Title: Geochemistry of uranium in apatite and phosphorite
Full Citation
Permanent Link: http://ufdc.ufl.edu/UF00000524/00001
 Material Information
Title: Geochemistry of uranium in apatite and phosphorite
Series Title: Geochemistry of uranium in apatite and phosphorite
Physical Description: Book
Creator: Altschuler, Z.S.
 Record Information
Bibliographic ID: UF00000524
Volume ID: VID00001
Source Institution: University of Florida
Holding Location: University of Florida
Rights Management: All rights reserved by the source institution and holding location.
Resource Identifier: ltqf - AAA1635
ltuf - AME1672
alephbibnum - 002436516


This item has the following downloads:

Binder1 ( PDF )

Full Text

eochemistry of Uranium

n Apatite and Phosphorite

TAhis report concerns work done on behalf CST F
of the U. S. Atomic Energy Commission 'A
and is published with the permission 0
of the Commission



FRED A. SEATON, Secretary


Thomas B. Nolan, Director

Altschuler, Zalman Samuel, 1919-
Geochemistry of uranium in apatite and phosphorite, by
Z. S. Altschuler, R. S. Clarke, Jr., and E. J. Young. Wash-
ington, U. S. Govt. Print. Off., 1958.
iv, 45-90 p. illus., plates, tables. 29 cm. ([U. S.] Geological
Survey. Professional Paper 314-D. Shorter contributions to general
Bibliography: p. 85-87.

1. Uranium ores. 2. Apatite. i. Title. (Series: U. 8.
Geological Survey. Prosessional Paper 314-D. Series: U. S. Geo-
logical Survey. Shorter contributions to general geology)

For sale by the Superintendent of Documents, U. S. Government Printing Office
Washington 25, D. C. Price 60 cents (paper cover)


ledgments .--------------- -------------
ogy of apatite---------------------------
tal chemistry ...---------------------------
opmposition of the sedimentary carbonate-apatites. -
graphic terminology -----------------------
.um content of apatite -----------------------
,?actors affecting equivalent uranium determinations-
,Uranium content of representative apatites ---------
Island phosphorites-secondary uptake of uranium
and fluorine- ------- ----------------------
bblems of occurrence of uranium in phosphorite -----.
Possible modes of occurrence --------------------
Absence of uranium minerals--------------------
Relation of uranium content to composition of
phosphorite -------------------------------
Fluorescence and uranium content -----------
Lack of preferential solution of uranium -------
Relation of uranium to P05 ------------------
Marine versus ground-water origin of uranium in
General statement and review-----------------
Evidence of young phosphorites and submarine
Necessity of geologic interpretation -----------
nciples governing uranium distribution in phospho-
Uranium in phosphorite during weathering
Florida Land Pebble phosphates-general ge-
ology -----------------------------------
Illustration of leaching----------------------
Postdepositional enrichment ----------------
Aluminum phosphate zone-lateritic weath-
ering and supergene enrichment -----
The Cooper marl-residual enrichment ----
Contrasting modes of enrichment in mild
versus intense weathering. --_--------
Brown-rock phosphates of Tennessee-re-
sidual concentration ------------------
Secondary emplacement of uranium in bones -------
Marine enrichment and reworking ----------------

Crystallo-chemical nature of uranium in apatite --------.
General statement -----------------------------
Substitution of uranium for calcium in apatite ------
Theoretical considerations ------------------
Relation of uranium to phosphorus --------
Oxidation state of uranium in apatite ---------
Chemical problems ---------------------
U(IV) content of sedimentary apatites ..----
U(IV)/U percent reflect geologic history... -
Primary marine uranium in young phos-
phorites is mainly U (IV) ---------------
Subaerial weathering alters U(IV) to
U(VI) in marine phosphorite -----------
Possible relation of U(VI) content to radio-
activity and age ----------------------
Secondary sources of U(VI) in phosphorite -
Relative structural favorability of U(IV) and
U(VI) in apatite--------------------------
Test of uranium and fluorine interdependence in
Effects of reworking on U(IV) content of apatite ----
Summary of evidence bearing on the primary state of
uranium in marine phosphorites ----------------
Uranium in igneous apatite --------------------------
General statement and analyses ------------------
Variation of uranium in apatite with uranium in host
rock -----------.........-------------------------
Oxidation state of uranium in igneous apatite ------
U(IV)/U(VI) ratio reflects magmatic equilibria -----
Relation of U(IV)/U(VI) ratio to age------------
Auto-oxidation -----------------------------
External weathering controlled by radioactive
decay ...-------------------------------
Conclusions ----------------------------------------
Summary --------------------------------------
Nature of uranium fixation in phosphorites ---------
Regenerative capture --------------------------....
Equilibrium fixation of uranium in igneous apatite. -
Selected references ---------------------------------
Index ..-----...------------------------------------












[Plates follow page 60]

PLATE 11. Sections and autoradiographs of sandstone, bone, and pebbles.
12. Contact of aluminum phosphate zone and overlying sand mantle, and mined-out area of Tennessee brown-rock phospl
13. Photomicrographs of arkosic sandstone with carbonate-fluorapatite cement from the Gas Hills area of Wyoming.

FIGURE 3. The structure of fluorapatite__--------------------------------------------------------------------
4. Relation of uranium to P205 in Mona Island cave phosphorites------------------------------------------
5. Relation of uranium to fluoridation in Mona Island cave phosphorites------------------------------------
6. Relation of uranium to P205 in phosphorites from Oulad-Abdoun Basin, French Morocco ------------------.
7. Stratigraphic relations in the Land Pebble phosphate field, Florida-------------------------------------
8. Sampling scheme used for study of uranium distribution in pebbles --------------------------------------
9. Distribution of oxides through aluminum phosphate zone at Homeland, Fla-------------------------------
10. Uranium distribution in various textural components at Homeland mine, Fla ------------------------------
11. Contrasting modes of uranium enrichment and weathering in phosphorites_
12. Sample locations for comparison of brown-rock phosphate with parent limestone in Tennessee---------------
13. Uranium versus excess fluorine in Phosphoria formation--------------- -------------------------
14. U(IV) versus excess fluorine in various phosphorites ---------------------------------------------------
15. Total uranium in rock versus total uranium in apatite for rocks from Boulder Creek batholith --------------.


TABLE 1. Resources of some well- known uraniferous phosphorites-------------------------------------------------
2. Chemical analyses of sedimentary apatite pellets from the Bone Valley formation ---------------------------
3. Semiquantitative spectrographic analyses of sedimentary apatite pellets -----------------------------------
4. Thorium content of Florida pebble phosphorite ---------------------------------------------------
5. Uranium content of representative apatites -------------------------------------------------------
6. Partial analyses of cave phosphorite from Mona Island, Puerto Rico-------------------------------------
7. Comparison of fluorescence and uranium content in composite samples of apatite nodules from the Bone Valley
formation, Boyette mine, Florida--------------------------------------- --------------------
8. Size distribution and uranium contents of pelletal phosphates from Khouribga, French Morocco--------------
9. Correlation coefficients among eU, P205, F, CO2 and organic matter in samples of the Phosphoria formation..---
10. "Inferred uranium contents" of Polish phosphorites-----------------------------------------------------
11. Phosphate and uranium contents of bottom samples from the Gulf of Mexico ------------------------------
12. Relative contents of uranium and P205 in marine phosphorites-------------------------------------------
13. Chemical analyses showing distribution of uranium within phosphate pebbles ------------------------------
14. Examples of phosphate- and uranium-enriched hardpan in Cooper marl near Charleston, S. C----------------
15. Comparison of phosphatized and primary Cooper marl from Lambs, South Carolina ------------------------
16. Variations of enrichment with weathering in phosphatized Cooper marl, South Carolina --------------------
17. Semiquantitative spectrographic analyses of phosphatic hardpan from the Cooper marl----------------------
18. P2Os and uranium content of weathered and fresh Tennessee phosphates, Akin mine, Columbia, Tenn--------
19. Uranium content of representative elastic apatite from the land-pebble phosphate field, Florida-
20. Iron and uranium content, in weight percent, of representative apatites from the Bone Valley formation, Florida.
21. The recovery, in weight percent, of uranium(IV) from 1.5 M phosphoric acid solutions of natural and synthetic
apatite -----------------------------------............-------------------------------------------------------
22. Percentages of uranium and of U(IV) in sample B.L.-3 before and after oxidative heating-
23. Total and tetravalent uranium content of sedimentary apatites -------------------------------------------
24. Experiment showing reduction of U(IV) by phosphoric acid solutions of hematite-rich apatite samples--------
25. Data on excess fluorine and U(IV) in miscellaneous apatite and phosphorite.----------------------------
26. Relative contents of U, U(IV), and P205 in phosphatic terrace gravel in the vicinity of Charleston, S. C-------
27. Uranium and U(IV) content of igneous apatites -------------- -----------------------------------





Apatite contains only traces of uranium, yet as apatite is a
minor constituent in most rocks and the major constituent of
a few very large deposits, it accounts, paradoxically, for both
dispersal and concentration of uranium in nature.
Uranium typically makes up 0.001 to 0.01 percent of primary
igneous apatite and 0.005 to 0.02 percent of sedimentary marine
apatite. Marine reworked apatite becomes enriched in uranium
to as much as 0.1 percent. This is demonstrated by the greater
uranium content of the texturally more complex phases within
a single deposit.
Uranium can be secondarily leached from or introduced into
apatite by ground water. These secondary changes are indi-
cated by pronounced concentration gradients within single peb-
bles of apatite as well as by the redistribution of uranium
among different mineral hosts in leached and altered deposits
of phosphorite. The postdepositional enrichment of uranium in
phosphorite may be entirely residual as in the Tennessee
"brown-rock" deposits, or greatly enhanced by ground-water
additions as in the South Carolina phosphates. Moreover, the.
pattern of enrichment reflects the conditions and intensities
of weathering. Highly acid lateritic weathering has caused
supergene enrichment of uranium in the aluminum phosphate
zone of the Bone Valley formation. In contrast, surficial en-
richment characterizes the moderately weathered Cooper Marl
of South Carolina. Isolated fossil bones or phosphate pebbles
may contain almost one percent of uranium as a result of
ground water enrichment. Such enrichment is comparable to
the postdepositional uptake of fluorine by bone or insular
Phosphorite, and in some places the processes are synchronous
and show good mutual correlations as they are codependent on
the some ground water source.
It is proposed that uranium replaces calcium in the apatite
structure. This is indicated by several lines of investigation.
Uranium and calcium contents are parallel in sections of leached
and altered phosphorite. Ionic radii of tetravalent uranium
(0.97 A) and divalent calcium (0.99 A) are virtually identical,
and much of the uranium in igneous, sedimentary, and bone
apatite is found to be tetravalent. Petrographic and chemical
analyses and nuclear emulsion studies have shown that uranium
in apatite is disseminated rather than locally concentrated.
In addition, phosphate deposits are essentially devoid of
uranium minerals.
The igneous apatites contain from 10 to 66 percent of their
uranium as U(IV). In apatite from a suite of related igneous
rocks both the total uranium and the U (IV)/U (VI) ratio vary
as the total uranium in the rock. Thus, the U (IV)/U (VI) ratio
in igneous apatite reflects the prevalent equilibrium conditions
in the crystallizing magma.

In marine phosphorite thus far investigated the tetravalent
uranium ranges from a few percent to more than 90 percent
of the total uranium. Taken alone such statistics suggest a
great variation in the initial U(IV)/U(VI) ratio of uranium
emplaced in apatite. Thus, experimental evidence that bone and
apatite pellets can remove uranyl uranium from solution sug-
gests that much of the U(VI) found in natural apatite may be
of primary origin, fixed as chemically adsorbed uranyl anions.
It is proposed, however, that uranium in marine apatite is
emplaced primarily as U(IV), structurally fixed. This follows
from the fact that the higher U (IV)/U (VI) ratios are found in
the younger unweathered marine apatites, and in apatites re-
cently reworked by marine transgression. U(IV) is readily
oxidized to U(VI), and postdepositional weathering, facilitated
by radioactive decay, has most probably lowered the initially
high U (IV)/U (VI) ratio in many older phosphorites.
Apatite, by effectively removing the small amounts of U (IV)
produced in sea water by reduction of (U02)', causes more
U (IV) to be produced for its own uptake. As the marine apatite
is far from saturated with respect to uranium, it thus Interferes
with the attainment of equilibrium while fixing an unusual
quantity of U(IV). The name regenerative capture is proposed
for this type of concentration in which the fixation of an insig-
nificant valence species interferes with the equilibrium produc-
ing the ion and thus generates a continuing supply for further
uptake and results, ultimately, in unexpectedly large build-ups
of the insignificant trace element in the host mineral.


Apatite contains only trace quantities of uranium,
typically several thousandths to a few hundredths of
one percent. However, as average rock has at most a
few parts per million and average ocean water only one
or two parts per billion, it is immediately apparent that
apatite is an important concentrator of uranium in
nature. The ubiquitous occurrence of apatite as an
accessory mineral in igneous rocks establishes its role
as an important agent, along with a few other minerals,
notably zircon, monazite, and sphene, in dispersing
uranium in rocks during primary differentiation. This
is particularly true because apatite, unlike many other
accessory minerals, is prominent in mafic as well as
felsic rocks. Thus, whereas the uranium content may
be low in the gabbros, the contribution of apatite to the
total budget of uranium in the early differentiates may

55; ~


be correspondingly higher. As apatite is also rather
common in shales and limestones in the form of phos-
phatic nodules, concretions, and fossil replacements,
it is evident that apatite plays a major role in the
geochemical cycle of uranium.
Beside contributing to the dissemination of uranium,
apatite paradoxically is also one of very few minerals
causing immense localizations of this element. It does
so by virtue of its occurrence in vast accumulations of
marine sedimentary phosphate composed essentially of

the mineral carbonate-fluorapatite (Altschuler
others, 1953). Table I illustrates the magnitude
economic potential of these deposits. Thus, as we annm
ally produce more than ten million tons of phospha(
fertilizer in the United States and as an appreciabi
part of this is processed chemically, an opportunity 1
afforded to mobilize and extract many hundreds of toi
of uranium each year. Indeed, uranium as a byprodu(
of triple superphosphate production is now recovered i
several plants in the United States (Barr, 1955).

TABLE 1.-Resources of some well-known uraniferous phosphorites -
Reserves minable at present Additional resources minable under Total resources
changed conditions
Area and formation PsOsI Apatite a U PsOs Apatite 2 U Apatite 2 U h
Thousands Thousands Millions Tho
Millions of long tons of long Millions of long tons of long of long of long
tons tons tons tons *

Florida Land Pebble field, Bone Valley formation,
matrix ...----..---------------------------------- 330 870 87 600 1,600 160 2, 500 25d
Hawthorn formation .....----------------------------- ------- ---------------4,000 11, 000 1, 100 11, 000 1,100
Western field, Phosphoria formation ---------------- 870 2,300 230 5, 800 15, 000 1, 500 17, 000 1, 700
French Morocco --------------------------------- 4,800 13, 000 1,300 6,420 16, 900 1,690 30, 000 3, ood

I Data on PsOs from Le Cornec (1951) for Moroccan deposits and McKelvey and
others (1953), for domestic deposits.
s Apatite data have been computed on the basis of an assumed average PsOs con-
tent of 38 percent in the carbonate-fluorapatite comprising these deposits. Generally
msch compilations show BPL (bone phosphate of lime or tri-calcium phosphate)
and as the BPL is in turn based on theoretical fluorapatite of 42 percent PsOs, it is

The principal aims of this report are three: to survey
the uranium content of a variety of apatite materials
and deposits; to portray the geologic and chemical con-
trols of uranium distribution in phosphorites of various
origins; to attempt an explanation of the occurrence of
uranium in apatite.
We shall be concerned largely with the migration and
fixation of an element that seldom exceeds 0.01 percent
in the rocks studied. However, it is best to begin with
a discussion of some of the restrictions which this low
concentration imposes on the approach taken. Owing
to its unusual properties of radioactivity and conferred
fluorescence, uranium, unlike most trace elements, may
often be followed in the field virtually as one follows
a visible, major, mineral component. The same prop-
erties have allowed the development of analytical
methods (Grimaldi and others, 1952) that permit accu-
rate detection of uranium at extremely low concentra-
tions. Unlike an investigation of a major element,
however, a study of a trace constituent, in an impure
material like phosphorite, is handicapped by lack of
information on bulk composition. Whereas it is true

both fictitious and misleading. This table gives PsOs and apatite that is minable,-
rather than tons of minable rock. The latter term does not correctly reflect absolute
composition due to its dependence on a complex of external factors such as grade
(amount of diluent), thickness, and overburden depth.
* Based on the conservative estimate of 0.01 percent U in the pure apatite. .

that one may obtain the general composition of the
mineral host with moderate accuracy-and even here
the carbonate-apatite problem is a vexing intrusion-
the tiny quantities of uranium dealt with suggest that
most of the trace element assemblage be quantitatively
determined. Otherwise, the immediate environment of
each uranium sample is but poorly known. Obviously,
it is impossible to overcome this difficulty in most cases
because of the prohibitive time and money require-
ments, and the fact that adequate analytical methods
are not available for many minor and trace elements.
Instead, reliance must be placed on a reconstruction of
conditions from field and petrographic facts and on
selection of several elements for specific analysis. Thus,
in studying the phosphorites, uranium contents are
evaluated mainly in relation to P205 content. Where
important, fluorine, iron, calcium, aluminum or organic
matter may also be studied. However, implicit in some
of the work that follows is the assumption that uranium
in phosphate nodules is associated mainly with the
apatite composing the nodules, and that trace to minor
amounts of included carbonates, clays, iron-oxides,


quartz, pyrite, and organic matter contain only insig-
nificant amounts of uranium.
Three lines of evidence validate this assumption:
1. Many phosphorite samples have been analyzed by
two different methods, one, using weak acids, in
which apatite dissolves readily, and clays, pyrite,
iron-oxides or quartz would remain insoluble or
dissolve only on prolonged treatment; the other,
using a mixture of hydrofluoric and nitric acids in
which all of the rock dissolves. In all samples the
uranium determined was essentially the same by
both methods.
2. Chemical analyses of the minerals associated with
apatite in the Florida and Moroccan deposits showed
their uranium content to be insignificant relative
to that of apatite.
3. Nuclear emulsion studies reveal that most of the
radioactivity in phosphate nodules comes from
The field and laboratory studies which this paper
summarizes have been part of the U. S. Geological
Survey's program of investigation of radioactive raw
materials for the Division of Raw Materials of the
U. S. Atomic Energy Commission. We have benefited
greatly from the unpublished studies of many of our
colleagues in the Geological Survey engaged in com-
plementary and parallel studies of phosphorite. We
are especially indebted to V. E. McKelvey and J. B.
Cathcart, whose regional studies have contributed much
important information on the geologic occurrence of
phosphorite and uranium; to W. W. Rubey, whose
earlier unpublished investigations and recommenda-
tions were largely instrumental in implementing the
Survey's program of phosphorite investigations; and
to the late John C. Rabbitt, whose reviews and sug-
gestions in the senior author's unpublished studies are
reflected in much that follows. We have also benefited
greatly from the advice and interest of R. M. Garrels
and F. S. Grimaldi in many helpful discussions of the
chemical problems, from the stimulating discussions of
regional weathering patterns with C. B. Hunt, and
from the thoughtful and constructive reviews of E. S.
Larsen, 3d, and R. P. Sheldon. The following Geo-
logical Survey colleagues furnished analyses: A. B.
Caemmerer, Frank Cuttitta, G. J. Daniels, G. Edging-
ton, Katherine Valentine Hazel, C. L. Johnson, Harry
Levine, A. R. Schrenk, and Helen W. Worthing.
We are deeply grateful to the mining and adminis-
tration staffs of the many companies engaged in phos-
phate mining in central Florida for their wholehearted

cooperation in allowing access to operating quarries
during fieldwork from 1949 to 1954.

It is important in considering the nature of uranium
occurrence in apatite and phosphorite deposits to under-
stand the possible variations in apatite composition.
Basically, apatite is a hexagonal network structure
composed of P04 tetrahedra, a simple anion (usually
fluorine) and a divalent cation, calcium. Beevers and
McIntyre (1946) liken the structure to a honeycomb
(fig. 3) with the vertices made up of vertical Ca-O
columns in which the oxygens are supplied by P04
tetrahedra intervening between columns and linking
adjoining columns together in a continuous structure.
The fluorine or hydroxyl ions occur vertically above
each other within the channels of the honeycomb, where
they are coordinated to additional calciums (three
about each fluorine) that occur in "caves" within the
irregular walls of the network (Beevers and McIntire,
Apatite is host to many substitutions by cations,
anions, and anionic radicals that resemble its normal
constituents in size and charge (McConnell, 1938). As
a result it may depart significantly from the composi-
tion of simple igneous fluorapatite, depending on its
environment of genesis. The fluorine position may be
occupied wholly or in part by hydroxyl. (Chlorine may
also substitute for fluorine; however, as their size dif-
ference is appreciable, a structural rearrangement
occurs in chlorapatite and it has only limited misci-
bility with fluorafa'ft.) In fossil bone, hydroxyl is
eventually replaced by fluorine through ion exchange.
Minor to major amounts of Sr, Mn, Mg, and Pb are
known to replace calcium in apatite, and V04 and AsO0
occur as traces substituting for P04 in apatite and form
analogues of normal apatite in the pyromorphite series
(Palache and others, 1951).
Additional variety is effected by the opportunity for
coupled diadochy in which cationic or anionic replace-
ments causing loss or gain in valence are balanced by
replacements of opposite kind. Thus, in apatites con-
taining rare earths the imbalance created by the substi-
tution of tervalent cerium for calcium is quantitatively
compensated by substitution of tetravalent silicate for
phosphate or by the monovalent sodium for calcium
(Borneman-Starinkevitch, 1938; Volodchenkova and
Melentiev, 1943). An analogous situation exists in
many other phosphate mineral groups. For example,
monazite, the cerium phosphate, contains thorium and
silicate in roughly equivalent amounts.



