosphate deposits 
of the United States can be related to periods of volcanic activity 
known from pyroclastic deposits. 
These conclusions suggest a possible explanation for the wide differ- 
ences in the fluorine content of sedimentary rocks of different ages 
and the variation from place to place in rocks of one age. Fluorine 
from volcanic sources would be supplied intermittently and would be 
unevenly distributed in space. Its presence in the rocks would not, 
however, necessarily ensure its entrance into the ground water. In 
volcanic-ash deposits the fluorine would be present in minerals such 
as apatite, tourmaline, and some of the micas. That washed out of 
the atmosphere as a rain-water solution of hydrofluoric acid would 
enter into new combinations when it reached the ground, much of it 
probably forming fluorite. Because minerals such as these are rela- 
tively insoluble, fluorine, even though present in considerable amounts, 
could not enter the ground water until the minerals containing it had 
been decomposed. Gwynne** has suggested that much of this may be 
accomplished by the attack of sulfuric acid formed by the decom- 
position of pyrite. In sedimentary rocks pyrite is commonly found 
in fairly large amounts in association with lignite and coal beds, 
evidently formed by reaction between iron minerals and hydrogen 
sulfide produced by the decay of organic material. Fluorine in soluble 
forms should therefore be expected in large amounts where two condi- 
tions prevail: (1) an abundance of fluorine-containing minerals, sup- 
plied either directly from magmatic sources, directly from volcanic 
sources (probably furnishing the greatest quantity), or indirectly by 
the weathering and erosion of igneous rocks; (2) an abundance of 
pyrite, usually found in association with concentrations of organic 
material, to facilitate the decomposition of the fluorine minerals. If 
this hypothesis is correct, it should be possible to demonstrate that 
many aquifers yielding fluorine-rich waters were formed during 
periods and in areas of marked volcanic activity, and that they also 
contain abundant pyrite. 
Such a correlation requires first of all a knowledge of the partic- 
ular aquifer from which a water supply is derived. From a purely 
practical viewpoint also, this would be desirable as an aid in determin- 
ing the sources with the most beneficial fluorine concentrations and 
those to be avoided because of harmful fluorine concentrations. Too 
often, however, the exact source of the water is unknown. Many 
well logs are missing or incomplete. Even when they are available 
and accurate, the well often passes through several water-bearing 
layers; and unless it is cased and the exact nature and depth of the 
#% Gwynne, C. S., Geological significance of fluorine in Iowa well-waters, Pan-American 
Geologist, vol. 62, pp. 189-140, 1934. 
