212 ANNUAL REPORT SMITHSONIAN INSTITUTION, 1946 
In most cases, also, more analyses of the individual water supplies 
are needed. It is well known that the fluorine concentration of waters 
fluctuates during the year, the maximum value often being several 
times as great as the minimum. This is true especially for surface 
supplies but also to a lower degree for ground-water supplies. For 
epidemiological studies it is the average value for the year that will 
be of greatest significance. Most of the fluorine analyses now availa- 
ble, however, represent single samples taken at random times through- 
out the year. Some of them may deviate from the average value 
enough to be misleading. 
A different problem is introduced by the marked local variations in 
the factor tobe mapped. The fluorine content of water depends largely 
on the nature of the rocks through which or over which the water has 
passed. Wells only a few feet apart may differ greatly in the 
fluorine content of their water because they were drilled to different 
depths and obtain their water from different aquifers. Supplies 
nearby may be obtained from surface waters, with a still different 
fluorine content. Experiment with a limited area, North and South 
Dakota (fig. 1), has shown that the mapping of all available fluorine 
analyses is impracticable except for detailed local studies. The varia- 
tion within short distances is so great that the regional pattern of 
distribution is blurred, and the construction of isolines is impossible. 
Representation of the individual analyses by point symbols, the only 
alternative, presents a confusing picture and requires a large-scale 
map. A complete representation of the distribution of fluorine in 
water supplies would have to be three-dimensional. 
As a solution to this problem, Dr. Wright suggested that only the 
maximum fluorine content known for each county be mapped. Figures 
2 and 3 show the results for North and South Dakota, the first for 
all water supplies that have been analyzed for fluorine, the second for 
communal supplies only. On these maps a regional pattern of distri- 
bution of fluorine is revealed, and isolines have been drawn for the 
values 0.5, 1.0, and 1.5 p. p.m. These degrees of concentration were 
suggested by Dr. Dean as being of greatest significance as regards 
dental health. The highest dental-caries experience rates are associ- 
ated with fluorine concentrations of 0.0 to 0.4 p. p.m. Resistance to 
dental caries is noticeably greater when the concentrations are 0.5 to 
0.9 p. p. m. and reaches its maximum at concentrations of 1.0 to 1.4 
p. p.m. With the latter values there is a very slight development of 
mottled enamel in a small percentage of the population, and higher 
values cause increasing development of this malady. 
For the country as a whole the lack of information for large areas 
precludes the drawing of isolines. In plate 1,C and D, therefore, color 
is used to represent the known maximum fluorine content of water 