P04 tetrahedron, 0,, Fluorine, singly Calcium Relation of unit cell
superimposed and superimposed to hexagonal network
FIGURE 3.-The structure of fluorapatite showing progressive stages of completion as viewed looking down the C axis. (Based on drawings of
Beevers and McIntyre, 1946.) a, Incomplete structure showing hexagonal framework and unoccupied channels; b, addition of 3 channel
calciums and a fluorine occupying same plane; c, position of second planar group of 3 channel calciums and 1 fluorine, at a different vertical
height In the structure; d, complete structure.

The fine-grained microcrystalline carbonate-apatites
of the insular and marine phosphorites are best under-
stood in terms of such coupled substitutions. This
apatite is generally characterized relative to fluor-
apatite by a deficit in P205 content of 3 to 6 percent,
an excess of F, OH (or both) of 0.5 to 1.0 percent, and
by the presence of 2 to 3 percent of carbonate. The
exact mode of occurrence of carbonate in apatite is a
controversial question. Hendricks and Hill (1950) have
proposed that the carbonate is adsorbed on the surfaces

of discontinuities within the apatite crystals. It is sig-
nificant, however, that carbonate fluorapatite is demon-
strably smaller in unit-cell dimensions than fluorapatite
(Altschuler and others, 1953). It is felt therefore that
the structural difference revealed by the characteristi-
cally smaller cell must reflect the major and equally
characteristic chemical deviations, and that carbonate-
fluorapatite is a structurally distinct species as proposed
by Gruner and McConnell (1937). Lacking a precise
determination of the position of carbonate in the struc-
ture we shall adopt provisionally the structural formula
which best rationalizes the chemical composition, as


follows: Caio(PO4,COa)6F2-3. Thus excess fluorine
(or hydroxyl) serves to balance the charge difference
created by the substition of (COs)-2 for (PO4)-3.
The chemical and spectrographic analyses in tables 2
and 3 illustrate the chemical nature of the sedimentary
carbonate-fluorapatites. Theoretical fluorapatite has
the following composition:
CaO = 55.5 (includes F for 0 = 1.6)
P205 = 42.3
F= 3.8
The analyses portray the deficiency of P205 relative
to CaO (augmented by Na, Mg, Sr and other
divalent metals shown in table 3), and the excess of F
plus (OH). It should be noted in table 2 that the car-
bonate is substantially all nonleached and therefore
assignable to apatite.
Sedimentary carbonate-hydroxylapatite occurs as
nodules or massive deposits in caves or islands and is
indicative of continental origin. It usually originates
as a replacement of limestone by guano solutions. In
the presence of fluorine, carbonate-fluorapatite is the
stable form and it precipitates as such in marine sedi-
ments (Kazakov, 1937). Furthermore, fluorapatites

TABLE 2.-Chemical analyses of sedimentary apatite pellets from
the Bone Valley formation
[Analyst, R. S. Clarke, Jr.]

B. L.-3, Bonny Lake Wa.-10, Watson mine,
mine, Ridgewood, Fla. Fort Meade, Fla.

1 2 3 1 2 3

Acid insoluble.-- -- --------- --------- 6.6 ----------------- 2.4 ---------
CaO..............-----------------.............. 49. 5 ..... 52.9 51.5 ..... 52. 7
P20o ----------------------- 34.9 ----- 37.3 36.6 --. 37. 5
CO2 ...--------------------------- 2.1 2.0 2.2 1.9 1.7 1.09
SiOs (total)---------.... ------.... 6. 7 -----.. --- 2.9 --..- --.-----.
Si02 (soluble) -. ----.. -- --------- .8 .9 1.1 1.1
S 3 --..---------------------.. 3 ----- .3 .1 ----- 1
A1203----------- -------------. 1.4 --. -- 1.5 1.1 ----- 1. 1
FeOs3 -------------------------- 4 ----- .4 .9 ----- .9
Na --------------------------- 1 ----- .1 .2 .. 2
K20 ---.---.-.---------------- .1 ----- .1 .2 ----- 2
20 (+)------------------------ 1.6 ----- 1.7 1.8 ----- 1.8
H20 (-)--------------------- 1.0 .6 1.1 .7 .5 .7
-..-- -- -----....... ...... 016 ----- .017 .0075 ----- .0077
F..------------------------------ 3.8 4.1 3.8 --..--- 3.9

Total--------------------- 101.9 ----- 102.6 101.7 ----- 102.1
F=O 1.6 ----- 1.7 1.6 ----- 1.6

Corrected total---------- 100.3 --- 100.9 100.1 ... 100. 5
Nonleachable C02 -------- --------- 1.8 --..---- --------- 1.6 ---------

1. Analysis by complete decomposition of sample by solution in HNO3 and fusion
of insoluble residue with Na2C O3.
2. Partial analyses of same material. Acid insoluble determined after boiling sample
for 20 minutes with 1+3 HC1. Nonleachable carbonate determined after treat-
ment in 0.5 M tri-ammonium citrate (Silverman, Fuyat, and Weiser, 1952).
3. Corrected analysis, free of insoluble residue. Microscopic examination and the
two sets of SiOs figures establish that acid insoluble is essentially quartz.

TABLE 3.-Semiquantitative spectrographic analyses of sedimentary
apatite pellets

Weight B. L.-3 Wa.-10 3 Va.-7 3 K.-1 4 K.-15

Over 10.0 Ca, P -.. Ca, P ..... Ca, P....... Ca, P---.. Ca, P.
10.0-5.0--.. -- Si---- ---------- ------- --------------
5.0-1.0...---- Al...------ -------------- Si. ----------- Si.
1.0-0.5:..:.::.:.... ....... Si, A1, Fe._
1.00.5 ---- ------- A, Na, Fe -- Al, Fe, Mg, Na, Si, Al, Na, Mg, Al.
0.5-0.1 ------ Mg, Fe, Na, Mg---
Na. Na. Mg.
0.1-0.05-- Sr, Ti --- Ti, V, Sr Sr, Y, La, Cr, V, Y, Fe, Sr, Cr,
0.05-0.01... Pb, Mn, Mn, B, Y,
.-. Cr. Cr. Ti, Zr, B. Fe, Sr. V. Y. Ni.
Cr. Cr. I
0.01-0.005 .. Ba .-. Ba, La, Ni. Mn, V, Ba, Ni, La, Ba, Mn, Ti, Ba,
Ni, Yb. Ti, Zr. La, Cu.
0.005-0.001. Cu, V, Y Zr, Yb, Cu Cr, Pb, Cu, Cu, Mn, YbZr, Mo, Yb.
Sc. Yb, Ag. Ag.
0.001-0.0005.. Zr. --...---
0.0005-0.0001 Yb g Be ..... Ag-.........

1 Bone Velley formation, Bonny Lake mine, Ridgewood, Fla.; analyst, Katherine
V. Hazel.
2 Bone Valley formation, Watson mine, Fort Meade, Fla.; analyst, Katherine V.
3 Bone Valley formation, Varn mine, Fort Meade, Fla.; analyst, Katherine V.
4 Daily production samples, Khouribga, French Morrcco; analyst, Helen W.

are less soluble than hydroxylapatites, and carbonate-
fluorapatite can replace previously formed hydroxyl
varieties in bone and island phosphorites. This last
fact was demonstrated in the classic investigations of
Carnot (1893) and is the basis of the current interest
in water fluoridation for the prevention of caries, as
well as being partly responsible for the dramatic revela-
tion of the Piltdown forgery (Oakley, 1955b; Weiner
and others, 1953).


Many of the petrographic terms common in phos-
phorite literature are ambiguous. The terms used in
the present report are defined here. Most marine phos-
phate deposits contain oviform particles ranging in
size from granules to fine sand (4.0 to 0.06 mm) and
composed almost entirely of carbonate-fluorapatite.
These may display oolitic structures; many of them are
replaced fossils; many show no textures or structures.
Reflecting the variety of possible origins of the deposits,
the names oolites, nodules, pellets, and granules have
been frequently employed for all of the ovules. The
term "oolite" properly denotes only small round par-
ticles evincing accretionary textures; the term "granule"
is primarily a size term; and the term "nodule" is too
nonspecific to be useful here as it usually applies to
round irregular masses of any size and may embrace
rock as well as monomineralic particles. In keeping
with similar usage by V. E. McKelvey and his col-
leagues (oral communication, 1950) the term "pellet"


is used as a general term for oviform sedimentary
apatite particles.
The term "phosphorite" has gained increasing accept-
ance as a rock name synonymous with the cumbersome
term "phosphate rock." It is used here to denote rock
or specimen containing substantial sedimentary apatite.
Thus, one may speak of a "phosphorite sample," a
"phosphorite deposit," or a "phosphorite pebble." The
last would denote a round pebble-size rock fragment,
rich in apatite but containing other minerals such as
quartz, clay, and calcite. In contrast the term "apatite
pebble" denotes a pebble-size, round or rounded, frag-
ment of virtually pure apatite.

The occurrence of uranium in phosphate nodules,
bones, and other apatite materials was first demon-
strated by Strutt in a remarkable set of analyses based
on radioactivity determinations (Strutt, 1906, 1908).
Strutt pointed out that the determinations were not
affected by thorium which is virtually absent in fossil
bone or sedimentary apatite, as is borne out by the
analyses of phosphorite pebbles given in table 4.

TABLE 4.-Thorium contents of Florida pebble phosphorite
[Analyses from U. S. Geological Survey laboratory]

Locality Percent Th Percent U

South Pierce mine, Brewster-...--...-............-------------------........ <0.0005 n. d.
Carmichael mine, Plant City-.............-------.......------ .001 1n. d.
Bonny Lake mine, Mulberry..-....----...........------......----------- .0007 0.017

I Although uranium was not determined on these samples, similar materials from
the same localities contain 0.01-0.02 percent of uranium.

Available analyses of igneous apatite show three or
four times more thorium than uranium, and if rich in
rare earths, igneous apatite may contain as much as
0.38 percent of Th, as reported in apatite from Mine-
ville, N. Y. (McKeown and Klemic, 1956).
Despite the lack of thorium in phosphorite, it is
nevertheless preferable to base a discussion of uranium
on chemical rather than radiometric determinations.
This is particularly true in view of the ease with which
apatite can be postdepositionally leached or enriched
in uranium and thus be out of radioactive equilibrium
regardless of its age. In addition, due to a great varia-
tion in the size of the inherent crystallites of the sedi-
mentary apatites (Jacob and Hill, 1953), it is difficult
to maintain the standardization necessary to establish
comparable equivalent-uranium results among a variety

to materials, regardless of the fact that they may all be
ground to the same apparent size prior to radiometric
assay. It is perhaps for these reasons that the equiva-
lent-uranium figures presented in the excellent sum-
mary of H6bert (1947) are somewhat high. The
discrepancies in H6bert's and in Guntz's data (1952)
have been pointed out by Davidson and Atkin (1953)
in their very thorough review and discussion.

Table 5 presents typical and commonly encountered
values obtained in a variety of apatite materials studied

TABLE 5.-Uranium content of representative apatites
[Asterisk Indicates values considered by the authors to be typical for such material.
Analyses from U. S. Geological Survey Laboratory] Uranum
A. Fluorapatite, Caio(PO4)6Fs, from crystalline rocks (rC
1. Alkali syenite, Renfrew, Ontario, Canada..--..- 0. 0022
2. Gabbro, Henderson, N. Carolina.------------ 0006
3. Shonkinite, Mountain Pass, California---.....----. 0049
4. Magnetite deposit, Durango, Mexico.-------- 0009
5. Magnetite deposit, Mineville, New York -----. 079
6. Syenite, Kola Peninsula, U.S.S.R.---------- <. 001
7. Tonalite, southern California batholith-----..--. 012
8. Granodiorite gneiss, Boulder Creek batho-
lith, Colorado ---------------------- .0069
9. Quartz monzonite, Boulder Creek batholith,
Colorado-----.------------....................-----------. 0053
10. Hornblende biotite granite, Boulder Creek
batholith, Colorado -----.------------- .0078
11. Mica granite, Boulder Creek batholith,
Colorado ---------------------------- .0047
12. Quartz diorite gneiss, Boulder Creek batho-
lith, Colorado ----------------------- .0049
B. Carbonate-fluorapatite (F=> 1 percent), Caio(P04,
COs)6(F,OEH),-s from marine sedimentary phos-
la. Pellets, Hawthorn formation, Florida -------- *. 0061
b. Do-...-------------..........----------------- .0075
c. Do ------------------------------- .0045
2a. Pellets, Bone Valley formation, Florida --.--- *. 009
b. Do..........------------------------------- .012
c. Do.............------------------------------. 019
3a. Pebbles, Bone Valley formation, Florida -----... 011
b. Do---.------....--------------.....-------.... *. 016
c. Do --....------.----.----...--------------.... 024
4a. Pellets and oolites, Phosphoria formation,
Idaho -----------....... --------------- *. 0061
b. Do------------..--........--...--....------------...... .016
5a. Pellets and replaced fossils, Khouribga, French
Morocco -----------.......-------------- .-- 007
b. Do---....-------------------------- *. 013
c. Do.....-------..---.....--..-----------------........ .023
6. Nodules dredged from the sea
floor off southern California Samp No.
69-----. 0089
(Samples, those analyzed for 106 -----. 0068
Dietz. Emery. and Shepard 121 0041


158 -
162 ---

. 0125


TABLE 5.--Continued
C. Fossil Bone, carbonate-fuor-hydroxylapatite (F=<
1 percent) Caeo(POi,COs)a(OH,F)s-_
1. RW-4468, mammalian (sp. unknown), May-
port, Florida---------------------------.......0. 074
2. W-3841, mammalian (sp. unknown), Miles
City, Montana-..--------...............--------------....--. 83
3. Titanothere bone, Hell Creek formation, Camp
Cook, South Dakota ---------------------- .015
4. Shark tooth, Bone Valley formation, Florida ... 009
5. Manatee rib, Bone Valley formation, Florida- 006
6. Conodont fragments, Khouribga, French
Morocco ------------.-------------------. 079
D. Guano-derived phosphorites, carbonate-hydroxyla-
patite Caie(P04,COs)s(O-H)s-
1. Angaur Island, Pacific Ocean ----------------............... 001
2. Mona Island, Puerto Rico- ------------------ 001
3. Cuthbert Lake Rookery, Tampa Bay, Florida <. 001

by the U. S. Geological Survey. In all cases the analyses
are chemical 1 and represent the contents of individual
mineral specimens or pure mineral concentrates rather
than total rock and hence are of additional value in the
geologic literature and do not duplicate the excellent
summaries of uranium in certain phosphorite deposits
(Rusakov, 1953; H6bert, 1947; Davidson and Atkin,
1953; McKelvey and Nelson, 1950; McKelvey and
others, 1955; Thompson, 1953; Cathcart, 1956).
It is apparent from the above data that uranium
content in apatite materials as a group varies from
barely detectable traces to the concentration of a minor
constituent, approaching 1 percent. Nevertheless, within
major occurrences of the same nature, as exemplified by
the igneous rocks of the Boulder Creek batholith, or all
of the immense marine phosphates, the level of concen-
tration is of the same order of magnitude, 0.OOX to
0.03 percent. This range in uranium content may be
taken to embrace the great bulk of apatite in the
lithosphere and departures from it may be safely con-
sidered unusual and examples of special history.

The unusually low uranium contents of island phos-
phorite, as illustrated by the Angaur and Mona Island
analyses (table 5) are attributable to the lack of
uranium in the parent limestones from which they
originate by subaerial replacement. Only rarely do
The uranium analyses reported in this paper that have been per-
formed in the U. S. Geological Survey laboratories have been done by
fluorimetric methods. Grimaldi and others (1954) discuss the accuracy
of these methods pointing out that they may be as good as 4 percent
of the uranium content. They state that under routine conditions,
"The error generally is greater and may range from 8 to 15 percent
of the uranium content. When errors occur, the results are generally
low." Most of the analyses in this paper have been done by a
fluorimetric-extractlon procedure. For this reason, as well as because
of the nonroutine nature of the determinations, it is safe to assume
them to be accurate in the range of 5 percent of the uranium present.

island phosphorites contain more than 0.002 percent
uranium and where they exceed this value the uranium
in all likelihood has been acquired postdepositionally
from percolating ground water or from sea water
with which such deposits are sprayed or washed by
wind and wave. It is of interest in this regard to com-
pare the fluorine and uranium contents of such mate-
rials, as fluorine, not being a primary constituent of
subaerially formed phosphorites, is a certain indicator
of postdepositional replacement.
Table 6 and figure 4 show that the percentage of
uranium bears least relation to the P205 content or,
inferentially, to the amount of apatite present in the
rock. Thus specimens 3, 7a, and 8 are all virtually pure
apatite rock yet their uranium contents differ by several
hundred percent. In contrast, uranium generally in-
creases as fluorine, or as the ratio of fluorine to phos-
phorus does. The lack of a constant ratio between the
increases is easily explained by the fact that two inde-
pendent structural emplacements are involved, F for

TABLE 6.-Partial analyses of cave phosphorite from Mona
Island, Puerto Rico

POl),' F2 U' FU 000
Sample (percent) percentt) (percent) PX00 P; oX10,000

Mona6 --------- 26.3 0.07 0.0003 0.27 0.11
7b -------- 15.8 .06 .0003 .38 .19
3 --------- 41.4 .05 .0005 .12 .12
4 --------- 32.4 .24 .0011 .74 .34
8 ---------38. 6 1.08 .0028 2.80 .73
7a ------.. 37.1 .65 .0014 1.75 .38

I Analyst, A. B. Caemmerer.
SAnalyst, 0. Edgington.
SAnalyst, C. L. Johnson.

o 10 20 30
Fiouna 4.-Relation of uranium to PsO in Mona Island cave


OH, and (presumably) U for Ca. That there are so
few analyses of the required precision available for
such comparisons is in itself a reflection of the gen-
erally recognized lack of uranium in phosphate deposits
of continental origin. Nevertheless, the facts that sec-
ondary uptake of fluorine and uranium are codependent
and related to the same geological process and time of
emplacement seem indicated by figure 5, in which the

10 20 30
rnmuRE 5.-Relation of uranium to fluoridation in Mona Island cave

admittedly few points exhibit a trend that originates
at the point of zero content for both percent of
fluoridation and uranium content.
Similar accord between uranium content and the
percent of fluorine of Cenozoic bones, in which the
fluorine is unquestionably of secondary origin, has been
noted in studies of Oakley (1955a, b).
If we note also that marine apatite nodules of recent
origin contain appreciable uranium (table 5), it seems
evident that the large marine phosphorites derive much,
if not all, of their uranium from the ocean water at
the time of their formation.

Two large problems are posed by the analyses given
in table 5. In what form and manner does uranium
occur in phosphorites? How may its variations within
the same deposits or among different deposits of essen-
tially similar material be explained? Thus, the apatite
nodules from Florida and Morocco are compositionally
and petrographically quite similar, yet those from
Morocco contain 2 to 3 times as much uranium. Isolated
pebbles from the Land Pebble field contain as much
as 0.2 percent of uranium in contrast to the normal

To explain the nature of uranium in phosphates
several mechanisms may be postulated. The uranium
may be present in a separate phase such as U02, UF4,
or Ca(U02)2(PO4)2.8H20 (autunite). It may be
present as isolated U+4 ions or (U02)+2 radicals,
chemically adsorbed on surfaces or internal discon-
tinuities (Hendricks and Hill, 1950). It may be present
by structural substitution of U+4 in apatite as proposed
by several investigators (Michael Fleischer, oral com-
munication, 1949; McKelvey and Nelson, 1950; David-
son and Atkin, 1953; Goldschmidt, 1954) on the
grounds of theoretical plausibility in view of the fact
that the ionic radii of U+4(0.97A) and Ca+2(0.99A)
(Green, 1953) are virtually identical. Lastly, it may
be substituted structurally in apatite as (U02) +2 since
it has been demonstrated that the uranyl radical will
be abstracted from solution by glycol-ashed bone. Such
uranium uptake occurs by displacement of two moles
of calcium per mole of uranium and also renders sur-
face phosphate "* * nonexchangable to the extent of
two moles per mole of uranium." (Neuman and others,
1949a; 1949b). These are puzzling facts for divalent
uranyl. Moore (1954) has shown that phosphate can
also remove appreciable (U02) +2 from cold water
solution and that the extracted uranium cannot be
washed out.

It is a fact of signal importance that uranium min-
erals are virtually unknown in phosphorites. Uranyl
phosphates and vanadates are common in other deposits
and they are readily found by virtue of their habit as
highly colored "paints" and the fact that many of the
species fluoresce brilliantly, thus allowing detection of
pinpoint disseminations. Yet these minerals have been
noted in only three or four local concentrations in
phosphorites (Altschuler and others, 1956; Arambourg
and Orcel, 1951; McKelvey and others, 1955).
In the Bone Valley formation, the mineral autunite
was found in a single layer of leached and altered rock,
2 feet long and 6 inches wide. It occurred in thin square
tablets with perfect (001) and prominent (010)
cleavages. Its indices are No = 1.603 0.002 and
Ne = 1.585 0.002. Flakes give a uniaxial figure and
show blue to violet interference colors when tilted. The
mineral shows a bright emerald-green fluorescence on
exposure to light of 3650 A wave length and fainter
fluorescence of the same color at 2537 A. X-ray data
showed the mineral to be autunite and the optical
properties, although not characteristic of ideal autunite,
are nevertheless attributable to autunite as described
by Fairchild (1929) for synthetic materials.


Within the leached rock in which it occurs, the
autunite coats cavities or is scattered throughout the
porous cement. The mineral is absent from the interior
of solid pebbles. In addition, phosphorite completely
free of autunite (established by lack of fluorescence
after powdering), but from the same rock, contains
appreciable uranium and the autunite was found to
account for only 5 percent of the total uranium in the
rock. Thus, even where their presence is established,
uranium minerals are of secondary origin and too
insignificant to account for all of the uranium present.

