FLUORINE CONTENT OF HUMAN BONES (1)
H. Hamaguchi and M. Tatsumoto
Mr. Murotabe of,the Tokyo University Anthropology Department, who had
bee-n endeavoring to apply chemical analyses to the studies of anthropology,
reported in 1945 on the calcium and phosphorus content of human bone from
the shellmound of Homii  The chemical composition of the hun bone is gen-
erally as follows: approximately 40% organic substances (after dehydration)),
and 60% inorganic constituents, of which approximately 85% is calcium phos-
phate and the remainder chiefly calcium carbonate. Taken as a whole, the
past studies have revealed the following proportions of inorganic consti-
tuents:
SiO2 less than 1%          C02 around 5%              Cl around 0.1%
CaO around 52%      (Fed Al)203 less than 3%        P205 around 42%
Na2O around 1$              MgO less than 2%    CaO:P205 = 10,
F     around 0,1%           K20 around 0.1%
Carnot (1891-2) pointed out that the relationship between the content
of F and P can be used as an indicator to deduce the age of the fossilized
bone, When the respective F:P ratios of the fossilized bones and apatite
are compared, it wili be observed that the older the geological age of the
fossilized bone, the greater the ratio as shown in Table I.
Bemmelen (1897) expressed the opinion that the enrichment of fossilized
bone with F is due either to the conversion of CaCG3 to CaF caused by the
alkali fluoride in the subterranean water or to the adsorption of F on P.
A matter of interest to the archaeologist is Sidersky's discovery (1934)
that the ratio of F.:P in the bones of animals increases in proportion to the
number of years after death (e.g., F is 0.52% of P in present-day bones but
increases to 8,85% in bones from primary geological age) and that consequently
this ratio may be used in estimating the age of the bones. Boyle et-al. (1939)
also conducted some studies on changes in the content of organic substances,
carbonate salts of alkali earth metals and F during the aging processo
We were very fortunate in belng able to conduct some studies with
human bones of known geologic age to see whether the facts mentioned above
in connection with fossilized bone can be verified for bones about 4000
years old. The bones were given us by the Tokyo University Anthropology
Department,
Experimental Procedure
From among the gravimetric, titration and colorimetric methods available
for the determination of small amounts of F. we chose the titration by thorium
nitrate using sodium alizarin sulfonate as indicator. This method requires
41
TABLE I
Ratio of F/P 0  in fossilized bone to F/Pi(Carnot)
- 1.07 ^ 0.9
Tertiary
(
)
(
)
(
)
(
)
(
)
(
Quaternary
Unfossilized
bone
1.90
30.24
36.81
1025
33.24
1e83
35o20
33.83
.o205
40,28
0.0892
OoO892
:   0.0892
:   0o0892
:   0.0892
0.0892
TABLE II
F Content of Human Bones (Tibiae)
Loss of wt.
Period       Upon Heating
F       F     *
Content P205 x 100
Kokufu shell mound
Todoroki  "    n
Yatsukuri n    n
Yoyama         n
Tsukumo   n    n
Inariyama n    i
Okitsu - cave
Tagagun - hole
Kanekoyama - hole
Early Jomon
n    a
Late Jomon
if  fn
Final Jomon
it   Is
Tomb
Late
n
1949
Period
Tomb Period
n     nit
content was determined by M. Ashikaga
yet unpublished by him, P content is
Nos. 7, 9, 10 are incomplete and were
of the Chemistry Department and
40-44% and the value declines with
based on P205 content of 44%.
42
Paleozoic
= 0,70
_ .0.' 63
0 .O65
= 0,58
0035
= 0.06
Sample
No.
Location
1
2
3
4
5
6
7
8
9
10
12985
25 71
17055
21.82
11.05
14o06
19.04
30.77
24.11
37,53
* The P
is as
age.