In efforts to understand the occurrence and variation
of uranium in phosphorites, as well as to concentrate it
for commercial use, many attempts have been made to
correlate it with physical and chemical properties of
apatite and phosphorite. Materials from the Bone
Valley and Phosphoria formations and from the
Moroccan deposits have been fractionated according
to their variation in specific gravity, magnetism, color,
luster, and particle size. None of these separations led
to any positive or consistent correlation with uranium.
This is not surprising, however, as it is not to be
expected that uranium present in the order of O.OX or
O.OOX percent, or 1 atom per several thousand unit cells
of apatite, could sensibly alter the properties of the
host rock.
The property of fluorescence is a notable exception,
however, and it was found that a consistent relation
could be observed between intensity of fluoresence and
uranium content within certain samples. In table 7,

TABLE 7.-Comparison of fluorescence and uranium content in
composite samples of apatite nodules from the Bone Valley for-
mation, Boyette mine, Florida
[Analyst, Harry Levine]

Fluorescence color Percent U Number of

Yellow --------------------------- 0. 017-0. 021 2
Yellow to brown -------------------. 009 1
Pink or peach------------ .006-. 010 2
Purple or lavender----------------- 003-. 006 4

uranium analyses are given for composite samples of
apatite nodules hand-picked on the basis of their fluo-
rescence under ultraviolet light of 3650 A wavelength.
Each of the composite samples in table 7 consisted
of many pellets or pebble fragments, and the fluores-
; cence was characteristic of the entire pellet, or pebble,

rather than pinpoint or spotty. The correlation of
increased uranium content with enhanced fluorescence
is, therefore, further indication of the absence of sepa-
rate uranium phases in phosphorites, even below the
limits of microscopic detection, as it illustrates a change
in an inherent property of the host mineral, apatite.

In solution studies of rock from the Bone Valley and
Phosphoria formations with nonoxidizing solvents such
as sulfuric, phosphoric, and citric acids, it was found
that the percent of uranium extracted was always pro-
portional to the percent of P205 extracted. This result
was attained whether the samples were slimes or
crushed pebbles, whether calcined or treated raw, and
with highly acid solutions or with small volumes of
dilute weak acid (Igelsrud and others, 1948, 1949).
It seems evident that uranium in phosphate deposits
cannot be accounted for by discrete uranium phases,
by absorption, or by loosely adsorbed ions, radicals, or
phases. This is indicated by three lines of evidence-
the lack of uranium minerals, the fluorescence of
uraniferous apatite, and the inability to concentrate or
separate uranium by physical or chemical means in
the laboratory.


Most investigators have found that uranium varies
approximately as P205 content in phosphate deposits
but that a close correlation does not exist between the
two. The Moroccan deposits illustrate both the general
accord and the lack of detailed agreement. Figure 6 is
a scatter diagram in which P205 is plotted against U
for 18 samples representing the average daily produc-
tion from various beds and locations in the Oulad-

U 15
5 10

z 5

1 5 10 15 20

25 30 3>

FIGUc 6.--Relation of uranium to PsOs in phosphates fr6m Oulad-
Abdoun Basin, French Morocco.




SI I -


Abdoun Basin. They thus represent large tonnages.
The tendency for the two factors to be related is ap-
parent from the plot, in which most of the points fall
between narrow limits, bounding a region of well-
defined slope. However, when only high-grade samples
are studied (those grouping closer together and con-
taining more than 0.011 percent uranium in figure 6)
more scattering is noted.
The approximate dependence of uranium on phos-
phate content is to be expected on two different grounds:
1. Phosphorites are notably devoid of separate uranium
mineral phases.
2. Uranium is intimately associated within apatite as re-
vealed by lack of preferential solution and by fluo-
Furthermore, the other mineral constituents of the
major phosphorites, generally clays, carbonates, and
quartz, almost universally contain less than 0.001 per-
cent of uranium. A few samples of clay contain more,
but such clays have been phosphatized or contain pri-
mary, admixed apatite, and their phosphatic material
can be dissolved with dilute nitric acid leaving a pure
clay residue that is relatively uranium-free. The ratio
of uranium to phosphate dissolved is the same as in the
apatite associated with such clays.
In view of the essentially exclusive association of
uranium with the apatite within unweathered phos-
phorites, almost perfect positive correlation between
uranium and any parameter of apatite content would
be expected within a group of related rocks. However,
detailed studies of the relations among uranium and the
other chemical constituents of apatite show frequent and
striking departures from pronounced positive correla-
tion. The prevailing relation in the Bone Valley forma-
tion is for coarse pebble to be richer in uranium and
poorer in P205 content than the finer pebble. Cathcart
(1956) demonstrated this relation and found that a
moderate negative correlation (-0.64) exists between
P20s and uranium for samples of +150 mesh pellets
and pebbles throughout the Land Pebble Field. Such

TABLE 8.-Size distribution and uranium contents of pelletal
phosphates from Khouribga, French Morocco
[Analyst, Frank Cuttitta]

Diameter Weight Uranium
Sieve size (mesh) mm percent of percent

32 ---------------------------- 0.630 2.0 0.015
44----------------------------- .437 2.3 .015
84 .210 35.2 .016
120---------------------------- 149 28.4 .015
200---------------------------- .074 25. 9 .016
325---------------------------- .044 2.9 .015
<325-------------------------- <. 044 3.4 1.012

i The finest fraction contains clay diluent.

negative correlation has not been found elsewhere and
indeed, even within the land pebble field deposits con-
taining 10 percent P205 will contain less uranium than
deposits with 20 percent P20O. In the Moroccan deposits
uranium is uniformly distributed through all size
grades, as table 8 demonstrates. In the Phosphoria for-
mation McKelvey and Carswell (1956) have demon-
strated that maximum uranium content is regionally
associated with areas of thickset and highest grade
phosphate accumulation.
An equilibrium condition may exist between apatite
and uranium in sea water during the precipitation of
apatite; however, such equilibrium may vary greatly
with time and marine environment. Thus, assemblies
of data from different parts of the same formation may
represent a variety of different equilibria and an aver-
age of such varied groups of data may have the effect
of masking, rather than demonstrating, a universal
relation. The work of Thompson (1953, 1954), who has
explored the relation of uranium to apatite composition
in a series of closely spaced, contiguous samples from
the Phosphoria formation, possibly overcomes this
difficulty. Her results are reproduced in table 9.
The results developed, and noted by Thompson, are
that samples with a higher average uranium content
show the best positive correlation with P205 content.

TABLE 9.-Correlation coefficients among eU, PsO5, F, COs, and organic matter in samples of the Phosphoria formation (Thomp-
son, 1953, 1954)

Number eUX101 percent PsOs percent Correlation coefficients
Locality and sample of
Range Average Range Average eU/POs eU/organic CO,/PiO F/P206

Brazer Canyon WT 605, 604--...----........----..........----------..-.. 51 8-29 13 12-34 26 +0.2 -0.2 +0.52 +0.98
Brazer Canyon WT 603-------------------------- ---.....-. 24 7-36 20 3-34 26 +.8 -.6 +.93 +.99
Coal Canyon WT 700.....----....--------.....----------------..............------..12 6-16 31 1-34 20 +.9 -.7 +.98 +.99
Trail Canyon WT 365-----..........--------------------.....-------- 26 4-28 8 20-35 30 +.1 +.5 +.29 +.95
Reservoir Mountain WT 910------------------------------....... 31 7-30 16 31-36 34 -.2 -.6 n. d. n. d.


The same samples also show the best positive correla-
tion between CO, and P205 and the best negative cor-
relation with organic matter. All samples showed
excellent positive correlation between F and P205.
This last is expectable as the phosphate mineral is a
carbonate-fluorapatite and although free fluorite is
common in the Phosphoria, it was not detected in any
of these samples (Thompson, 1954). One is struck by
the spread in the correlations between eU and P20s,
and by the fact that the samples from Reservoir Moun-
tain, which show negative and very poor correlation,
do not differ greatly in average eU content from samples
WT 603 which show strong positive correlation.
Equally arresting is the fact that the two groups of
samples showing best correlations do so for all of the
constituents compared. This suggests that all of the
factors are covarying with the same set of environ-
mental conditions and therefore that these two sets
of samples have sustained the least postdepositional
change. This follows from the fact that the rocks are
polymineralic, containing organic matter, carbonates,
and apatite as separate phases. As the relations among
these in primary marine deposition would differ con-
siderably from those that would prevail under sub-
aerial weathering or metamorphism, the CO2, eU, and
P205 correlations in sets of samples WT603 and WT700
may all reflect the carbonate-fluorapatite content in
relatively unaltered rock. On the other hand the poor
correlation between CO2 and P205 in the other sets may
reflect secondary alteration in terms of the highly
mobile calcite, and, as during such alteration uranium
may be added or subtracted, and apatite may be
affected, eU-P205 correlations are also notably poorer.

Evidently, comparisons based on composition alone
do not yield consistent explanations of the occurrence
of uranium in phosphorites. Davidson and Atkin
(1953) after surveying the bewildering variation in the
available chemical data, concluded that in most in-
stances the uranium in phosphorites is postdeposi-
tionally emplaced from percolating ground water as a
proxy for calcium. This explanation adequately ac-
counts for the uranium content of isolated bones and
some guano deposits, and may account for some part of
the uranium of other materials, but it is not tenable
when applied to the preponderant occurrences-the
marine phosphorites, plus the phosphatic shales and

The theory of ground-water percolation as the major
source and mode of uranium emplacement demands
that uranium content increase significantly with age.
L6noble, Salvan, and Ziegler (1952) observe that radio-
activity increases with age in the Moroccan deposits.
However, they stress the minute nature of the variation,
and Guntz (1952), in discussing the same deposits,
emphasizes their regularity of grade, which suggests
to him uranium coprecipitated during sedimentation.
It is significant also that the Senonian Moroccan apatite
contains appreciably more uranium than apatite from
the Permian Phosphoria, which is much older, and has
had a more varied history.
In a comparison of the radioactivity of nine Polish
phosphorites, ranging in age from Cambrian to Upper
Cretaceous, Pi6nkowski (1953) noted that "* * the
most ancient phosphorites were feebly radioactive;
whereas the activity increases progressively in the
younger formations," (table 10). Pi6nkowski did not

TABLE 10.-"Inferred uranium content" ("in arbitrary: units") of
Polish phosphorites (Pienkowski, 1953)

"Geologic stage" Age, in 106 Uranium

Middle Cambrian--------------------............-----.................------------... 400 3.7
Middle Cambrian ------------------.-..--.------ ... 370 9.4
Upper Ordovician----------------...................... ..... ..... 350 5.3
Lower Cenomanian ---....--- --- --......... ..--- --- 100 63.0
Lower Cenomanian .......................... .... 100 46.0
Middle Cenomanian- ------------------ ------ --... 95 47.6
Middle enomanian ..... ...... ...... 95 47.8
Maestriehtien ..........................------------------------------------------ 75 138.7
Upper Cretaceous -------------------......... -..... ... 60 132.5

explain this relation and, as his data fall into three
groups rather than continuously over the time span
examined, we believe that a progressive change has
not been significantly demonstrated. It is germane to
the present discussion, however, that large steplike
increases in uranium content occur in the younger
phosphorites. This contradicts the theory that ground
water is the major source of uranium in phosphorites.

Regarding the question of marine versus ground-
water origin, it is most noteworthy that apatite nodules
dredged from the Gulf of California (table 5), contain
as much uranium as the pellets from the Permian Phos-
phoria and Miocene Hawthorn formations. The Phos-
phoria and Hawthorn materials have been exposed
subaerially, whereas the nodules from southern Cali-
fornia waters have remained submerged and, although


they may have undergone submarine reworking, most
of them are thought to be of Quaternary age (Dietz
and others, 1942).
Further information on the magnitude of ocean-
derived uranium in recent deposits may be gleaned
from the analyses of slighty phosphatic materials
dredged from the Gulf of Mexico during recent U. S.
Geological Survey studies (H. R. Gould, written com-
munication). Table 11 presents 10 of many hundreds
of similar analyses. The samples represent contem-
porary accumulations of very slightly phosphatized
shell and marl, and it is most significant that the ratio
of uranium to phosphate in them far exceeds that of
most high-grade marine phosphorites.

TABLE 11.-Phosphate and uranium contents of bottom samples
from the Gulf of Mexico
[Analyses from U. S. Geological Survey Laboratory (Gould, H. R., written
communication, 1953)]

North West
latitude longi-

2650.1' 8316.1'
26052.1' 8333.5'
2650.7' 8346.1'
2650.1' 83057.9'
26051.5' 84008.0'
26052.2' 84032.7'

26o34.8' 8415.0'

26031.3' 8357.9'

26018.7' 8312.5'
26o20.0' 83022.1'

U P20o
(per- (per-
cent) cent)

0.0002 0.13
.0002 .11
.0002 .11
.0002 .09
.0003 .28
.0002 .24

.0002 .20

.0002 .18

.0001 .09
<.0005 .08


Shell sand.
Shell sand.
Algal sand.
Algal sand.
Shell sand.
Foraminiferal sand and
Foraminiferal sand and
Foraminiferal sand and
Shell sand.
Quartz sand.

Table 12 lists analyses and U/P205 ratios of repre-
sentative large deposits and individual samples of
apatite, all of materials not demonstrably enriched
The unusually high U/P205 ratio of the phospha-
tized bottom samples may be due to some special
regional condition such as oceanographic factors affect-
ing CO2 concentration which in turn can influence
uranium solubility (Piggot and Urry, 1941; Bachelet
and others, 1952), or perhaps unusual uranium concen-
trations in the sea water of the Gulf as a result of the
Pleistocene weathering and leaching of the land pebble
phosphate deposits.
Although the mechanism causing the high uranium
concentration in the phosphatized shells and marls can
only be conjectured, the fact remains that enrichment
of such magnitude can occur in a purely marine en-
vironment. Some nodules dredged from coastal Cali-

TABLE 12.-Relative contents of uranium and P205 in marine

Sample U P205 U/P20s
(percent) (percent)

Gulf of Mexico bottom samples I

Range ---------------- 0.0001-0.0003 0.09-0. 28 -----..
Arbitrary mean -------- 0. 0002 0. 185 1/925

Pacific Ocean off southern California 2

162 Phosphate nodules- 0. 0125 22. 43 1/1800
69 Do----------- .0089 29. 56 1/3300
158 Do----------- .0081 29. 09 1/3600
106 Do----------- .0068 29. 19 1/4300
183 Do- .0051 29. 66 1/5800
121 Do- .0041 28. 96 1/7000

Hawthorn formation, Florida 3

Primary apatite pellets- 0. 0075 37 1/4950
Do --------------- .0045 37 1/8200

Khouribga, French Morocco 4

Phosphorite---------- 0. 015 33. 0 1/2200
Do--------------- .014 33. 6 1/2400
Do ------------- .013 33. 0 1/2540
Do --------------- .013 33. 6 1/2580

Land pebble field, Florida s

Clastic apatite compos-
ites----------------- 0.018 30.7 1/1760
Do --------------- .014 33.4 1/2380
Do--------------- .010 34.5 1/3450
Do--------------- .010 35.2 1/3520
Do --- 009 33.0 1/3660

i See table 11.
2 Pi05 and sample numbers from Dietz, Emery, and Shepard (1942); uranium on
the same sample splits by B. S. Clarke, Jr.
3 Uranium by R. S. Clarke, Jr.; PsOs, assumed average for such pure apatite
nodules, will yield conservative U concentration.
4 Daily production averages at four stations in Oulad-Abdoun Basin, French
5 Weighted averages of nodules and pebbles from five areas each representing a
minimum of 100,000 long tons of phosphorite (data converted from Cathcart, fig.
3A, 1956).

fornia waters contain as much uranium relative to
phosphate as the relatively rich large deposits of
Florida and French Morocco.


Neglecting the trivial occurrences in guano and bone,
spectacular as the actual uranium concentration in the












latter may be, it can be stated that most of the uranium
in marine sedimentary phosphorite is of marine origin.
The neutral to basic nature of apatite makes it sus-
ceptible to alteration under the prevailing acidities of
most vadose waters. It is thus to be expected that the
primary marine uranium may be redistributed by vari-
ous secondary processes, as indicated by conflicting and
varied relations of uranium to composition in the same
and among different deposits, and by the evidence of
laboratory uptake of uranium by bone and phos-
phorite. Therefore, it appears that the uranium con-
tent of apatite can be explained only in terms of the
total geologic history of the deposits. In the following
sections of this report a number of specific deposits are
examined, and the geologic processes that have gov-
erned their uranium content are elucidated in hopes of
developing a set of principles to explain the range and
variation of uranium in most phosphorites.

The land pebble phosphates of Florida present an
opportunity to study apatite in a variety of geologic
circumstances and thus to evaluate its uranium content
in terms of both petrography and petrology. These
deposits are mainly in the Bone Valley formation of
Pliocene age (Cathcart and others, 1953) and subordi-
nately in the underlying Hawthorn formation of early
and middle Miocene age (MacNeil, 1947). The surface
of the Hawthorn is irregular, its rock is solution-pitted,
and it contains many small and large slumps, within
which the overlying Bone Valley thickens (Cathcart,
1950). The Hawthorn was exposed and weathered dur-
ing late Miocene time, developing an irregular topog-
raphy and accumulations of phosphatic residue con-
taining primary apatite nodules, inherited as such from
the Hawthorn, and secondarily phosphatized limestone
pebbles. During a period of Pliocene marine trans-
gression this residuum of the Hawthorn was reworked
into the unconformably overlying Bone Valley forma-
tion, with additions of quartz, clay, and probably
phosphate (Altschuler and others, 1956).
The Bone Valley formation is about 30 feet thick
and consists of two units (fig. 7). A pebbly and clayey
sand characterized by graded bedding comprises the
lower two-thirds and is the unit mined. Grading
upward from this is a massive-bedded, less phosphatic

quartz sand, which together with a surface mantle of
quartz sands, is discarded as overburden during mining
The upper part of the Bone Valley formation has
been leached, altered to aluminum phosphates, and sec-
ondarily enriched in uranium in a widespread, trans-
gressive zone that, though discontinuous, underlies
several hundred square miles in the land pebble district.
This aluminum phosphate zone is the result of lateritic
weathering and ground water alteration, and sections
through it exhibit a progressive variation in mineral
content and texture (fig. 7). In typical sections pebbles
of carbonate-fluorapatite still occur at the base, incipi-
ently leached and altered. The calcium-aluminum
phosphates crandallite and millisite are found in the
middle of the zone. At the top, the aluminum phos-
phate wavellite predominates (Altschuler and others,
In the land pebble deposits we may thus study and
contrast the uranium contents of sedimentary apatite
of the following types:
1. Simple structureless nodules from the Hawthorn formation.
2. Nodules from Hawthorn that have been reworked into
the Bone Valley and are thus in a second cycle of
3. Simple pebbles of phosphatized limestone from the Bone
Valley formation.
4. Complex reworked pebbles from the Bone Valley formation
illustrative of several subeycles of sedimentation.
5. Leached and altered pebbles.
6. Secondarily enriched pebbles.

In a conglomerate rock composed of pebbles resting
in a matrix of finer grained material, the pebbles
usually cause zones of increased permeability in adjoin-
ing matrix by disturbing the packing arrangement of
the smaller grains. These areas would be zones of better
circulation and as such would be the preferred loci of
any leaching or enrichment that occurs within the rock.
Thus the uranium, or the secondary phosphate, might
be distributed according to the texture within the
phosphate pebble rock. Graton and Fraser (1935)
described this type of control and noted its influence
in gold and copper deposition.
To test the influence of texture on the distribution of
uranium, six pebbles were selected and analyzed for
percent of uranium in zones from the surface to the
center. The pebbles were selected from the aluminum
phosphate zone of the Bone Valley formation and the
data given in table 13 are described according to the
sampling scheme shown in figure 8.


.. . . . . ... . . . . . .. .. Sand of

. . . . . . . . .. Pleistocene(?)
". . .' . ." ." . . ges

o -- ---, .- a. .

.. * / < / ,...

*J . ---* *- oo /,,*
_--- -- -- - \ >,,. 0 ,0 "0 : 3o8

'- - ---------------*
S* *. * * 0.**. "oc-000 00 O0 00 0X


S . . .. . .. . . . . .. . . .. . . . . .-.- -
|, ..;.o .;.- . .A : ; g

.*** ** ** *. ** * *d* e * **** ***

*. *. *.*.. *...e o ---.- -
9* P *,* *---* * E. -- *. e I

*- d I |. do db
0 I-- I \0 01 0M .0 o b

Phosphatized Leached and vesi-
clay cular rock
SI Scale. in feet
Limestone Dolomite
FIGURa 7.-Stratigraphie relations in the Land Pebble phosphate field, Florida.


B Outside zone

<- C Median zone

D Central zone

Fxouna 8.--Sampling scheme used for study of uranium distribution In pebbles.

Table 13 reveals a striking increase in uranium con-
tent from the surface inward, in all of the pebbles.
Except for zone B in pebble 1, this increase is constant
and of steep, although nonuniform gradient, and values
at the center are easily 100 percent higher than those
at the outside.
This graded distribution suggests leaching of the
uranium from phosphate pebbles of originally more
uniform richness. It is significant that the sharpest
increase in uranium content is in every case, between
zones A and B. Inside the B zone the increase is
smaller. This is precisely what would be expected as
the surface zones would be more easily attacked.
It is evident that subaerial leaching can effect large
changes in the primary uranium contents of marine
deposits. The leaching illustrated above occurred after
deposition, in the course of ground-water alteration.
Leaching can presumably be equally effective during
subaerial transport and reworking of similar materials.
In this latter case the leaching would not be confined
primarily to conglomeratic material and would prob-
ably show greater effects in the finer materials, as these
have a greater ratio of surface to volume. That leach-
ing did affect the rock's uranium content could prob-
ably be demonstrated by the same means or by the use
of autoradiography.

The Bone Valley formation yields instructive exam-
ples of enrichment as well as of leaching. Unaltered
phosphorite in the lower part of the Bone Valley con-
tains an average o pg f .uramum and 10 to
15 percent of P2Os. In contrast, typical rkfom the
aluminum phospHaeione has 0i.012 percent of uraniumii
and 8 to 12 percent of PAOs-roughly a two-fold enrich-
ment in the piosphaticfrc _.
There is much variation in the composition and
uranium content of the aluminum phosphate zone.
Nevertheless, the distribution of uranium within it
closely follows the distribution of calcium (fig. 9). The
uranium content is greatest in the base of the zone,
declines gradually toward the middle of the zone, and
is relativeIy6Tow-in tEjipperihalf. The zone of basal
enrichment corresponds to the zone containing apatite
pebbles, and analyses for uranium in pebbles, cements,
and mineral concentrates within the entire zone show
that the apatite pebbles are the richest uranium com-
ponents of the zone and that the pebbles in the base of
the zone are richer than those in the middle. Figure 10
illustrates these relations and shows the contribution of


TABLE 13.-Chemical analyses showing distribution of uranium
within phosphate pebbles
Corrections were made for the percentage of quartz in each portion tested. Analyses
from U. S. Geological Survey laboratory]
Pebble Zone Percent U Remarks

1 A 0. 011 Selected at random; a small flattened
B 023 spheroid 2.5 by 2 by 1 cm. The pebble
C 020 is a microconglomerate, made up of
D .021 precipitated phosphate, phosphate pel-
lets, quartz grains, and still smaller
pebbles of similar composition. The
other pebbles are similar in character.
2 A 007 Selected at random.
B .026
C .028
D .033
3 A 020 Selected because of high fluorescence, a
C .028 small fractured pebble of originally high
roundness; original central portion is
now exposed surface, hence only surface
and median zone were sampled.
4 A .011 Selected because of high fluorescence.
C .032
5 A 012 Selected because of high fluorescence; sur-
B 019 face encrustation of brown phosphate
D 026 which was obviously deposited by circu-
lating solutions since some of it occurs
as a circular ridge outlining the area of
previous contact with another pebble.
6 A 007 Selected because of high fluorescence.
B .013
C .015

the various textural components to uranium content of
the total rock in a section through the aluminum
phosphate zone.
Additional evidence of secondary enrichment may be
obtained by comparing the uranium contents of com-
posite samples of pebbles from rock of originally iden-
tical petrography above and below the base of the
aluminum phosphate zone. Pebbles from within the
zone were found to contain 0.056 percent of uranium
in contrast to their anologues just outside the zone


FIGURBE A, B. Phosphatic quartz sandstone from the aluminum phosphate zone, Florida. A, Plane polarized light; B, crossed
nicols. Original clay cement (a) is now highly porous and almost completely replaced by lathlike wavelite.
The large and smoothly bounded cavity (b) is the site of a previous apatite pebble that has been removed by
leaching. The cavity is now rimmed and partially filled by layered crandallite, with admixed kaolinite and
goethite. The concavely stratified bottom filling of the cavity indicates the downward direction of the original
weathering solutions.
C, D. Photograph and autoradiograph of fossil bone, both X1%.
E, F. Autoradiographs of reworked and first cycle pebbles.


which had only 0.023 percent of uranium. Further-1
more, individual pebbles from the base of the zone'
contain as much as 0.25 percent of uranium. Concen-
trates of crandallite and millisite from the middle of
the aluminum phosphate zone generally contain from
0.03 to 0.05 percent of uranium. In contrast, pure!
wavellite, the dominant mineral in the upper part of
the zone, contains only 0.002 to 0.004 percent of
uranium. Despite- the low uranium content in wavel-
lite the upper part of the aluminum phosphate zone
generally contains from 0.005 to 0.01 percent of uranium
(fig. 9). This fact has led some authors to assume that!
"* * radioactivity is as characteristic of wavellite
* * as it is of apatite * *" (Davidson and Atkin,
p. 13). However, most analyses of the wavellitic part;
of the aluminum phosphate zone reveal the presence of
trace to minor amounts of calcium (fig. 9), reflecting,
the presence of small amounts of crandallite that can
be seen microscopically, and in which much of the'
uranium resides.
The aluminum phosphate zone is commonly under-,
lain ferruginous hardpans that are relict of old
ground-water levels. It is also characterized b -y sec-
ondary leaching and vesicularity and by mineral
changes that progress vertically through the section.
At the bottom, the zone is characterized by softened,
and bleached apatite pebbles. In the middle, holes
represent former pebbles, and crandallite has replaced
clay cement and has been deposited from solution
within pebble cavities. Upward, wavellite needles re-
place the clayey crandallite deposits and the original
clay cement (pl. 11, figs. A, B). This alteration has
caused a depletion of CaO, silicate, and some phosphate
in the upper part, where there is a complementary
increase in A1203 (fig. 9) combined in acid alumi-
num phosphate, wavellite (Als(OH)3(PO4)2.5H20).
Wavellite yields downward to intermediate Ca-Al
phosphates (crandallite and millisite), and finally to
the basic apatite at the base of the zone. These features
imply a history of lateritic alteration by acid ground


qLL.. "y


-. ... "
A.-Details of contact between aluminum phosphate zone (white) and overlying ferruginous -mand mantle, Clear Springs Mine,
Bartow, Fla. Photo illustrates the rubbly and concretionary nature of the weathered zone and the irregular and deeply
embayed contact, which, with the residual concretions visible in the overlying sands, reveals the previously higher level
of contact.

B.-Mined out area of Tennessee brown-rock phosphates, Akin mine, Columbia, Tenn., showing control of rectangular joint system
in the origin and distribution of the residual phosphates. Pinnacles and blocks of unaltered, crossbedded limestone of the
Bigby formation occur in rows flanking the valleys created by weathering and mining along thie master joints. The sub-
sidiary joint set may be seen dividing and separating individual blocks. Phosphate has been mined from all of the depressions.



Photomicrographs of arkosic sandstone with carbonate-fluorapatite cement from the Gas Hills area of Wyoming. A, Thin section
alone, plane polarized light, showing quartz grains (a) and intergranular cement (b). B, Photomicrograph focused in the
plane of the superimposed nuclear emulsion to show alpha tracks emanating from the apatite cement.



I II I I i I
Ho20 Ho20A Ho21 Ho22 Ho23 Ho24 Ho25


P2 05s

u 30------------J^ V---------- /. ..- "-""--------
,i // "

aL A A1203

w 20


-- .

2 4 6 8 10 12

FIGURE 9.-Distribution of oxides through aluminum phosphate zone at Homeland mine, Homeland, Fla.

I 1 I I I I
Ho20 Ho20A Ho2l Ho22 Ho23 Ho24 Ho25




o4*/\ \


.02 / '

.01 _

2 4 6 8 10 12

FIGURE 10.-Uranium distribution in various textural components at Homeland mine, Homeland, Fla.


water. The pattern of uranium distribution parallels
that of downward leaching and alteration. Uranium
is liberated during the leaching and solution of apatite
pebbles in the upper part of the zone. It is kept soluble
in the acid ground water and emplaced in the highly
porous, partly leached apatite pebbles at the base of the
zone where the acids would be neutralized by the yet
unaltered calcium phosphate. The correspondence be-
tween the curves for CaO and uranium (fig. 9) and the
basal enrichment of uranium in apatite (fig. 10) sug-
gest that the uranium substitutes for calcium in the
apatite structure. This presupposes that the uranium
in apatite is tetravalent, a supposition which is dis-
cussed on page 69.
The theory of supergene enrichment presented above
demands that originally there was 2 or 3 times more
apatite than now occurs in the aluminum phosphate
zone. This prerequisite is not supported by the obvious
field relations as, in unaltered exposures, it is possible
to see a primary stratigraphic change to a less phos-
phatic clayey sand, from lower to upper Bone Valley
(fig. 7). There is evidence, however, that an originally
thicker section was available to provide the present
enrichment (pl. 12, fig. A). Briefly, the contact between
the upper part of the Bone Valley and the surface
mantle of loose quartz sand is irregular in detail and
the sands of both zones are virtually identical in size,
sorting, and heavy-mineral content. Therefore, a large
part of the loose sand mantle is a residue of weathered
Bone Valley and the present contact is a lowered one.
Unusual and widespread subaerial enrichment of
uranium in apatite also occurs in the phosphate deposits
of the Cooper marl in the vicinity of Charleston, S. C.
Being composed principally of limestone, the Cooper
marl is only slightly phosphatic. However, it has been
postdepositionally weathered and phosphatized, and in
the exposures in the vicinity and along the banks of
the Ashley River, between Drayton and Middleton
Gardens, the deposits are characterized by a thin upper-
most zone of nodular induration. The induration re-
sults in irregular, discontinuous encrustation and the
rock is replaced by secondary phosphate. It is obvious
that the process was accompanied by moderate leach-
ing-the indurated fragments that are loose or easily
broken out of this zone are marked by cavernous irregu-
larity and conspicuous molds of pebble and fossils,
although its fine texture is similar to that of the
unweathered underlying rock.
The unaltered Cooper marl "* * commonly con-
tains about 75 percent lime carbonate and about 2 per-
cent lime phosphate," (Rogers 1913, p. 187). The

Cooper marl was exposed and eroded during Oligoceuq
to Miocene time and later exposed and weathered dur.
ing Pliocene to Pleistocene time (Rogers, 1913, and
Malde, in press). During the last period of weathering
the calcite was both leached out of the section and
replaced by apatite. The fact that apatite could repre
cipitate as a stable phase during weathering suggest
that the ground water alteration was only mildly acidj
This is further indicated by the absence of alumini"
phosphates or even the intermediate calcium-aluminmu
phosphates, despite the fact that the region contain
appreciable clay.
An immediate consequence of the re-forming 01
apatite at the site of alteration is the refixation of an]
uranium mobilized by prior solution. This established
a pattern of residual enrichment. The pattern may S
construed as follows:
1. Mildly acid ground water dissolves calcite and apatite.
2. Some, if not much, of the phosphoric acid in solution I
immediately neutralized by reaction with the remain
calcite of the marl, producing secondary apatite.
3. Most of the uranium in solution is taken up by the newly
formed apatite.
This scheme indicates a disproportionate enrichme
of uranium as a result of its preferential uptake in th
apatite structure. It has been possible to develop cor
elusive data in support of this in the phosphates of th
Cooper marl. Malde in his study of the Cooper ma
noted that "* * the uranium content is rather
form, ranging between 0.001 and 0.005 percent, rarely
as much as 0.007 percent and averaging about 0.003.
In contrast to these observations of the prevail
tenor in the unaltered marl, analyses in table 14 ill
trate the overall magnitude of enrichment in the har
pan collected in the area between Rantowles Creek an

TABLE 14.-Ex-amples of phosphate- and uranium-enrich#
hardpan in Cooper marl near Charleston, 8. C.

Locality Percent U PercaI

Magnolia Gardens, exposure on west bank of
Ashley River---------------------------................. 0.048 221
Andrews, borrow pit on west side State High-
way 161, one-half mile north of Andrews... 061 28.&
Ashley River banks, west bank at road from
Olive Branch Church....-------------------- .048 24.
Ashley River banks, east bank 1.25 miles south
of Lambs-----------.................-------------------- .038 25.1
Bulow Mines, 2 miles southwest of Fort Bull
on Fort Bull-Red Top Highway ---------.. .043 27.1
Micah Jenkins Nursery, U. S. Highway 61,
one-half a mile west of St. Peters Church --- 087 24.

I Analyst, G.7 Daniels.
3 Analyst, A. B. Schrenk.


Ashley River. The analyses show enrichment of both
apatite and uranium, especially of uranium. The selec-
tive enrichment of uranium is shown by comparing
analyses of unaltered marl and the overlying fragments
of cavernous hardpan from the same exposure (table
15). It is significant that the enrichment in P205 is
only 6-fold whereas that in uranium is 15-fold.

TABLE 15.-Comparison of phosphatized and primary Cooper marl
from Lambs, S. C. (300 yards above old railroad terminus on west
bank of Ashley River)

Material Percent U 1 Percent

(a) Unaltered marl ------------------------- 0.005 4. 0
(b) Phosphatized hardpan ----------------- 074 25. 9
(c) Enrichment factor (b/a) ---------------- 15 6

1 Analyst, G. 3. Daniels.
2 Analyst, A. R. Schrenk.

Another illustration of preferential uptake of
uranium during weathering of apatite is obtained by
comparing different nodules from the same hardpan
samples of phosphatized Cooper marl. At several
localities contiguous fragments differed markedly in
color, from dark brown to light tan or gray. The
lighter color was ascribed to the bleaching effects of
oxidative subaerial weathering. On the assumption
that the lighter fragments had thus undergone more
of such weathering, light and dark fragments were
analyzed separately with the thought that differences
in uranium content would be shown and that differ-
ences in other constituents would reveal the relation of
the uranium content to weathering history. Results
of chemical and of semiquantitative spectrographic
analysis of the same samples are given in tables 16
and 17.
The two samples, obtained from localities about 4
miles apart, are representative of hardpan and weather-

ing conditions over a wide area; nevertheless, the chem-
ical changes illustrated by the two comparisons are
remarkably uniform for secondary subaerial processes.
The spectrographic analyses show that the differences
between the contiguous dark- and light-colored mem-
bers of each pair are due to weathering. In each pair
the relative quantities of silica and aluminum increase
and the iron content drops in the light member. The
changes are not large and are of comparable magnitude
in each pair. The fact that the carbon content is the
same for both members of one pair and is even greater
in the light-colored member of the second pair in table
16, proves that the color change was not brought about
by oxidation of organic matter as originally supposed.
Instead, it is obviously the result of leaching of ferric
iron. The persistence of ferrous iron in each case indi-
cates that the leaching was accomplished by reducing
solutions, a fact further indicated by the persistence or
increase in organic carbon.
It is suggested that mildly acid humate solutions,
derived from the organic-rich soils that are common
in the Ashley River flood plain, removed the iron by
the formation of organo-ferric complexes and thus
effected a partition between ferrous and ferric iron.
The solutions dissolved out apatite, slight amounts of
the other soluble constituents, and calcite. This is dem-
onstrated by the uniform decrease in P205 content in
both pairs. This caused a slight residual enrichment
in the relatively insoluble constituents, quartz and clay;
hence silica and aluminum both increase in the light
members (table 17).2 Thus, the light members of each
pair reflect substantially more contact with ground
water solutions, and during such contact appreciable
quantities of uranium were taken out of solution by the

It Is possible that in the Ninemile sample only the relative propor-
tions of SI and F have been reversed. However, in view of the system-
atic changes in Fe and Al being demonstrated for both samples, It is
believed that the Si in the light member of the Ninemile sample
actually has Increased as In the unambiguous case of Lambs.

TABLE 16.-Variation of enrichment with weathering in phosphatized Cooper marl, South Carolina
Location Sample
U' P30s 2 Total FeO Fe2sO 3 Organic
Fe carbon

Ninemile Station, Southern Railroad, half a mile north- Dark fragment ------- 0. 035 27. 0 1. 04 0. 40 1. 04 0.5
west of Goodrich, stream bank, east of railroad trestle. Light fragment -------- 11 26. 9 .57 .38 .40 5
Roadcut one mile south of Lambs on Charleston road-..... Dark fragment -------- 064 27. 2 .73 .34 .66 6
Light fragment -------- 12 25. 5 .35 .38 .07 1. 1

1 The uranium analyses were made on the filtrate of a nitric acid leach. Additional
analyses on the total sample indicate that all the uranium occurs in the soluble,
Phosphatic fraction of the sample. Analyst, G. J. Daniels.
I Analyst, A. R. Schrenk.

SAnalyst, Alexander Sherwood. (Total iron and FeO determined chemically;
FesOs computed by difference.)
4 Analyst, A. B. Caemmerer.


TABLE 17.-Semiquantitative spectrographic analyses of phosphatic
hardpan from the Cooper marl

[Order of precedence gives relative percent content; thus in the dark member at Nine
mile, F is greater than Si, and Mg is greater than Sr. Analyst, Katherine E.

Ninemile, South Carolina Lambs, South Carolina
Range, in percent
Dark (less Light (more Dark (less Light (more
weathered) weathered) weathered) weathered)

>10 ------------ Ca, P Ca, P Ca, P Ca, P
5-10.......-- ..----- --------- ------ ---------------- Si.
1-5...............------ F, Si, Na Si, F, Al, Na Si, F, Na A], F, Na
0.5-1............----- Al, Fe ........-----....... A, Fe ---- --
0.1-.5...--.........--- Mg, Sr, K Fe, Mg, K, Mg, K, Sr Mg, Fe, K,
Sr, U Sr, U.
o.05-.1 ..........---- Ti Ti Ti Ti
0.01-05------.........--. B, Ba, V, Y B, Li, Ba, V, Li, B, Ba, V, B, V, Ba, Y,
Mn, Y. Y. Li.
0.005-01.-------- LI, Mn, Ni Ni Ni -------
0.001-005 ----. Zr, Cr, Yb, Cr, La, Zr, Zr, Cr, Mn, Ni, Sc, Zr, Cr,
La, Sc. Sc, Yb. La, Yb, So. Mn, La, Yb.
0.0005-.001..........---- -..........------ ---------------- ---------------------------
0.0001-.0005 ------ ---------- ------------- -- ---------------- ---------------
o.00005-.0001..... Be Be Be Be
0.00001-.00005....----..----.........------- ------- -----------------

apatite. The fact that the bulk chemical changes are
small (for example P205 data and the date table 17)
proves that residual enrichment cannot explain the 2-
to 3-fold uranium enhancement of the light over the
dark nodules. Patently, the residually enriched u rani um
of the nodules throughout the region is frequently aug-
mented by significant additions of uranium from
ground water.


In comparing uranium distributions within the Bone
Valley formation and the Cooper marl, it becomes evi-
dent that two largely different patterns of regional
enrichment prevail during weathering, despite the fact
that the primary mechanism of enrichment is the same,
namely, the takeup of uranium from ground water
during weathering. The patterns of enrichment are
governed by the mineral assemblages which in turn
reflect the diverse conditions and environments of
weathering. Figure 11 summarizes and contrasts the

Strong acid weathering
(Florida, Senegal, Nigeria, and soils)

Lowered contacts

Unaltered Altered

o *Uranium
__ Altered enrichment

-- o Al phosphates


0o 0 Ca, Al phosphates
..0 0 0 0 0 0 0

a o o "o o oooo oo ooo o a o Uranium
0 0 o0 0 o 0 o oo o o 0 o 0 enrichment
o o 0 o o o 0 ? 0
0 0 0 0 40 C ,

Mild acid weathering
(South Carolina)

Altered "Unaltered

_ -S" --. 0.-- . .

0- o
ao o o


-r -- a




Florida, Senegal, Nigeria
Nature of new phases........... Acid; Al and Ca, Al phosphates
Process................----------.... Phosphatization of clay
Type of enrichment............. Supergene
Locus of uranium enrich-.......At base of weathered section,
ment and hardpans controlled by groundwater



South Carolina
Basic to neutral; secondary apatite.
Phosphatization of limestone.
Near surface of active weathering.

FIGURE 11.-Contrasting modes of uranium enrichment and weathering In phosphorites.




In South Carolina the process is a simple one of
simultaneous enrichment of uranium and production of
apatite at the zone of weathering. The weathering does
not seem to have been very intense-the zone it has
affected is quite thin and the mineral transformations
are not profound, involving merely the leaching of
calcite and apatite and the reprecipitation of apatite.
As a result the zone of enrichment, coinciding with the
zone of appreciable weathering, is restricted to the
uppermost part of the section and migrates only as
weathering reduces the rock to inert constituents
(fig. 11).
In Florida, and in Senegal (Capdecomme, 1953) and
Nigeria (Russ, 1924), the prominence of aluminous
phases implies a regimen of highly acid, lateritic
weathering such as would be expected in these tropical
areas. The development of purely aluminous phosphate
(wavellite) in the upper parts of the weathering zone
in turn excludes the concentration of uranium, which
migrates to where calcic phosphates occur (fig. 11),
and, accordingly, uranium content increases progres-
sively down the section, reaching moderate concentra-
tion in the crandallite zone and causing a supergene
enrichment in the base of the entire zone where
weathering is incipient and apatite still abounds.
The deduction of two different intensities of weather-
ing, strongly and weakly acid, from rocks of two dif-
ferent gross petrographies, might be questioned on the
grounds that the South Carolina deposit, containing
primary calcite, would have so much greater neutral-
izing capacity than the Florida deposit that acid phases
could not be generated. However, it is noteworthy that
the Sengal deposits contained limestone and dolomite
but were nevertheless altered to crandallite and other
aluminum phosphates. In addition it has been shown
that the excessive enrichment of some of the Cooper
marl nodules must represent ground-water enhance-
ment beyond the pure residual enrichment of adjoining
nodules. This ground-water enhancement implies con-
siderable additional ground water seepage; yet the
apatite, a basic to neutral compound, has remained as a
stable phase despite the fact that clays abound in the
overlying marls and metasomatic alteration of apatite
to aluminum phosphates is rather common under
similar conditions in soils. Presumably the acidity was
quite low, as even traces of aluminum phosphates or
intermediate phases are absent in the hardpan of the
Cooper marl.