0011
0,26
0.25
0,27
0021
0.22
0024
0*21
0,22
0.07
0.27
0.62
0.62
0.65
0,52
0,52
(.o54)
0.48
(0.50)
(0015)
- .01
the removal of interferring substances prior to F determination. For this
purpose distillation of hydrofluosilicic acid as described by Willard and
Winter (1933) is suitable, The method has been widely applied and improved
by many later workers,  We chose the method employed by Okuno (1941) who
investigated the F content of hot spring water, and by McClure (19395, who
investigated the F content of bone ashes0
The following distillation apparatus is employed. Distillation flask B
has a capacity of 100 ml. and is covered with asbestos to preserve heat. A
thermometer is hung in the flask by a platinum wire to measure the tempera-
ture of the distilling liquid. Flask A contains water with a small amount
of NaOH and is used to generate the steam. The amount of steam going to B
is controlled by a pinchcock D. To heat flask B, electricity rather than
the burner is preferred to minimize the fluctuation in the temperature. The
temperature is kept at 1351404 C so that 150 cc, of distillate comes over
in 1 to lj hours,
Before distillation, about 20 cc, of HNO3, a few glass beads and a
small amount of AgNO3 (enough to catch all chlorides in the sample) are added
to flask B and blank distillation is carried outo The volatile impurities in
HN03 which emerge as a result of the initial abrupt rise of temperature are
discarded and then 150 cc. of distillate is collected, Then, a good portion
of the HN03 is taken up with a pipette and the weighted amount of the sample
powder is addedo The HN03 in the pipette is used to wash down the powder,
Distillation is continued until 150 cc. is collected0 This step is followed
by the distillation of a known amount of purified NaF until 150 cc, has
been collected. The receiver C contains 20 cc. of distilled water alkalized
to phenolphthalein with NaOH,
The slightly alkali distillates are placed in a platinum evaporating
dish and heated to dryness on a steam bath. The residue is dissolved in
10 ml. of 50% alcohol, after which H10 (1:50) is added dropwise to remove
the red phenolphthalein color, 0.1 cce of 0,05% solution of sodium alizarin
solfonate is then added; if this indicator turns orange more H01 is added
until a yellow color appears,  1 cc. of mono ClHO -NaOH buffer (pH 305) is
added and the solution titrated with a standard thorium nitrate until the
color changes from yellow to pale pink, Incidentally, the titration of the
blank is carried out beforehand to correct the amount of thorium nitrate
consumed, The blank also serves as a standard for the end point, The
correction for the recovery is made from the data on distillation of the
known amount of Fo
The procedure may be elucidated by the actual experimental data,   The
distillate on 0,5813 gm. of sample was dried and taken up with 10 cc. 50%
alcohol. It required 1.62 cc. of thorium nitrate for titration0 The blank
took 0.03 cc. (1 drop)0 Therefore the actual amount required is 1.59 cc,
The distillate from 1 cc. of standard NaF (F 1,82 mg.) required 4.45 cc,
The standard thorium nitrate solution has a concentration of 2,9024 gmn
Th(N03)4 4H20 per 1 and requires 2,51 cc, to titrate against 1 mg, of F.
Therefore, the recovery of F is 97% and the F content in the sample is 0,11%.
43
A drop of thorium N03 is 0.03 cc. which is equivalent to 0.01 mg. of F.
When the sample is 0.1 g. accuracy of more than 0.01% is not attainable.
Ruaisusi
The results of F analysis are shown in Table II. The values we obtained
for the F content are within the range of F. J. McClure's (1939) of about
0.01-0.5% for bone ash, and of Carnot's (1891-2) 0.17-0.3% for fresh bones
(human, sea-cow, cow, elephant).
The samples are all tibiae and the list is in chronological order, with
the oldest being at the top. Table II indicates that the F content is lowest
in recent bones and higher in all others, with the exception of the oldest
Kokufu-shellmound bones, the proportion of F in the other 8 samples varying
from 0.21 to 0.27%. If one were to push the data a little, one might say
that there is a tendency for Nos, 2, 3, 4 to be somewhat higher in F
content than those after No, 5. This difference is more clearly brought
out in the last column where the ratio of F content to P content is shown,.
The variation in F:P ratio may in part be accounted for by the tendency for
the P content of older bones to be proportionally lower0
Although at first glance it seems that the findings of Carnot (1891-2),
Bemmelen (1897) and Sidersky (1934) can be verified with the samples dealt
with here, but definite conclusions cannot be drawn since the number of samples
was too small. The opinion of the authors is that the determination of the
age of bones by measuring the F content is not practical when the differences
in age are as small as those involved in our experiment,
We are deeply indebted to the Anthropology Department for the bone
samples, to Dr. Hasebe as mediator, and to Mr. Uchiyama and Mr. Tanabe for
their helpful suggestions. A part of the research fund was donated by the
Department of Education,
44
NOTZi
(1) This article originally appeared in Zinr'iugaku Zassi (Journal of the
Anthropological Society of Japan), Vol. 61:1-4 (1950).
BIBLIOGRA
Bayle, E., L. Amy, M. Rondeau du Noyer
1939
1939
Bull, soco chime, 6, p. 1011.
Chem, Abst., 33, p. 6757.
Bemmelen, J. Mi. van
1897   Z. anorg. allgem. Chem., 15, p. 90.
Carnot, Ado
1891-2  Comt, rend., 114 p. 1189; 115, p. 243, p. 337.
Hisateru Okuno
1941   Journ. Chem. Soc, Japan, 62, p. 234.
McClure, F. J.
1939   Id. Eng. Chem., Anal. Ed. U,1 p. 171.
Sidersky, D.
1934   Bull. assoc, chim. sucr, dist., 51, p. 514.
1935   Chem. Abat., 29, p. 1366.
Giichi Tanabe
1944 J. Anthrop. Soco. Japan.
1945   J. Anthrop. 59, p. 1.
Willard, H. Ho, 0. B. Winter
1933   Ind, Eng. Chemo,, Anal. Ed. 5, po 7.
45