The brown-rock phosphates of central Tennessee are
a prominent example of residual phosphate accumula-

tion. They were therefore studied for residual concen-
tration of uranium. The brown-rock deposits occur
mainly as a residual blanket surmounting the Bigby
formation of middle Ordovician age (Hayes and
Ulrich, 1903), a crossbedded limestone in which apatite
occurs sparsely concentrated within topset and forest
intercalations, and also weakly disseminated through-
out the limestone. The deposits ".. . occur only in areas
where Bigby-Cannon limestone directly underlies the
soil..." (Alexander, 1953) and are absent when mining
extends into hillsides where fresh limestone is encoun-
tered. It is evident that the deposits were formed dur-
ing Cenozoic weathering by preferential solution of
the limestone (pl. 12, B). In many areas the residuum
occurs as filling of solution-enlarged joints locally
known as "cutters." Within such "cutters" it is possible
to sample both the weathering concentrate and the
adjoining, closely related parent material.8 Analyses
for one such set of samples from the Akin mine of the
Tennessee Valley Authority at Columbia, Tennessee
are presented in table 18. The relation of these samples
to the stratigraphy is shown in figure 12.

TABLE 18.-P3Os and uranium content of weathered and fresh
Tennessee phosphates, Akin mine, Columbia, Tennessee
[Analyst, R. S. Clarke, Jr.]
P2sO U

(a) Limestone of the Bigby formation --.. 1. 60 0. 00004
1. 60
(b) Brown-rock phosphate.------------- 27. 0
27. 4 .00074
(c) Enrichment factor (b/a) ----------- 16 18

0 ~- .



0 Sample location
FIGURE 12.-Sample locations (D for comparison of brown-rock
phosphate with parent limestone in Tennessee.

Exact stratigraphic equivalence of these samples cannot be assumed
due to draping and compaction within the "cutters."


The figures above demonstrate purely residual enrich-
ment. The differences between the concentration factors
of 16 and 18 are negligible in view of the difficulties in
analyzing such small quantities of uranium. The strik-
ing contrast of the above to results from the South
Carolina samples (table 15) where the enrichment
factor for uranium is far greater than for P205, serves
to emphasize the partly external origin of the ground
water enhancement of uranium in the South Carolina

Determinations of the radioactivity and uranium
content of bones have been made by several authors,
and it can be seen from their publications, and from
the analyses in table 5, that bones exhibit the gamut
of uranium content in apatite. The highest radio-
activity for any apatite material previously reported is
by Davidson and Atkin (1953), who listed a specimen
of fossil fish from the Old Red sandstone containing
0.55 percent eU. Bowie and Atkin (1956) have shown
that the unusual radioactivity of other fossils from the
same region and formation is due in appreciable
measure to thorium. They show, however, that the
thorium does not occur in the bone apatite but rather
in hydrocarbon concentrations within the bone cavities.
Although the eU values of fossil bones from the Old
Red sandstones may thus not accurately reflect uranium
in the bone apatite, fossil bones are nevertheless, among
the most uraniferous apatites. The value of 0.83 percent
uranium (table 5) is the highest uranium content of
which we are aware for any apatite substance (bones
or mineral specimens).
Lacking the economic and geologic interest of the
phosphorites, bones have not been extensively studied
for uranium. However, certain well-established gen-
eralizations may be drawn from available information.
Analyses presented by H6bert (1947), Oakley (1955a,
b), and Jaffe and Sherwood (1951), all show fresh bone
of marine or terrestrial organisms to be essentially non-
uraniferous. In addition, comparisons of radioactivity
of Cenozoic human and animal bones (Oakley, 1955a,
b), reveal that uranium content increases with time, an
indication of postburial uptake. As in the guano-
derived phosphorites, uranium uptake in bones parallels
fluorine uptake; apatite of all types is a favorable
receptor for both of these elements.
The above facts, and the experimental extraction of
uranium from solution by apatite (Moore, 1954) and
by glycol-ashed modern bone (Neuman and others,

AA procedure for removing tissue and organic matter.

1949), clearly demonstrate that the uranium of fossil
bone is due entirely to secondary enrichment from
ground water or marine sources. As Davidson and
Atkin (1953) have pointed out, the most uraniferous
fossil bones come from continental gravels of high
permeability. In the highly quartzose and felspathic
terrestrial deposits, a fossil bone may be the most, and
possibly the only, favorable host for uranium (and
fluorine). In such circumstances, ground water charged
by uranium from regional sources may effect unusual
concentrations of uranium in isolated bones or phos-
phorites nodules and cause an enrichment far greater
than that of the enclosing rocks. Thus, Smith and
Bradley (1952) report dinosaur bones from the
Cloverly formation which contain from 0.085 to 0.135
percent of uranium, whereas the enclosing rocks contain
only 0.001 to 0.005 percent of uranium.
Denson and Gill (1956) have shown that the highly
uraniferous lignites of the Fort Union and Hells Creek
formations in northwestern South Dakota and eastern
Montana have derived their uranium from percolating
ground waters which were (and still are) enriched by
leaching of the overlying tuffs. Undoubtedly, the fossil
bones from these formations have become uraniferous
through the same agency. The titanothere bone from
the Hells Creek formation (table 5) exemplifies this
type of enrichment. Probably Miles City fossil
W-3841 (table 5), from an unknown surface exposure
in the same region, derived its unusual uranium content
in the same manner.
Petrographic evidence of the mode of such post-
depositional enrichment is shown by an autoradiograph
of specimen W-3841 (pl. 11, fig. D). The most uranif-
erous (the lightest) parts of the bone are the peripheral
zones and the cavernous and highly permeable internal
structures. In contrast to these areas, the intermediate
dense areas are of relatively low grade. Obviously,
uranium concentration bears no relation to the great
difference in apatite concentration. In addition, it was
determined by X-ray powder diffraction analysis of
different portions of the specimen that the apatite is
of the same type throughout, and further analyses on
separated material showed that the carbon-rich fraction
has less uranium than the mineral fraction. Evidently
accessibility to solutions was the main control on the
localization of the uranium.
Unlike the contrast between bones and enclosing rock
in terrestrial deposits bones and other phosphatic fos-
sils in unaltered marine phosphorites are generally
similar in uranium content to the apatite pebbles and
pellets from the same deposits. This supports the argu-
ment that the unusual postdepositional enrichment of
continental fossils is attributable to special provenance


rather than special mechanism. Analyses of 12 fossil
manatee rib fragments and shark teeth from mines in
the Bone Valley formation showed a range in uranium
content of 0.004 to 0.016 percent (Jaffe and Sherwood,
1951). The P205 contents of the same specimens ranged
from 36.1 to 37.5 percent, indicating that all were rela-
tively pure carbonate-fluorapatite. The manatee and
shark specimens represented species that ranged from
Miocene through Pliocene and many of the Bone Valley
fossils have been reworked from the Hawthorn forma-
tion. Nonbiogenic apatite of virtually identical com-
position and geologic history has essentially the same
range of uranium content. The simple sand- and
granule-size apatite pellets from the Hawthorn forma-
tion (primary) and the Bone Valley formation (re-
worked and possibly primary) are pure carbonate-
fluorapatite containing 36 to 38 percent P205. Their
uranium content (table 5) ranges from 0.004 to 0.012

It has been shown that postdepositional alteration can
cause large variation in the uranium contents of apatite
deposits and that many of these variations can be
ascribed directly to ground water leaching or enrich-
ment. This does not, however, explain the apparently
primary variation in unaltered materials. This problem
has two aspects, illustrated by tables 12 and 19. Table
12 shows that similar materials from different deposits
can differ significantly in uranium content. Thus, the
Moroccan deposits, composed of primary apatite
nodules of simple texture, have a much higher uranium
to phosphate ratio than the comparable primary, simple
nodules of the Hawthorn formation.
Table 19 demonstrates, in terms of the Bone Valley
and Hawthorn formations, the extent to which the
components of the same deposits may differ in uranium
The following facts are evident from table 19:
1. Pebbles, which are complex in texture, are richer as a
class than pellets, which are simple in texture.
2. Pellets from the Bone Valley formation are richer as a
class than those from the Hawthorn formation.
The texturally complex pebbles represent materials
that have gone through two or more cycles of erosion
and deposition. In addition, the nodules of the Bone
Valley formation differ from those of the Hawthorn
formation only by the fact that the former have been
reworked from the latter. The differences in uranium
content cannot be explained in terms of relative concen-
tration, as it has been shown (Cathcart, 1956) that the
coarse pebbles as a class contain less P205 than the finer

TABLE 19.-Uranium content of representative clastic apatite from
the land pebble phosphate field, Florida
[Analyst, R. S. Clarke, Jr.]
Sample Formation Locality U (percent)
Individual apatite pebbles

B.L.-1 Bone Valley-- Bonny Lake mine, Ridge- 0. 011
wood, Fla.
B.L.-2- -----do------- ---- do------------------- .018
B.L.-3- -.----do------- ----- do------------------- .016
Ho-14 -----do ------- Homeland mine, Homeland, .032
Ho-15- -- do----------do do------------------ .022
Ho-16- -----do -----------do------------ 015
Ho-17- ----- do ------- ----do------------------- .021

Apatite pellets (composite samples)

Hawthorn ... Sarasota, Fla -----------0. 0059
-----do------- Venice, Fla --------------- 0067
-----do ------- Land pebble field, Fla ------ 0045
-----do -----------do------------------- .0071
-----do -----------do ------------------- .0075
Va-1 Bone Valley Varn mine, Fort Meade, Fla 008
Va-7 -------do ------------do ------------------ .011
Va-10 -----do ------- ----do ------------------- .0115
Ho-12 .-----do ------- Homeland mine, Homeland, .0127
Ho-13- -----do -----------do ----------------- .0123
Ho-14- -----do ------- ----do------------ .0089

materials. This fact is explained by the complex nature
of the pebbles which generally contain quartz and
carbonate diluents.
The coarse pebbles of the Bone Valley formation are
more abundant in the base of the deposits than higher
in the section. They commonly have internal unphos-
phatized remnants of dolomite and calcite. In addition,
they contain plastic quartz and primary phosphate pel-
lets and other pebble fragments. Both the primary and
the second-cycle laminations or textures within the
pebbles are discordant to the boundaries.
All these features, coupled with field evidence of
stratigraphic unconformity between the Bone Valley
formation and the underlying Hawthorn, indicate a
record of marine transgression with many pulses of
reworking. It seems reasonable to suppose that the
re-exposure to sea water sustained by the reworked
Bone Valley material allowed additional uptake of
uranium. Thus, the uranium content of the complex
pebble components in the Bone Valley is enhanced by
increments of uranium during successive cycles of
reworking. The pellets in the Bone Valley, also re-
worked from the Hawthorn, correspondingly contain


more uranium than the pellets in the Hawthorn. Within
the Bone Valley, however, the pellets, as expected, con-
tain less uranium than the pebbles as they evidently
have not been "worked" as frequently. Thus they are
more commonly found above the basal conglomeratic
zone, and they lack the textural complexity of the
The hypothesis of uranium enrichment during re-
working can be tested by comparing the autoradio-
graphs of a complex pebble and a pebble of first gen-
eration (pl. 11, figs. E. F). The complex pebble is not
uniformly radioactive, in contrast to the prevailing
uniformity of the simple pebble. Moreover, within the
reworked pebble the most radioactive parts are the
oldest or reincorporated materials. The strong periph-
eral enrichment shown by the fossil bone autoradio-
graph (pl. 11, fig. D) shows the distribution of uranium
that would be expected if the pebbles owed much of
their uranium to ground-water enrichment.
The absence of such peripheral enrichment or of any
gradational enrichment from the surface inward,
regardless of texture, supports the conclusion that
uranium was not introduced by ground water. The
intensity of radioactivity varies with the textures
created by reworking, providing proof of uranium
emplacement during the marine reworking. Clearly
in itself reworking served merely to expose the mate-
rial to the marine source of uranium. Fortunately,
however, reworked materials provide a demonstration
that increased exposure to such a source results in
increased uranium content.
The effect of total time of exposure to sea water may
explain differences in uranium content of similar apa-
ties from different marine deposits. Such discussion is
perforce speculative as it is impossible to assess the
chemical oceanographic factors controlling the uranium
solubility or concentration. Nevertheless, it is possible
to develop a reasonable explanation based on available
geologic facts. For example, the pellets of the Florida
and the Morocco deposits are chemically alike (table 3),
yet the nodules from Morocco contain 2 to 3 times as
much uranium as those from Florida. The analyes of
Morocco deposits in table 12 represent phosphatic marls
containing plastic apatite, clay, and minor chert. They
are practically devoid of detrital quartz and the bulk
of their particles are of fine sand size (table 8). In con-
trast, the Florida phosphates are poorly sorted con-
glomeratic sands and clays, and their pellets are com-
monly coarse sand to granules in size. The Florida
phosphates were undoubtedly much more quickly de-
posited than the Moroccan phosphates and in view of
the compositional similarity of the two, the difference

in their uranium content may be largely a reflection of
the difference in the length of exposure to sea water.

It seems clear that uranium in phosphate deposits
cannot be accounted for by discrete uranium phases,
by absorption, or by loosely adsorbed ions, radicals, or
phases. This follows from several lines of evidence in
the preceding sections:
1. Lack of uranium minerals.
2. Fluorescence of uraniferous apatite.
3. Lack of correlation of uranium content with any physical
property such as size, magnetism, color; and inability to
concentrate uranium by physical means.
4. Inability to selectivity leach uranium.
5. Lack of abnormal concentrations in specific petrographic
types of apatite.
Moreover, nuclear emulsion studies of many phos-
phorite samples from the Bone Valley and Phosphoria
formations have shown that the uranium atoms are
intimately dispersed within apatite rather than locally
segregated. Plate 13 is a photomicrograph showing the
alpha track distribution obtained from an apatite-
cemented arkose from the Gas Hills area of Fremont
County, Wyo. Concentrates of cement from this sand-
stone are found to contain tenths of one percent of
uranium. The tracks shown in the photomicrograph
are the recorded paths of alpha particles emitted dur-
ing the disintegration of individual radioactive atoms.
The intimate and homogeneous distribution of uranium
that can be inferred from the lack of clustering or of a
preferred pattern in the track distribution is typical of
unaltered phosphorite. The uranium distribution has
been illustrated with the Fremont County apatite be-
cause its high uranium content yields clearer results
than are normally obtainable with lower grade mate-
rials, which need exposures of several months and fre-
quently result in fogged emulsions. The Fremont
County apatite serves the additional advantage of
demonstrating that even in highly uraniferous apatite
random dispersion prevails rather than local concen-
The evidence summarized above and the nuclear
emulsion results indicate that uranium occurs as an
integral part of the apatite structure. The evidence
from the aluminum phosphate zone of the Bone Valley
formation in which uranium and calcium contents are
parallel (fig. 9), despite the fact that one is a mere
trace and the other a major constituent, indicates that
uranium substitutes for calcium in apatite.


The substitution of uranium for calcium is crystal-
lographically plauisible in terms of U+4 which has an
ionic radius (IR = 0.97 A) virtually identical to that
of calcium (IR = 0.99 A). The problem of maintain-
ing electrostatic neutrality in substituting a tetravalent
for a divalent ion may be reasonably discounted in this
case for two reasons. First, the content of uranium in
apatite is far too slight to create any appreciable struc-
tural dislocation. Thus, a uranium content of 0.01 per-
cent in apatite is equivalent to one atom of uranium for
every 2362 unit cells [Calo(P04)6F2] or to 1 atom of
uranium for every 23,620 atoms of calcium! Second,
the deficiency of positive charge created by other re-
placements common in apatite is orders of magnitude
greater than would be required to compensate for the
excess of positive charge caused by the replacement of
calcium by uranium. For example, from 0.1 to 1.0
percent of sodium is common in apatite as a replace-
ment for calcium, and fluorine plus hydroxyl are in
excess in carbonate-fluorapatite to an extent of 0.5 to
1.0 percent.
The suggestion has been made that uranium ions may
substitute for phosphate tetrahedra or the phosphorous
ion in apatite. Several arguments militate against this
1. The absorption of uranyl by apatite in the labora-
tory has been found to render phosphate non-
exchangeable rather than to proceed by dis-
placement of phosphate as it does of calcium
(Neuman and others, 1949a, b).
2. The uranous ion (IR = 0.97 A) is too large to
substitute for phosphorus (IR = 0.35 A) in
PO4 tetrahedra, and it normally requires a
much larger coordination with oxygen, as dis-
played by the eight-fold coordination of U(IV)
in U02 (Wells, 1950).
3. In the uranium-enriched aluminum phosphate
zone of the Bone Valley formation, the uranium
is concentrated preferentially in the calcium
and calcium-aluminum phosphates and rela-
tively impoverished at the top of the zone (fig.
9) where only wavellite, the pure aluminum
phosphate occurs. Although the wavellite
occurs as an alteration product of apatite and
crandallite (fig. 9), chemical analyses of pure
wavellite reveal that very little uranium is re-

trained by the wavellite in the course of its
replacement of the pre-existing uraniferous
minerals (Altschuler and others, 1956).

Is the uranium in apatite tetravalent? To answer
this question and questions concerning the industrial
recovery of uranium, a method was developed for the
analysis of U (IV) in apatite. The details of the method
are treated in Clarke and Altschuler (1958); however,
as the results are treated below, it is of interest to dis-
cuss some of the chemical problems involved in develop-
ing the method as they bear cogently on the geo-
chemistry of uranium.
As the total quantity of uranium in most apatite is
very small, the analyses of that portion of it which is
tetravalent presents two problems. First, the sample
must be dissolved under conditions that allow neither
oxidation nor reduction of its uranium. Second, the
minute traces of U(IV) must be separated as a uranous
cupferrate. The separation is easily accomplished by
the use of titanium, which also forms an insoluble
cupferrate, and hence can be used as a collector or
coprecipitant of the traces of uranous cupferrate
present. The initial problem of maintaining the origi-
nal valence state of uranium in solution is more vexing,
however, as it is aggravated by the presence in apatite
of many elements which directly or catalytically influ-
ence the oxidation state of uranium. Iron (present in
virtually all apatite) and cerium (present in most
igneous apatites) can, depending on their states, oxidize
or reduce uranium. In addition, air oxidation of
uranous solution is catalytically augmented by ions of
silicate, molybdenum, copper, and cobalt (Pannell,
1950), all of which may be present in apatite in trace
quantities. Of all these, iron is the element of greatest
concern owing to its ubiquitous occurrence and its quan-
tity in phosphorite relative to uranium, as illustrated
in table 20.

TABLE 20.-Iron and uranium content, in weight percent of
representative apatites from the Bone Valley formation, Florida
[Analyst, R. S. Clarke, Jr.]
Sample Location Material Fe Fe(I) U

B.L. 3-......
VA 5.......
WA 10..
P.V. 5 .--.

Bonny Lake mine-.......
Varn mine---------............
Watson mine...----...
Peace Valley mine------

Pebbles ..-----------..
Pebbles, nodules-......
Nodules ---------
.-- do.-------------

0.14 0.016
.06 .021
.04 .0076
.06 .006


It is known from theoretical and experimental con-
siderations that the iron and uranium couples can react
in the following manner:
2Fe(II) + U(VI) 2Fe(III) + U(IV)
The extent to which this interaction takes place in
nature would be governed greatly by the pH of the
medium, the relative concentrations of the individual
members of the couples, and the presence of other oxi-
dents, reductants, and stabilizing compounds, capable
of influencing the direction of the reaction. Of consid-
erable interest is the determination by Baes (1954) that
the reduction of U(VI) to U (IV) by Fe(II) is greatly
enhanced in phosphoric acid solutions at room tem-
perature. This production of U(IV) was augmented
by increasing the phosphoric acid concentration and
further promoted by the presence of fluoride ion (Baes,
1954), which forms stable complexes with both Fe(III)
and U (IV). Schreyer (1954) has shown that uranium
(IV) forms stable orthophosphate complexes in phos-
phoric acid solutions to an extent that limits air
oxidation of U(IV).
Additional verification of the efficacy of the above
reactions in acid solutions is obtained from experience
in the industrial recovery of uranium as a byproduct of
phosphoric acid in triple-superphosphate production.
In this process, apatite is treated with sulfuric acid
to produce phosphoric acid and gypsum. The final
product, monocalcium phosphate is manufactured by
treatment of additional apatite by the phosphoric acid.
Uranium is extracted from the phosphoric acid phase
of this process by use of alkyl pyro-phosphoric acids
(Ellis, 1952). It has been found in the course of re-
covery studies that the uranium must be sequestered
from the phosphoric acid as the tetravalent ion. To
effect complete recovery, the acids are passed over steel
filings or punchings prior to extraction.
Thus, in view of the ability of ferrous iron to reduce
U(VI) to U (IV) and the readiness with which both
U(IV) and Fe(III) form stable compounds with the
fluoride and orthophosphate ions, one could produce
U(IV) merely in the course of dissolving apatite in
order to make the analysis!
Most of the above experimental information on the
independent behavior and mutual interactions of
uranium and iron in phosphoric acid systems was
obtained at room temperature and at concentrations of
phosphate relative to uranium and iron comparable to
those prevailing in apatite and many natural solutions.
The acidities in the laboratory studies, however, are so
much greater than those prevailing in nature as pos-
sibly to invalidate the application of the results to the
interpretation of uraniferous apatite. Moreover, as it

is impossible to define all of the variables that have
affected the primary state of uranium in apatite, or all
of the factors which may have altered it after emplace-
ment, it is still necessary to determine chemically
whether apatite contains tetravalent uranium. Further-
more, a method for this determination is also valuable
as a means of assessing the relative quantities of
hexavalent and tetravalent uranium during various
stages of the industrial recovery process.
In devising a method for the analysis of U(IV) in
apatite it was necessary to depress the oxidation and
reduction, respectively, of the tetravalent and hexa-
valent uranium inherently present in apatite, and to
establish that the analytical recovery is both high and
consistent. Lacking proof that U(IV) actually occurs
in apatite, experiments were performed on solutions of
natural apatites and uranium-free synthetic apatites to
which known spikes of uranous and uranyl solutions
were added. Table 21 is a representative summary of
a number of these experiments.

TABLE 21.-The recovery, in weight percent, of uranium (IV)
from 1.5 M phosphoric acid solutions of natural and synthetic
[Assembled from Clarke and Altschuler, 1958]
Initial Uranium Iron added
4 uranium added U(IV)
Material deter-
W Total U U U Fe Fe
U (IV) (IV) (VI) (II) (III)

1 Synthetic apatite--.--- --------- 0.050 --------------.-----..........--........ 0.050
2 Synthetic apatite-......-----------............. .048 ------------ 0.100 .050
3 Synthetic apatite---------------------- 0.030 0.100 ........--------.0025
4 Apatite pebbles (B.
L.-3), Bonny Lake,
Fla --------. -- 0.016 0.012 .050 .......-..............--------------- .061
5 Apatite pebbles (B.
L.-3), Bonny Lake,
Fla---------------.......-....016 .013 .048 ............--------------.. 100 .060
6 Apatite pebbles (B.
L.-3), Bonny Lake,
Fla .-..-.......--------..------ .016 .013 .048 ---.... .100 ....... .056

It was determined (table 21, samples 1 and 4) that
uranous uranium can be recovered from apatite. By
additions of ferrous and ferric iron it was further
determined that the reduction of uranyl (table 21,
samples 3 and 6) and the oxidation of uranous uranium
(table 21, samples 2 and 5) are both negligible. It
should be noted that the percent of recovery of U(IV)
in table 21 is virtually complete in all cases where
U(IV) was added, and that it is less than 10 percent
where only U(VI) was added (table 21, sample 3).
A further demonstration that the U(IV) values
obtained are not merely governed by the experimental
conditions is the production of more U(VI) and less
U(IV) by treatment of the sample prior to analysis.


Sample B.L.-3 was heated at 6000 C for 4 hours in air.
This treatment would be expected to oxidize at least
part of the U(IV) present. A comparison of values
.obtained before and after treatment is given in table 22.
Thus, by changing the inherent U(IV) in a predictable
manner prior to analysis, and confirming the change by
recovery, independent proof is had that the inherent
U(IV) normally obtained is real, and not a function
'of the method of recovery.

TABLE 22.-Percentages of uranium and of U(IV) in sample
B.L.-8 before and after oxidative heating
[Analyst, R. S. Clarke, Jr.]



U (IV)

B.L.-3-- Unheated --------------- 0. 016 0. 013 81
B.L.-3-- Heated at 4500 C, 4 hours- 016 0085 53
.3 002 1
B.L.-3-- Heated at 600* C, 4 hours- .020 :. 002 12
Total U not determined on heated sample. Percent reduced U thus slightly high
due to being uncorrected for weight loss on heating.
* HsPO solution of sample.
HCl-NHjOH.HCI solution of sample.

Having established the validity of the method, the
consistency of the results and the applicability of the
method to a variety of materials was next studied. It
was found that igneous fluorapatite, due to its much
greater crystal size and its lack of carbonate, did not
dissolve as readily as sedimentary apatite. As a result
two different methods of solution have been used. Sedi-
mentary apatite is dissolved in cold 1.5 M phosphoric
acid and igneous apatite is dissolved in cold 1.2 M
hydrochloric acid containing 1.5 percent hydroxylamine
hydrochloride. The U(IV) recovery from phosphoric
acid is virtually complete, as demonstrated in table 21.
The U (IV) recovery from hydrochloric acid was estab-
lished from spike experiments to average 79 percent.
The range in percent recovery was from 67 to 87 per-
cent; however, the standard deviation from the average
of 79 percent was found to be only 5.4 percent (Clarke
and Altschuler, 1958). Therefore, in analyses based on
hydrochloric acid solutions a 20 percent correction will
be applied to the value obtained, for a close approxi-
mation of the real value. It was found (Clarke and
Altschuler, 1958) that materials with more than trace
amounts of oxidizing or reducing elements, like cerium
or iron, may not yield good results.
The U(IV) contents obtained on a variety of sedi-
mentary and igneous apatites are given in tables 23
and 27.

The analyses in table 23 reveal a wide spread in the
percentage of tetravalent uranium in apatite. The
remainder, in all cases, is safely presumed to be hexa-
valent in view of the instability, and the absence in
natural substances, of tervalent and pentavalent
uranium (Garrels, 1955) and of the fact that mono-
valent and divalent uranium are completely unknown in
natural solutions (McKelvey and others, 1955).
Several prevalent notions are open to serious ques-
tion as a result of these analyses. It was heretofore
generally assumed that the traces of uranium are of
the same chemical nature in all apatite deposits; that,
once fixed in apatite, uranium is not changed except
insofar as the host mineral is altered; and that the
uranium within any single specimen of apatite is
entirely hexavalent or tetravalent. These assumptions
must all be discarded.
Despite the variety displayed in table 23, the distri-
bution of U(IV) proportions is not haphazard and
several important generalizations emerge if the data
are regarded in the light of geologic occurrences.
Initially, to establish the validity of subsequent in-
ference, it should be noted that materials with identical
histories yield identical results. Thus, samples Ho-14
through Ho-17 in table 23 are composites of the same
petrographic fraction from each of four successive beds
in the Bone Valley formation. The particular strata
are thin and therefore have the same general relation
to the zone of aluminum phosphate alteration higher
in the section. Although the total uranium content
varies by as much as 100 percent among the samples,
the percentage of U(IV) is virtually identical through-
out, within the limits of the fluorimetric analyses of
The reflection of geologic history in the U(IV)/U
percentages is further illustrated by several other
groups of samples. Nodules from several localities in
the Pacific Ocean off the coast of southern California,
have sustained different degrees of submarine Irework-
ing and are of somewhat composite age, having also
sustained accretion to differing extents since Miocene
time. They are similar but not identical. Likewise
their U(IV) /U percentages are similar but not identi-
cal. The apatite pellets from the Hawthorn formation
were collected from two beaches along the central west-
- Florida coast. Both samples represent present-day
3 marine reworking of Miocene apatite. Their U (IV) /U
percentages are virtually identical.


TABLE 23.-Total and tetravalent uranium content of sedimentary
[Analyst, R. S. Clarke, Jr.]

ample Material Total U U(IV)
Sample (percent) (percent) U

Bone Valley formation (Pliocene)

B.L.-1.....- Black pebble ..--- .----------.- 0.0011 0.010 91
B.L.-2 --- Dark pebble-------- ---- .0089 .0056 63
B.L.-3 ----- Fine pebble composite .---------- .016 .013 81
Ho-14 4-8 mm pebble composite..-------. .032 .016 50
Ho-15 --- -----do--------------------------- .022 .011 50
Ho-16--------do --------------------------- .015 .007 42
Ho-17 ..-------- do ...----------------------- .021 .010 48
Va-7 .- Apatite pellets ------ ------- .011 .005 45
Va-7a---.----- do --------------.---------------- 009 .006 67
WA-10 ...- do---......- ------------.- .0075 .003 40
PV-5 ..--..---. do----...... .--------------. .007 .003 43

Hawthorn formation (Miocene, recently reworked on Gulf Coast beaches)

GP-1 --- Apatite pellets ------------ 0.0067 0.0039 58
GP-2 ....----- do------ ---------- .0059 .0037 63

Tennessee phosphates, Bigby formation (Ordovician)

Ak-2-..... Limestone, phosphatic ..---........ 0.00004 0.00001 25
Ak-1 ..-... Brown-rock phosphate (Cenozoic .00074 .00002 3

Moroccan phosphates (Senonian)

Mor-3.....-- A daily production average. ------ 0.012 0.0018 15
Mor-5-.....--..do- .-----------............ .012 .0015 13
Mor-11-.........do-----... .-------------. .008 .0006 8

Phosphoria formation (Permian)

Conda-300 Apatite pellets ------ ------- 0.0063 0.0015 24
RAH-185-.. .do ----------- 017 .003 18

Pacific Ocean, southern California (post-Miocene)

69 ---.------... Phosphate nodules..................-------- 0.0089 0.0061 69
106 -----------.... do......-------------------------- .0068 .0040 59
127 -----------do ......--------------------------- .0041 .0028 68
158 -------.........----do ....... --------------------- .0081 .0050 62
162.---------..........--do -------....-------------------- .0125 .0093 74
183 ------ -----do ------.. --------------------- .0051 .0028 55

Fossil bone

1 ------ Hells Creek formation------------- 0.015 0.0004 3
RW-4468-- Taxpayer sample ----...............-----------.... .078 .054 70-76
W-3841.........-do.........------ ..-------------------- .85 1.40 ----- -
.82 .50 61

Phosphatic arkose

Gas Hills, Wyo-...................----------------- 0.74 10.14 -----
.17 23

1 Determination from HC1 solution; uncorrected for 20 percent deficiency in re-


In marine materials that have not been postdeposi-
tionally oxidized and that are of relatively young age,
more than half of the uranium is tetravalent. The
nodules from the Pacific serve as an example as they
have been interpreted by Dietz, Emery, and Shepard
(1942) to have been submerged since their formation
in post-Miocene time. Samples B.L.-1, B.L.-2, and
B.L.-3 (table 23), in which 63 to 91 percent of the
uranium is tetravalent, were collected from clayey,
impermeable strata in the basal conglomerate zone of
the Bone Valley formation. Sample B.L.-1 has minute
pyrite crystals and petroliferous organic matter dis-
persed throughout; however, the clay in which it rested
was devoid of organic matter and pyrite. The pebble
thus reflects the reducing conditions of its time and
environment of formation and has not been postdeposi-
tionally weathered. Its content of U(IV), 91 percent
of the total, is the highest we have found. Samples
B.L.-2 and B.L.-3 are from higher strata, contain less
organic matter, are lighter colored than B.L.-1, and
have no pyrite. Less of their uranium is tetravalent.


Material that is obviously weathered contains pro-
portionally less U (IV) than comparable or equivalent
unweathered material. This may be illustrated by com-
paring the U(IV)/U percentage obtained from the
slightly phosphatic limestone of the Bigby formation
with the much lower value obtained from the equivalent
"brown-rock" phosphates derived from the Bigby by
subaerial residual concentration. See table 23 and fig.
12 for relations of the samples.) A second illustration
is obtained by comparing samples Ho-14 through
Ho-17 with samples B.L.-1, B.L.-2, and B.L.-3. The
two groups were collected from different mines (Home-
land and Bonny Lake) and from different parts of the
stratigraphic column and, hence, are not precisely
equivalent. They are comparable, however, in their
common origin, in their textural complexity, and in
their overall composition. The Homeland samples were
collected from an oxidized zone in the middle of the
Bone Valley formation. Their associated clays are
mottled and limonite stained, and the pebbles consti-
tuting the samples are light colored, and somewhat
friable in contrast to the dark pebbles from the green,
clayey, conglomerate at the base of the Bonny Lake
mine. The appreciably lower U (IV) percentages in the
Homeland materials demonstrate that some uranium
may be oxidized in place in the host apatite. The
enormous internal porosity of most phosphorite


* (Hendricks and Hill, 1950; Jacob and Hill, 1953)
* would greatly facilitate such oxidation, which, in fact,
is visible in the form of limonite impregnations and
micro-concretions dispersed throughout the apatite
It may be that the U(IV) /U percentages in marine
phosphorites are a function of radioactive disintegra-
tion and thus, indirectly, of age. This hypothesis pre-
supposes that all, or most, of their uranium is initially
U(IV), a suggestion favored strongly by the analyses
of samples from the Bonny Lake mine and of the
Pacific Ocean nodules. The thesis further demands
that the U(VI) present is produced by auto-oxidation,
or by some secondary effect whose onset is governed,
both as to rate and intensity, by the progress of radio-
active decay. If the U(VI) content of apatite is related
to age, rather than an independent result of oxidative
weathering, it should be possible to demonstrate a pro-
gressive increase in U(VI) content in a series of pro-
gressively older apatites. Such tests may soon be
practicable for igneous apatites but their application
to phosphorites is unfortunately prevented at present
by the difficulty of obtaining materials assuredly free
of oxidation from normal weathering. In addition to
the difficulty of obtaining such materials, the few cri-
teria at our command for establishing such freedom
from oxidative weathering may be relatively insensi-
tive. Such criteria are the presence of syngenetic
glauconite, pyrite, and organic matter, or the presence
of reduced ions in the apatite. Rarely, it may be pos-
sible to develop preclusive geologic evidence, such as a
history of continual protective burial.
In exploring the possible relations of U(VI) content
to age in phosphorites, only three groups of samples
may be compared with a modicum of assurance that
they have not been weathered. They are (table 23) the
Pliocene Bonny Lake pebbles, which contain pyrite and
organic matter and were found in green clay beds; the
Quaternary Pacific Ocean nodules, which have re-
mained submerged since their formation; and possibly
the pellets from the Permian Phosphoria formation, as
these contain minor amounts of organic matter. It may
be significant that the two Phosphoria samples show
considerably lower U(IV)/U percentages than the
other two groups, and that the two latter groups are
moderately close in value. However, the extreme view
that all of the uranium in apatite is initially U(IV)
is not supported by the data. Although, in some sam-
ples (as in the Bonny Lake pebbles), favorable re-
ducing conditions may cause substantially all the

uranium in apatite to be taken up as U(IV), the exist-
ence of significant amounts of U(VI) in the young and
recently dredged Pacific Ocean nodules suggests that
the initial U(IV)/U(VI) ratio in apatite is influenced
by the oxidation conditions in the marine environment.
It is also possible however that submarine, postdeposi-
tional oxidation can lower the initial U(IV)/U(VI)
Only three analyses for U(IV) have been made on
fossil bone; however, the spread in relative proportions
of U(IV) and U(VI) (table 23) is indicative of the
diversity that might be expected in view of the con-
ceivable variations in the composition of the transport-
ing ground water from which the uranium is emplaced,
in the environment of deposition, and in the state of
degradation of the organic matter that may be asso-
ciated with the bones.
Although it is true that apatite readily abstracts
U(IV) from solution, in a strongly oxidizing environ-
ment the mere presence of apatite is not necessarily
effective in producing or even retaining U(IV). To
the contrary, it was found that samples of Morocco
apatite pellets and fossil bone that are richly encrusted
by hematite have the capacity to oxidize uranous ions
in phosphoric acid solutions, as the experiment in table
24 demonstrates. Iron free synthetic and natural apa-
tite used as "controls" do not cause such oxidation.

TABLE 24.-Experiment showing reduction of U(IV) by phosphoric
acid solutions of hematite-rich apatite samples
[Analyses in weight percent. Analyst, R. S. Clarke, Jr.]
Initial uranium U(IV) U(IV)
Material added deter-
E Total U U(IV) mined

Synthetic apatite ........-----------. -------------------- -------- 0.050 0.050
Bonny Lake pebbles, B.L.-3 --------------...... 0.016 0.013 .050 .061
Moroccan apatite, hematitic Mor-11-.---------- .008 .0009? .050 .046
Fossil bone, hematitic Hell Creek 1 ------------- .015 .0004? .050 .020

The data given in table 24 show that the completing
capacities of phosphate and fluoride ions for Fe(III)
and U(IV) respectively are not adequate to prevent the
oxidation of U(IV) in the presence of excess ferric
iron. As this is true in solutions of freely mobile ions,
it indicates that slowly dissolving apatite in the en-
vironment of such reactions would not prevent the
oxidation of U(IV) by excess ferric iron. They also
show that the very low values for U(IV) in samples
Mor-3, 5, and 11 and Hell Creek 1 (table 23) may be
due to oxidation effects during analysis, although it is
to be expected that they would be low inherently. In


any case the method is not trustworthy for materials
with more than trace quantities of oxidizing impurities.
It is clear from the above experiment, however, that
high U(VI)/U percentages may be anticipated where
uranium is emplaced primarily from ground water.
The phosphatic arkose sandstone from the Gas Hills
region of Fremont County, Wyo., is a striking illustra-
tion of secondary uranium fixation by apatite. X-ray
powder diffraction analyses show the material to be a
carbonate-fluorapatite. Deposits of this material occur
as cements scattered through the arkosic sandstones and
conglomerates of the upper part of the Wind River
formation.5 Associated with them are "paints" of
autunite and other secondary uranyl minerals; how-
ever, the sample used was shown by nuclear emulsion
study and inspection under ultra-violet light to be free
of uranium minerals. The apatite cement is attributed
to secondary phosphatization, and the uranium is
attributed mainly to secondary emplacement in the
porous cement (R. G. Coleman, oral communication,
1956). This interpretation is supported by the occur-
rence of the uranyl phosphate and vanadate minerals
throughout the region.
The alpha-track film of the specimen (pl. 13) reveals
intimate and homogeneous dispersion of uranium
throughout the cement, and a virtual absence of any
uranium atom concentrations. It cannot be determined
whether much or most of the uranium was originally
fixed as U(IV) and subsequently oxidized or whether
the U(VI) now present was emplaced substantially as
uranyl. The local abundance of uranyl minerals and of
limonite films in rocks other than phosphorite indicates
that oxidizing solutions have permeated the area and
have introduced uranium as uranyl.
The material emphasizes that U(VI) like U(IV)
can be thoroughly dispersed in apatite and may be
emplaced in a manner analogous to that demonstrated
by Neuman and his coworkers (1948, 1949a, b). They
have shown that glycol-ashed bone will fix appreciable
quantities of uranyl ion by exchange with surface
calcium. This reaction was found to be surface limited
(Neuman and others, 1949b) and involved the dis-
placing of 2 moles of calcium for every mole of
(UO2) +2 taken from solution and the rendering of 2
moles of phosphate nonexchangeable. They suggest the
formation of a compound analogous to uranyl pyro-
phosphate to explain these results.

B All Information on occurrence has been obtained from R. G. Cole-
man, who kindly supplied the specimen, the second most uraniferous
known to us.

Regarding the form in which uranyl complexes with
apatite, several types of compounds should be con-
sidered. Uranyl pyrophosphates, or uranyl orthophos-
phates of the torbernite type could both be stable within
the range of conditions presumable oxidativee, weakly
alkaline to weakly acid). In addition, complexes of the
type U02HP04 could presumably form by surface reac-
tion with apatite; F. S. Grimaldi (oral communication,
1955) verifies that UO2HPO4 could be stable under
such conditions. We believe that uranyl orthophosphate
compounds are more probably the form in which pri-
mary U(VI) would occur in phosphorites. These would
not require the polymerization entailed in pyrophos-
phate formation and would avoid the anomalous ex-
change of 2 divalent calciums for 1 divalent uranyl
group. It is quite possible that an appreciable part of
the U(VI) present in unweathered phosphorites is
chemisorbed on the surfaces of growing apatite crystal-
lites and, upon the complete development of the apatite
nodules, the U(VI) appears to be intimately and ran-
domly dispersed throughout the apatite. Sheldon
(written communication, 1957) also favors the chemi-
sorption of some hexavalent uranium in the rocks of
the Phosphoria formation.

The above mechanisms have one characteristic in
common. Their surface-limited nature is virtually pre-
dictable in view of the asymmetry and large size of the
uranyl anion (3.4 X 1.4A, Connick and Hugus, 1952)
relative to the calcium ion (IR = 0.99 A). Thus,
although the reactions so clearly documented by Neu-
man and his colleagues involve ion exchange, the results
obtained indicate extensive structural rearrangement.
It is therefore difficult to visualize any intra-crystalline
solid solution by such a mechanism, first, because of
the structural dislocation involved in formation of
pyrophosphate and displacement of two calcium for
one uranyl ion, and second, because the large uranyl
anion could not migrate to any extent through the
apatite structure. The channels within the hexagonal
network of the fluorapatite structure, being barely wide
enough to accept the fluoride ion (IR = 1.36 A, Green,
1953), could not accommodate the uranyl anion. Sixty
percent of the calcium ions in apatite line the vertical
channels and would thus be inaccessible to postdeposi-
tional substitution by uranyl. On the other hand, the
U (IV) ion could fit easily, and migrate a short distance
within the channels, which are not as fully occupied


as the walls and hence permit greater mobility to
migrating ions. That such migration occurs is fully
substantiated inithe replacement of (OH) by F that
takes place in fossilization.
Structural considerations thus dictate that much
more calcium is accessible for replacement by U(IV)
than by U(VI). Furthermore, the U(IV) substitution
does not involve ,any structural dislocation and the
positive charge excess that would result is easily com-
pensated by other known replacements of greater mag-
nitude. It is not unusual, therefore, that samples as
uraniferous as RW-4468 and W-3841 (table 23) should
have as much as 70 and 61 percent respectively of their
uranium in tetravalent form, despite the fact that
ground water may be presumed to carry mainly uranyl
There are several compelling reasons for detailed
examination of the relations between excess fluorine
and uranium in marine apatite, as follows:
1. Excess fluorine is characteristic of marine sedimentary
apatite, in contrast to normal igneous fluorapatite.
2. Marine apatite is characteristically more uraniferous than
igneous apatite.
3. UF4 could be stable under the pH and Eh conditions of
formation, and of subaerial preservation, of marine
phosphorite. Although the formation of independent
fluoride minerals is conceivable only under conditions
of much greater salinity than those characterizing
marine apatite formation (Kazakov, 1937), the possi-
bility of a complex similar to UF4, as an integral part
of apatite, is strongly suggested by the following:
A. The calcium ions lining the channels in the apatite
structure are located in the same horizontal plane
as the adjacent fluorine ions.
B. These channel calcium are coordinated to fluorine
and oxygen (fig. 3), and as the excess fluorine could
reasonably replace oxygen, or lodge within the
channels in the apatite structure, the existence of
UP4 in apatite may be reasonably postulated, par-
ticularly as U*' may be considered to replace
4. The postdepositional uptake of fluorine and uranium by
various forms of sedimentary apatite and bone involve
separate replacements; however, their concomitant
operation may mask, or be augmented by, their mutual
completing ability.
A general correlation between total fluorine and total
ium was demonstrated in studies of seven Penn-
Ivanian phosphatic shales of Kansas by Runnels and
ers (1953). In all likelihood this reflects only the
that uranium content varies directly as apatite
Mten in unaltered rocks of similar composition in

which apatite is the only, or major, uranium host.
Similar relationship could be expected in a group of
phosphatic limestones and has been demonstrated for
Mona Island rocks (fig. 3) and in some suites of rock
from the Phosphoria formation (table 9).
The possible genetic relation between excess fluorine
and U(IV) is marine apatite should be tested in two
ways. On the assumption that all of the uranium was
originally fixed as U(IV), excess fluorine should be
compared with total uranium. Figure 13 is a scatter
diagram on which total uranium is plotted against
excess fluorine,6 as computed from a random selection
of 50 of the analyses given by Thompson (1953, 1954)
for Phosphoria samples from Trail Canyon, Utah;
Coal Canyon, Wyo.; and Reservoir Mountain, Idaho.
It can be seen that total uranium is independent of
excess fluorine in rocks of the Phosphoria.
Assuming that only the U(IV) demonstrably present
in apatite was emplaced as such, a comparison was
made (fig. 14 of the few data available (table 25)).
The wide scatter of the points in figure 14 indicates
that U(IV) and excess fluorine are independent of one
another in marine phosphorite. The data are admittedly
few; however, the well-defined trends that can be dem-
onstrated to exist among total fluorine, P2s0, and
uranium, for groups of samples no larger than this
(fig. 5, and Runnels and others, 1953) exemplify the
extent of mutual dependence that could be expected
between F and U(IV) if uranium occurred in apatite
as a result of UF4 formation.

The demonstrated secondary enrichment of uranium
in apatite during the course of marine reworking allows
the establishment of controls for studying the nature
of the uranium extracted from the ocean source. If the
uranium additions incident to reworking are mainly
U(IV), the U(IV)/U percentages should be higher
in the reworked materials of a deposit. Samples were
taken in the phosphatic terrace gravels in the vicinity
of Charleston, South Carolina, for tests of this thesis.
The phosphatic fraction of the samples consisted of
sand-size pellets and detrital bone apatite, mainly
water-worn shark tooth fragments, plus rounded
granules and pebbles of bone and phosporite. In addi-
tion, the samples contained quartz and shell fragments.
The coarser material was found to be texturally more
complex, indicating more reworking. Accordingly, each
Based on 3.8 percent of fluorine in pure fluorapatite and 38 percent
PsOs in carbonate-fluorapatite. Therefore excess fluorine is amount In
excess of theoretical F/P2Os weight ratio of 1/10.











0.1 0.2 0.3 0.4 0.5 0-


FIGURE 13.-Uranium vs excess fluorine in Phosphoria formation (data recomputed from Thompson, 1953).


0.1 0.2 0.3 0.4 0.5 0.6 0.7
FIouRs 14.--U(IV) vs excess fluorine in various phosphates.

TABLE 25.-Data on excess fluorine and U(IV) in miscellaneous
apatite and phosphorite
[All analyses from U. S. Geological Survey Laboratory; F and PsOs on Pacific nodules
are given in Dietz, Emery, and Shepard (1942)]

Sample Material Locality Remarks
F POs cess U(IV)

PV-5_- Apatite Peace Valley 3.8 34.5 0.3 0.0027 Bone Valleyforma-
pellets. mine, Fla. tion.
Wa-10...-do-.... Watsonmine, 3.8 36.6 .1 .0029 Do.
BL-3-.....do .... Bonny Lake 3.8 34.9 .3 .013 Do.
mine, Fla.
Mor-11 -do ... Khouribga, 4.05 33.4 .7 .005 Average sample,
Morocdo. daily production.
9---... Phos- Pacific 3.31 29.56 .35 .006 Sample members,
phate Ocean. these of Dietz and
nodule. others (1942).
106 --.....do------do-.----.. 3.12 29.19 .20 .004 Do.
S127--..... do -..---... do-. 3.07 28.96 .17 .003 Do.
8---......do...------. do- 3.15 29.09 .24 .005 Do.
162---do --------do -2.47 22.43 .23 .009 Do.
183 --.do..--------do-----3.36 29.66 .39 .003 Do.

sample was split into two fractions on the basis of size
and analyzed for P205, U, and U(IV). The results are
given in table 26.
' The samples from each locality were fractionated by
screening and the obvious shell fragments were hand-
'picked from the coarse fraction in each case. It was not
practical to remove shell fragments from the fine frac-
ion. Table 26 shows that the coarse fraction contains
P205 at all 4 localities and more uranium in 3 of
the 4. The lower P20s reflects the greater textural

TABLE 26.-Relative contents of U, U(IV), and P205 in phosphatic
terrace gravel in the vicinity of Charleston, S. C.
[Analyst, R. S. Clarke, Jr.]

U PsOs UX1000 U(IV) U
Sample Material (per- (per- P, O (per- per-
cent) cent) cent) cent-

Accabee Flats, Ashley River, Charleston, S. C.

Acc-1... +2 mm pebble. .....-----. 0.020 28.4 0.70 0.016 80
Ace-1..... Medium to fine sand .-..... .020 31.7 .63 .013 65
Acc-2-- .. Medium to fine sand -- --. .018 30.2 .60 .013 72

Micah Jenkins Nursery, Johns Island, S. C.

Micah-1.. +2 mm pebble ------------ .027 27.6 .98 0.021 78
Micah-1._ Medium to fine sand .....- .015 31.2 .48 .011 73

Ninemile Station, S. C.

Nine-1_-- +4 mm pebble-------------. 037 27.5 1.35 0.025
Nine-2---. Medium to fine sand-...... .021 31.4 .67 .010

Lambs, S. C.

Lamb-1.- +2 mm pebble ------------- .015 28.6 .52 0.0097 65
Lamb-1_. Medium to fine sand .....- .025 31.5 .79 .012 48

complexity and greater percentage of inclusions result-
ing from reworking. The generally higher uranium
content reflects the enhancement of uranium during
reworking. The fact that in all samples the U(IV) /U
percentages are significantly higher in the reworked
materials proves that the additional increments of
uranium are largely tetravalent.


The theoretically plausible occurrence of U(IV) in
apatite has been confirmed in these investigations by
demonstrating that U(IV) actually does occur and
indeed is the predominant ion species of uranium in
some phosphorites. In addition, considerations of the
crystal chemistry of apatite lead to the conclusion that
structural emplacement of (UO02)+2 is most probably
surface-limited and, therefore, not as favorable as struc-
tural emplacement of the U+4 ion. Nevertheless, the
fine-grained nature of most phosphorite results in an
exceptionally large inherent surface area (Hendricks
and Hill, 1950; Jacob and Hill, 1953) which can accom-
modate all of the uranium in typical phosphorite as
chemically adsorbed uranyl, as has been pointed out
by Hendricks and Hill (1950) and Sheldon (written
communication, 1957). It is thus important to examine
the significance of the wide range in the U(IV) /U


percentages of marine phosphorites (table 23). Do the
data accurately represent the primary state of the
uranium deposited in apatite ?
In answering the above question we must consider
laboratory and field evidence bearing on the relative
stabilities of U+4 and UO2+2 under subaerial exposure.
Garrels has shown (oral communication, 1955, and
McKelvey and others, 1955) that the reduction of
uranyl to uranous uranium can be effective at 250 C
and a pH of 4 or higher, for small amounts of uranyl
and a wide range of ferrous iron concentration. The
previously cited stabilizing effect of phosphate and
fluoride on this reaction can appreciably increase the
production of U(IV). It is also probable that U(IV)
may be produced from U(VI) by the oxidation of
organic matter or H2S. Both Gruner (1952) and Miller
and Kerr (1954) have successfully precipitated UO2
by such reactions at low temperature and pressure.
The presence of apatite in the milieu of such reactions
would fix the U(IV) so produced, by ion exchange
with calcium. Beyond this, by its extracting action,
apatite would actually cause further production of
U(IV) for its own fixation, as it would disturb the
achievement of equilibrium. This action would be even
more pronounced than the stabilizing effect of com-
plexing with phosphate or fluoride in solution as
equilibrium may be reached in the latter case.
All of the above reactions favoring production of
U+4 presuppose the interaction of uranium with
another reducing couple. On the other hand the rela-
tive instability of U+4, in the form of U02, to air
oxidation is revealed by the many secondary minerals
associated with vein deposits of uranium and the fact
that uraninite or pitchblende is virtually never pure
U02, but more commonly UOs, or some intermediate
value. We have shown experimentally (table 24) that
apatite pellets encrusted with iron oxide have the
capacity to oxidize U(IV) in solution to uranyl. From
the above evidence, as well as the fact that (U02) +2 is
the dominant ion under most natural conditions (Gar-
rels, 1955), it is reasonable to expect that weathering
causes a lowering of the initial U(IV)/U(VI) ratio
in marine phosphorites. The very features of excessive
porosity and surface area that permit contemplation of
adsorbed (U02)+2 in phosphorites also permit thor-
ough permeation by air and oxygen-bearing water.
Such extensive contact with oxygen could effect appre-
ciable alteration of U+4 to (UO2) +2, despite the fact
that other, more obvious, weathering changes might not
occur. External oxidation would be augmented by the
disruptive effects of radioactive disintegration on the
crystal structure and by the liberation of oxygen from

water as a result of apha particle bombardment
(Ellsworth, 1925).
Conclusive evidence that weathering has lowered the
initial U(IV)/U(VI) ratio is obtained in the Ten-
nessee phosphorites. There, the residually accumulated
"brown-rock" ores maintain essentially the same
P20s/U ratio as the underlying parent limestone (table
18), thus proving that uranium has not been added or
subtracted. In contrast the UT(IV) content has been
lowered from 25 percent to 3 percent of the total
uranium (table 23).
The fact that the highest U(IV) /U percentages (and
therefore the highest U(IV)/U(VI) ratios) are ob-
tained from youngest and least weathered rocks be-
comes doubly significant in view of the above evidence.
The high U(IV) contents of pebbles from the basal
Bone Valley formation of Pliocene age and of nodules
dredged off the coast of California (p. 72 and table 23)
suggest that the uranium fixed by apatite from marine
waters is largely tetravalent and that long exposure has
allowed oxidation to lower the U(IV) /U perventages
in the Phosphoria and Bigby formations.
Few samples of the Phosphoria and Bigby forma-
tions have been studied and the hypothesis of initially
higher U (IV)/U (VI) ratios must be tested with more
data and specific comparisons of weathered and obvi-
ously unweathered materials. It can be urged in sup-
port of the idea, however, that McKelvey and Carswell
(1956) infer a history of weathering from changes in
P205 and uranium contents in various exposures of the
Phosphoria formation, and they note unquestionable
instances of weathering in other exposures. In addi-
tion, it is difficult to assume that the U (IV) /U percent
of 25, for limestone of the Bigby formation in the Ten-
nessee phosphate field, represents the initial U(IV)
content, and not a lowered value, superimposed by
oxidative weathering. In view of the striking field
evidences of weathering in the origin of the Tennessee
"brown-rock" ores (pl. 12) and the substantial decrease
of their U (IV)/U percentage from 25 to 3, it is highly
probably that the immediately underlying parent lime-
stone, the Bigby formation, was also altered and that
its primary U(IV)/U percentage was appreciably
higher than 25.
It is not possible to suggest how much of the primary
uranium taken up by marine apatite was U(IV). It is
clear however that significant amounts of U(VI) are
present initially, even though UT(IV) may be the domi-
nant species, and that the primary UT(IV) /U(VI) ratio
in apatite is undoubtedly influenced by the Eh of the
marine waters.


In accordance with Goldschmidt's empirical rules,
U(IV) should be preferentially emplaced or "cap-
tured" in apatite due to its size identity with the Ca+2
ion and the greater binding capacity of its higher
charge. Considering the existence of such preferred
emplacement in sedimentary apatite (see p. 59), and
also the progressive increase in uranium content in the
younger products of differentiation (Larsen and Phair,
1954; Adams, 1954), apatite might be expected to con-
tain more uranium in the more felsic members of an
intrusive series. Unfortunately, too few groups of
related intrusive bodies have been investigated exten-
sively enough to test such generalizations. To demon-
strate the type of progressive concentration that might
be a consequence of capture, it is necessary to compare
the uranium contents and the relative amounts of apa-
tite and other uranium-bearing minerals in each rock
type, and also the uranium content and the relative
volumes of the different rock types. It is, in effect, the
problem of illuminating the total budget of uranium
before one can understand individual expenditures of it.
The determinations of U and U(IV) in apatite re-
quire as much as 100 milligrams of pure mineral. To
obtain this quantity of a small accessory mineral large
samples of representative rock specimens must be care-
fully processed through several time-consuming opera-
tions. The paucity of uranium analyses for apatite in
the literature reflects the prohibitive nature of the
task. The analyses 7 given in table 27 are hardly rep-
resentative of all igneous apatite. They, nevertheless,
give some idea of the range and typical values for
uranium content. The apatites, all cases, were optically
clear, subhedral to euhedral, and contained few inclu-
sions. An exception is the apatite from Mineville, N. Y.,
which is not properly igneous in origin, is exceedingly
high in rare earths, and has many inclusions (McKeown
and Klemic, 1956). Excluding the unusual Mineville
material, all the apatites fall within the range of 0.001
to 0.012 percent of uranium. Most of them fall within
the narrower limits of 0.004 to 0.008 percent of uranium.
The rocks sampled come mainly from a few batholiths
and the selection is heavily weighted in favor of inter-
mediate "acid" igneous types. As exemplified by the
Boulder Creek batholith, they reveal the fact that the
range in uranium content of apatites from a group of
related rocks is not large in contrast to the dispersion
that may obtain in sedimentary deposits. Significantly,

S'We are greatly indebted to H. W. Jaffe, David Gottfrled, E. S.
Larsen, Jr., and George Phair, for supplying many of the apatite

TABLE 27.-Uranium and U(IV) content of igneous apatites
[U(IV) was determined after solution of sample in HCI-NHsOH-HCO solution.
Values shown are corrected to compensate for the 80 percent recovery by this
method (Clarke and Altschuler (1958). Analyst, R. S. Clarke, Jr.]

Total U U(IV) U
Sample Material Area (per- (per- per.
cent) cent) cent-
Boulder Creek batholith

GP-17-..... Granodiorite...... Gold Hill, Colo-------....... 0.0069 0.0019 28
GP-19-..... Quartz monzonite- Magnolia, Colo-----...-- .0053 .0012 23
GP-34--.... Quartz monzonite- West of Boulder, Colo- .0078 .0019 24
GP-129.... Quartz diorite..... Tungsten, Colo-----...... 0049 .0012 24
GP-49 ...-- Granodiorite..--- South Beaver Creek, .0011 .00030 27
GP-50----- Quartz monzonite- West of Tremont .0048 .0011 23
Mountain, Colo.
GP-51 --..... Quartz monzonite- West of Tremont .0044 .0012 27
Mountain, Colo.
GP-63-..... Granite------ .. West of El Dorado .0023 .00024 10
Springs, Colo.
GP-61..... Quartz diorite..... Tungsten, Colo------... .0005 n.d.

"Silver Plume" dikes intrusive into Boulder Creek batholith

GP-28 --. Granite dike...... Gordon Gulch, Colo-. 0.00044 0.00014 32
GP-41.....----.do----------- Wallstreet, Colo------- .0022 .00027 12
GP-100---- .....do----------- Flagstaff Mountain, .0017 .00039 23

Idaho batholith

CPR-117-- Diorite----------...... Halley, Idaho.-------- 0.0059 0.0025 42
CPR-118.--..-do-----............ Horseshoe Bend, .0032 .0017 53
CPR-119-.......do-------- Quartzburg district, .0015 .00086 57
CPR-122_- Gnelssic granite -- Hamilton, Mont..-... .0013 .00086 66


Dur-1-..... Apatite-magne- Durango, Mexico----- 0.0010 0.0006 60
tite, deposit.
P-4-....... Quartz diorite-..... Southern California .012 .0057 47
batholith, Lake-
view Mountain,
P-22...... Gabbro-----...-----.. Henderson, N. ---... .00066 .00036 54
P-24...... Syenite---------- Renfrew, Ontario, .0023 .00052 23
53-BE-3a- Quartz monzonite- Shelby, N. C--------- .0061 .0032 52
53-BE-3b- Quartz monzonte Shelby, N. C ----------... 0065 .0041 63
Ko-1 -- Apatite massif--- Kola Penlnsula,USSR- .001 n. d.....
FK-3 --.... Apatite-magnetite Mineville, N. Y------ .079 n. d ......
G-25...... Shonkinte........ Mountain Pass, Calif. .0049 n. d. ..----

these analyses also show that the changes are not neces-
sarily progressive, as apatites from the more felsic
rocks are not more uraniferous than those from inter-
mediate rocks. The explanation may lie in the progres-
sive increase in the quantity of the other uraniferous
accessory minerals. Studies of E. S. Larsen, Jr. and
George Phair, and their colleagues confirm this sug-
gestion (oral communication, 1957).


It is noteworthy that the uranium content of the
apatites from the Boulder Creek batholith varies almost
directly as the uranium content of the host rocks.8
This is shown in figure 15. The rocks are slightly to
moderately different in composition and are from
localities as much as several miles apart. The fact that
most of the analyses conform so closely to a uniform
trend of concentration of uranium in apatite relative
to host rock indicates that the partition of uranium
between apatite and the crystallizing magma can main-
tain a constant ratio over a significant interval of the
differentiation span of the magma. That such equi-
librium does not necessarily prevail throughout differ-
entiation is indicated by the aberrant point on the
plot, GP-63.



40 -

Z 30




0 1 2 3 4 5
FIGURE 15.-Total uranium in rock vs. total uranium in apatite for
rocks from Boulder Creek batholith.

S We are indebted to George Phair for uranium determinations
of total rock for this comparison.

In the absence of analytical data the uranium in
igneous apatite has been reasonably presumed to be
entirely tetravalent (Tomkeieff, 1946; Goldschmidt,
1954; Larsen and Phair, 1954; McKelvey and others,
1955). The conclusion is virtually dictated by the
1. The common occurrence of uraninite in the same or re-
lated rocks;
2. The implications of uranium occurring largely in other
calcium and cerium-bearing minerals such as sphene,
monazite, xenotime, and allanite, a fact presumably
governed by the ability of U+' to substitute for the
Ce3" and Ca+2 ions;
3. The occurrence of cerium and thorium in igneous apatite;
4. Independent evidence of the effectiveness of reducing con-
ditions during the crystallization of the host rocks (Lar-
sen and Phair, 1954; Adams, 1954).
It is reassuring to demonstrate that U(IV) actually
occurs in igneous apatite in view of the geologic argu-
ments favoring this occurrence. Fewer data, gathered
earlier (Altschuler and others, 1954), showed that more
than half the uranium in some igneous apatite was
reduced. Appreciable U(IV) has been found in all
samples analyzed; however, with more comprehensive
sampling it is now seen that many apatites may have
more U(VI) than U(IV). It is not yet possible to
define the controls of the relative U(IV) and U(VI)
concentrations in igneous apatite. Many more analyses
are required, not only for determination of the valence
state of uranium in apatite, but also for other constitu-
ents of the parent rock and particularly for isotopes of
lead (see discussion p. 82). It is possible, however,
to discuss several alternative hypothesis on the basis
of the data presented in table 27.
Of paramount importance is the fact (noted earlier,
under sedimentary apatite) that consanguinous rocks
yield similar results when their apatites are analyzed
for U(IV). The virtual identity in the U(IV) /U per-
centages of all but one of the Boulder Creek samples
could not necessarily be anticipated unless apatites gen-
erally showed predominantly one or the other valence
state. Rocks from the Idaho batholith show slightly
more spread; the lithologies represented also differ
more. In attempting to explain the obtained distribu-
tions of U (IV) it is important to note that U (IV) per-
centages have been found to range from 10 to 66, and
it is possible that more analyses may substantially
widen the range.
It is conceivable that igneous apatite initially fixes
more U(IV) during its crystallization than it now
contains and that the increase in U(VI) is super-
imposed by normal oxidative weathering. However, the


lack of greater diversity in groups of related specimens
argues strongly against so simple a hypothesis. Thus
the total uranium content of the Boulder Creek apatites
range from 0.0011 to 0.0078 percent. In view of this
range and of the demonstrated possible range in U(IV)
(10 to 66 percent of the total), the regularity that
prevails in the U(IV)/U percentages of the Boulder
Creek apatites would be remarkable if it were caused
by so variable a process as weathering, in which local,
external factors (such as permeability and ground-
water composition) play so large a role. It is pertinent
to observe that the original rock samples and the apatite
separates were carefully screened for evidence of altera-
tion, and only fresh materials were used.
Two other hypothesis may be considered to account
for the consistent U (IV)/U(VI) ratios9 in groups of
apatites with much more diverse total uranium
1. The U (IV)/U (VI) ratios may be entirely a function of the
conditions prevailing during the crystallation of the
magma and be inherited unchanged.
2. The U(IV) /U (VI) ratios may be related to radioactivity
and to age. This second explanation does not preclude
the application of the first, as the percent of U(IV) may
have been initially determined by the prevalent equilibria
of the magma during crystallization and it may have
since been altered regularly by radioactivity.
Much evidence can be marshalled for the first of the
above viewpoints. Figure 15 demonstrates that the
uranium in the apatite is in equilibrium with the total
uranium in the rock and therefore, assuming the con-
fined crystallization of phanerocrystalline rocks, with
the magma from which these rocks crystallized. It
follows that the uniform U(IV) /U(VI) ratio in these
apatites varies directly with the total uranium in the
crystallizing magma. The progressive change in
uranium content with differentiation in a series of
related igneous rocks has been documented in the radio-
activity studies of Billings and Keevil (1946) and the
chemical studies of Larsen and others (1956). In addi-
,tion, Adams has shown that total uranium has a linear
relation to total K20 in the volcanic rocks of the Lassen
Peak region. In these rocks "potassium increases sys-
tematically during differentiation" (Adams, 1954).
Obviously uranium responds to the major chemical
equilibria that control differentiation. The above evi-
dence that trace percentages of U(IV) in apatite are
9 Previous sections were concerned with the amounts of the total
uranium that is tetravalent; hence the percentage U(IV)/U was used.
Here, relative quantities of U(IV) and U(VI) are discussed and their
ratio is used. Obviously the ratio and the percentage are based on the
Same figures and demonstrate the same trends.

directly related to uranium in the total rock (fig. 15)
reveals that the U(IV)/U(VI) ratio in apatite might
vary regularly with differentiation. Goldschmidt
(1954) pointed out the progressive development of
"magmatic oxidation" displayed by the increasing ratio
of ferric to ferrous iron in mafic to felsic rocks, and he
also noted that the iron oxide and sulfide mineral as-
semblages are compatible with the ferric-ferrous ratio.
It is reasonable to suppose that uranium in plutonic
apatite would similarly reflect the overall oxidation
capacity of the magma and that the U(IV)/U(VI)
ratio would change accordingly in a large series of
related rocks. Precise proof of this, however, must
await a much larger assemblage of data.
The hypothesis that the U(IV)/U(VI) ratio in
igneous apatite is a function of radioactivity and varies
regularly with age is more difficult to apply. Essen-
tially, it proposes that the U(IV)/U(VI) ratio is
governed by auto-oxidation. The concept of auto-
oxidation was first proposed by Ellsworth (1924, 1925)
who stated that the disintegration of UO02 or ThO2 to
Pb must automatically oxidize UO2, or other reduced
species present, in the following manner
UO2 Pb + 20
20 + UO2 + Pb -- UOs + PbO
Subsequent decay of the UOs thus produced will, in
turn, oxidize two more U02 (Holmes, 1948) as follows:
UO3 -+ Pb + 30
30 + 2UO2 + Pb -+ 2U03 + PbO
Thus, in net, one mole of U03 is produced for each
mole of radiogenic lead and, ideally, one might use
U(VI)'/U ratios to compute the ages of rocks and
minerals in substantially the same manner that lead-
uranium ratios are now used, as was proposed by
Lane (1934).
Three assumptions underlie the use of auto-oxidation
as an age method. First, all or most of the uranium
present in a mineral was originally tetravalent; second,
none of the U(VI) now present was produced by
weathering; third, uranium has not been differentially
leached or added to rock.
Tomkeieff (1946) found that UOs in uraninite cannot
be explained solely by auto-oxidation. In studying the
lead and UO3 contents of 52 collected analyses, he
demonstrated that natural uraninite varies greatly and
always contains more UOs than is demanded by the
auto-oxidation hypothesis.


These findings are not surprising in view of the ease
with which UO2 may be weathered. Indeed, Brooker
and Nuffield (1952) have found that surface oxidation
affects the peripheral zones of individual pitchblende
specimens enough to change their cell edge constants
relative to the material from the interior of the same
Despite the fact that pure UO2 is not known in
nature, it is possible that uraninite may crystallize
substantially free of UOa. If such material is un-
weathered and relatively free of water and Fe+2, Ce+3,
or other reduced species, it can be used to test the theory
that UOs is formd by auto-oxidation and that the
UOa/U ratio is an index of geologic age. Tomkeieff
(1956), by averaging analyses, found that auto-oxida-
tion might account for a significant part of the UO3
in uraninite. The demonstration by Holmes (1948)
that the Khito Island uraninites yield UO3 ages that
are in very good agreement with the lead-uranium ages
derived from the same specimens shows that auto-
oxidation may be the principal mechanism in producing
UOs3 in many minerals.
The uniformity of the UO3/U percentages of the
apatites from the Boulder Creek batholith and, to a
lesser extent, the Idaho batholith, plus the fact that
the apatite from the much younger Idaho batholith
rocks have much lower UO3/U percentages, suggest
that the U03 was produced by an internal process that
is a function of time, rather than superimposed by an
external process such as oxidative weathering.
It is difficult, however, to visualize the direct opera-
tion of auto-oxidation in producing' U03 in igneous
apatite in view of the widespread dispersion of uranium
in apatite (0.01 percent of uranium equals 1 atom of
uranium per 2400 unit cells of apatite) and the fact
that Fe and Ce in igneous apatite equal or exceed
We can evaluate auto-oxidation in apatite quite
simply by premising all initial uranium in apatite as
U(IV). On this basis the age of the apatite can be
determined by using the charts or the formula pub-
lished by Wickman (1939). As discussed above, we
may substitute the U(VI) /U ratio for the Pb206/Pb238
The U(VI)/U ratios of the Boulder Creek apatites
(excluding sample GP-63) yield ages between 4.0 and
4.5 billion years. Ages obtained by the ratio of lead to
alpha activity (Larsen and others, 1952) in zircon from
the same rock are close to 1 billion years (David Gott-
fried, U. S. Geological Survey, oral communication,
1956). This latter age demands a U(VI) content of
approximately 15 percent of the total U rather than
75 percent of it.

Ages based on U(VI)/U ratios of the Idaho batho-
lith apatites range from 2.4 to 3.6 billion years. Ages
from the lead-alpha activity ratios of zircon from the
same rock range from 60 to 110 million years (H. Jaffe,
U. S. Geological Survey, oral communication, 1957).
These would demand U(VI) contents of only 0.8 to 1.5
percent of the total U rather than 34 to 58 percent of it.
Despite the fact that the contribution of thorium has
not been taken into account in the above evaluations,
the results show so great an excess of U03 as to invali-
date the hypothesis that auto-oxidation produced the
UOs3, or even contributed significantly to it. It is pos-
sible, however, that radioactive disintegration can indi-
rectly facilitate and control the ratio of external oxida-
tion by solutions and gases present in the rocks and,
hence, produce an effect like auto-oxidation. Such con-
trol on crystal permeation by oxidizing agents could be
established by progressive increase in crystal damage
because of radioactive decay. Ellsworth (1925) pointed
out the following disruptive effects of radiation on
crystal structure:
1. Intracrystalline fracturing due to volume changes caused
by auto-oxidation of Ce, Fe, and U(IV),
2. Fracturing due to volume changes caused by radioactive
3. Structural dislocation caused by recoil forces in atoms
emitting alpha particles,
4. Dislocation caused by alpha particle bombardment.
The effectiveness of radiation damage in increasing
the cell dimensions of quartz and zircon has been dem-
onstrated by Hurley and Fairbairn (1953). In addi-
tion, oxygen liberated by the decomposition of struc-
tural or absorbed water as a result of alpha radiation
would effect further oxidation.
In summary, the relative U(IV) and U(VI) con-
tents of igneous apatites are most readily attributed to
oxidation conditions during the crystallization of the
host rocks. The consistency evident among some groups
of related rocks, however, indicates that initial UOs
contents may be progressively augmented by auto-
oxidation or external oxidation, at a systematic rate,
and it is of interest to investigate the relations of radio-
active decay to such contributions to UOs content. To
do so requires much more uranium data and a com-
parison of U(VI)/U ratios with radiogenic lead.

Uranium in apatite ranges from a few parts per
million (0.OOOX percent) in the insular phosphorites
to almost 1 percent in subaerially enriched deposits.


The most typical deposits, those comprising the greatest
concentrations of P205 in the lithosphere, are far more
restricted in range. Primary apatite, both igneous and
sedimentary, generally, has between 0.005 and 0.015
percent of uranium. Thus, igneous apatite seldom
exceeds 0.01 percent and primary apatite pellets from
the Hawthorn and Phosphoria formations typically
contain 0.005-0.008 percent of uranium; the rich
Moroccan phosphorites contain about 0.012 to 0.014
percent (with remarkable consistency).
The uranium in apatite may be leached or enriched,
and its enrichment may take place subaerially from
ground water or during submarine reworking. Marine
enrichment commonly enhances uranium content in
apatite to 0.02 to 0.03 percent, as shown in the Bone
Valley formation and the South Carolina terrace
gravels. The fact that unweathered marine phosphorite
seldom exceeds 0.03 percent of uranium, however, can
be attributed only to the infinitesimally small concen-
trations in the marine source, rather than to any limita-
tions of the apatite structure-or to the competition of
other ions for replacement sites. This is evident from
the much greater content of uranium in subaerially
enriched apatite, in which much additional uranium is
emplaced long after the normal suite of other trace
elements substituting for calcium. Thus pebbles in the
Bone Valley formation are supergene y enriched to
6.2 percent of uranilrum, aid-TossTi-bonel b6ntain as
iiucKias 0.83 percent of uriiim. ^Sigiincantty, tfe
U(IV) content in each of the above is more than half
of the total uranium.
The characteristics of geologically young marine
phosphorites enlighten us both as to the capacity and
the nature of marine sources of uranium. Phosphatic
marls from the Gulf of Mexico and phosphorite nodules
from the Pacific Ocean have a U/P205 ratio as high or
highIer than those of any of the world's major phos-
pate deposits that have been investigated, thus proving
that ocean water is an adequate source of uranium ini
typical-marine phosphorite.

The suggestion that uranium in phosphorites is domi-
nantly U(VI) fixed as a surface-coordinated uranyl
pyrophosphate has been advanced (Bowie and Atkin,
1956) to account for the lack of uranium in calcareous
sediments. However, the well-established ability of
carbonate to complex uranyl ion and thus increase the
solubility of uranium (Bachelet and others, 1952) more
probably accounts for the relative lack of uranium in
carbonate rocks. Furthermore, the surface-limited reac-
tions of uranyl ion with bone apatite demonstrated by

Neuman and others (1949a, b) are structurally much
more limited that the isomorphous replacement of Ca+2
by U+4, and it is not surprising that an apatite mate-
rial with as much as 0.83 percent total uranium should
contain more than half the uranium as U(IV). Some
of the U(VI) in unweathered phosphorites is probably
chemisorbed by surface reaction of uranyl on the grow-
ing crystallites of sedimentary apatite. Such reactions
are, in nature, more likely to involve orthophosphate
than pyrophosphate formation. In addition, the high
U(IV) content of young phosphorites suggests that
postdepositional oxidation has also produced appre-
ciable U(VI) in the older rocks.
Were mere adventitious extraction the sole factor in
uranium fixation by precipitating apatite, the ratio of
uranium to phosphate in apatite might approximate
their relative concentrations in the sea. Uranium in sea
water is 1-2 parts per billion. P20s in sea water of the
Continental Shelf is 50 mg/m3 (Kazakov 1937). The
U/P20O ratio in sea water is thus 0.02-0.04 and thereby
exceeds by far the ratio of U/P205 found in primary
marine phosphorites (table 12). In addition, compari-
son of U/P205 ratios both within and among the pri-
mary pellets and nodules from the Hawthorn forma-
tion, Florida, the Gulf of Mexico, the Pacific Ocean,
and French Morocco (table 12), reveals considerable
variety for comparable materials. Thus, direct and
indiscriminate fixation of uranium that is available
fails to explain the known uranium contents in apatite.
Nor can these known contents be explained merely by
citing an "affinity" of uranium for phosphate that is
demonstrated by the many uranium phosphate minerals,
as these are entirely uranyl minerals, and we have found
that in relatively young, primary phosphorite most of
the uranium is present in the tetravalent or, uranous,
It seems, indicated from the above, that the uranium
actually available to apatite must be limited and largely
contingent on special circumstances, presumably, the
production of U(IV). Garrels (1955) and Sheldon
(written communication, 1957) have shown from ther-
modynamic considerations of the isolated couple
(U02) +2 and U+4 that the uranyl concentration of
normal sea water would be many orders of magnitude
greater than that of the uranous ion. It is only in an
acid reducing environment that the U(IV) concentra-
tion becomes significant (Garrels, 1955). Eveiiun-ir
these conditions (pH = 6, Eh = -0.3), the
(UO2)+2/U+4 ratio is about 100 (Sheldon, written
communication, 1950); however, apatite cannot be con-
sidered to form from such water of relatively high


It follows from the insolubility of U (IV) in ground
water solutions that uranium is delivered to the ocean
as (U02)+2. However, uranyl ion may be readily
reduced at low temperature and pressure to U(IV) and
as the concentrations of ferrous iron and organic matter
in marine phosphorites is generally far greater than
the concentration of uranium, the necessary marine
reduction of U(VI) prior to apatite fixation may be
readily presumed.
The substantial ground-water and marine enhance-
ments of uranium in apatite reveal the potency of
apatite to remove uranium selectively from terrestrial
or marine solutions. U(IV) produced by reduction
reaction between (U02) +2 and Fe+2, or organic matter,
would be taken up as a proxy for calcium during for-
mation of apatite, or by ion exchange of calcium during
marine reworking of previously formed apatite.
Unweathered pebbles from the Bone Valley forma-
tion and nodules of Quaternary age from the Pacific
Ocean contain dominantly tetravalent uranium. This
indicates that the hexavalent uranium in sea water is,
in fact, reduced prior to emplacement in apatite.
Further proof of this is obtained from the demonstra-
tion that new increments of marine uranium in the
transgressively reworked Pleistocene terrace gravels
near Charleston, South Carolina are mainly U(IV).
In contrast to these findings, the marine phosphorites
containing prominent U(VI) have been obviously
weathered or are geologically much older (and prob-
ably weathered). It is therefore believed that the
U(VI) present in marine phospior-ite-si-derived
largely from postdepositional oxidation of U(IV).
In contrast to igneous apatite the occurrence of
uranium in marine sedimentary apatite is not governed
solely by the equilibrium conditions prevailing during
the precipitation of the apatite. This follows from the
evidence of secondary enrichment during marine re-
working; namely, the host mineral acquiring new incre-
ments of the trace element by re-exposure to the same
or similar solutions. It is also indicated by the wide
variation in the U/P205 ratios in unaltered primary,
or first-cycle, phosphorites, a fact which would other-
wise demand considerable variation in the uranium
content of the marine sources, for which there is no
evidence. It is proposed that apatite actively affects
the course of the reduction reactions which might nor-
mally produce U(IV) in the following manner:
(U02) +2 + Reductant U+4 + Oxidized reductant
The presence of apatite as a precipitating or solid
phase which can sequester the U+4 by solid solution or

ion exchange for calcium, interferes with the attain-
ment of equilibrium in the above reaction and drives it
instead in the direction:
(U02) +2 + Reductant -- U+4 + Oxidized reductant

Thus by removal of U(IV) in an insoluble phase,
apatite promotes an additional reduction of uranyl and
thereby insures a continuing supply of U(IV) for its
own uptake. The removal of U (IV) from sea water is
thus not dependent on continuous precipitation of
apatite, as uptake may proceed equally by ion-exchange
reaction with previously formed apatite. This is indi-
cated by the higher total U and U (IV) contents of
reworked phosphate nodules.
The name, regenerative capture, is proposed to
describe this type of trace element concentration in
which a stable complex or insoluble phase, by effec-
tively removing an ion species that is normally present
in insignificant quantities, prevents the attainment of
equilibrium in the oxidation-reduction reaction pro-
ducing this ion, and causes continuous production of
the ion. Thus, as long as removal continues, the supply
is maintained and a mineral is able to fix unusual con-
centrations of an ion species which, from mere stability
or concentration considerations, would not be expected
in large quantities. Regenerative capture can be main-
tained only as long as the host mineral remains un-
saturated with respect to the trace constituent. The
fact that apatite can concentrate as much as 0.5 per-
cent of U (IV) proves that saturation conditions are
seldom realized in typical marine apatite (0.01-0.02
percent of U).
The idea of regenerative capture clearly invokes no
new chemical principle. Its merit lies in explaining
unexpectedly large mineral concentrations of ions pres-
ent in insignificant concentrations in solution. Regen-
erative capture applies to uranium in apatite in two
important ways. It helps to explain the high U(IV)/
U(VI) ratios of some marine phosphorites despite the
fact that uranium in the ocean source most probably
occurs predominantly as U(VI). It implies that a
strongly reducing environment is not a requirement for
the fixation of J(IV) by apatite-but rather that appre-
cible 6UT(IV) may be fixed by aptatite so loniig as any
reduction of U(VI) is possible.

A completely different situation prevails in igneous
apatite. The presence of many other good hosts makes
the entire distribution of uranium much more complex
and the amount of uranium available to apatite is regu-


lated not only by its inherent capacity but by the
capacities of other hosts coextensive with it and thus
reflects the total equilibrium in the magma. This is
clearly evident from the systematic variation of both
the total uranium and the U(IV)/U(VI) ratio in
apatite with the total uranium in the rock, for suites
of related rocks.


Adams, J. A. S., 1954, Uranium and thorium contents of vol-
canic rocks, in Faul, Henry, ed., Nuclear geology, p. 89-98:
New York, John Wiley and Sons.
Alexander, Frank, 1953, Guidebook, field trip to central Ten-
nessee phosphate district-Southeastern Mineral Sym-
posium, Geol. Soc. America: Nashville, Vanderbilt Univ.,
Tennessee Div. Geology, 8 p.
Altschuler, Z. S., Cisney, F. A., and Barlow, I. H., 1953, X-ray
evidence of the nature of carbonate-apatite (abs.) : 19th
Internat. Geol. Cong., Algiers, Comptes rendus, sec. 11,
fasc. 11, p. 9.
Altschuler, Z. S., Clarke, R. S., and Young, E. J., 1954, Uranium
in apatite (abs.) : Geol. Soc. America Bull., v. 65, no. 12,
pt. 2, p. 1225-1266.
Altschuler, Z. S., Jaffe, E. B., and Cuttitta, Frank, 1956, The
aluminum phosphate zone of the Bone Valley formation,
Florida, and its uranium deposits: Proc. Internat. Conf.
on Peaceful Uses of Atomic Energy, Geneva, 1955, v. 6,
p. 507-513; U. S. Geol. Survey Prof. Paper 300, p. 495-504.
Arambourg, Camille, and Orcel, Jean, 1951, Observations pr6-
liminaires sur a presence d'un vanadate d'urane dans les
gisements de phosphate du Maroc: Acad. Sci. Paris Comptes
rendus, v. 233, no. 25, p. 1635-1636.
Bachelet, M., Cheylan, E., Douis, M., and Goulette, J. C., 1952,
Preparation et propri6t6s des uranylcarbonates; 2* note:
Uranylcarbonates alcino-terreux: Soc. chim. France Bull.,
p. 565-569.
Baes, C. F., Jr., 1954, The production of U" by ferrous ion in
phosphoric acid solution; the formal electrode potential of
the U*--U+6 couple: U. S. Atomic Energy Comm. ORNL-
1581 or AECD-3594. (Microcard copy on file in U. S.
Atomic Energy Comm. depository libraries.)
TBarr, J. R., Jr., 1955, Uranium production from phosphate rock:
Remarks at 39th Ann. Mtg. Florida Eng. Soc., Daytona
Beach, Florida, U. S. Atomic Energy Comm.
RBeevers, C. A., and McIntyre, D. B., 1946, The atomic structure
of fluorapatite and its relation to tooth and bone material:
Mineralog. Mag., v. 27, no. 194, p. 254-257.
Billings, M. P., and Keevil, N. B., 1946, Petrography and radio-
. activity of four Paleozoic magma series in New Hampshire:
Geol. Soc. America Bull., v. 57, p. 797-828.
orneman-Starinkevitch, I. D., 1938, On some isomorphic sub-
I; stitutions in apatite: Acad. Sci. U.R.S.S., Comptes rendus
(Doklady), v. 14, no. 4, p. 253-255.
wie, S. H. U., and Atkin, D., 1956, An unusually radioactive
t fossil fish from Thurso, Scotland: Nature, v. 177, no. 4506,
p. 487-488.
ker, E. J., and Nuffield, E. W., 1952, Studies of radioactive
'compounds, IV-Pitchlende from Lake Athabaska, Canada:
Am. Mineralogist, v. 37, p. 363-385.

Capdecomme, Laurent, 1953 Etude mineralogique des gites de-
phosphates alumineux de la region de Thies (S6n6gal) : In-
ternat. Geol. Cong., 19th, Algiers, Comptes rendus, sec. 11,
fase. 11, p. 103-117.
Capdecomme, Laurent, and Pulou, Rene, 1954, Sur la radio-
activitd des phosphate de la region de Thies (Sendgal):
Acad. Sci. Paris Comptes Rendus, v. 239, p. 288-290.
Carnot, Adolphe, 1893, Recherches sur la composition g6n6rale et
la teneur en fluor des os modernes et des os fossiles des
diff6rents ages: Annales des mines, ser. 9, v. 3, p. 155-195.
Cathcart, J. B., 1950, Notes on the land-pebble phosphate de-
posits of Florida, in Snyder, F. G., ed., Symposium on min-
eral resources of the southeastern United States, 1949 pro-
ceedings: Knoxville, Tenn. Univ. Press, p. 131-151.
-- 1956, Distribution and occurrence of uranium in the
calcium phosphate zone of the land-pebble phosphate dis-
trict of Florida: Proc. Internat. Conf. on the Peaceful Uses
of Atomic Energy, Geneva, 1955, v. 6, p. 514-519; U. S. Geol.
Survey Prof. Paper 30, p. 489-494.
Cathcart, J. B., Blade, L. V., Davidson, D. F., and Ketner, K. B.,
1953, The geology of the Florida land-pebble phosphate de-
posits: 19th Internat. Geol. Cong., Algiers, Comptes rendus,
sec. 11, fasc. 11, p. 77-91.
Clarke, R. S., Jr., and Altschuler, Z. S., 1958, Determination of
the oxidation state of uranium in apatite and phosphorite
deposits: Geochimica et Cosmochimica Acta, v. 13, nos. 2/3,
p. 127-142
Connick, R. E., and Hugus, Z. Z., The participation of f orbitals
in bonding in uranium and transuranium elements: Am.
Chem. Soc. Jour., v. 74, p. 6012-6015.
Davidson, C. F., and Atkin, D., 1953. On the occurrence of
uranium in phosphate rock: 19th Internat. Geol. Cong.,
Algiers, Comptes rendus, sec. 11, fasc. 11, p. 13-31.
Denson, N. M., and Gill, J. R., 1956, Uranium-bearing lignite
and its relation to volcanic tuffs in eastern Montana and
North and South Dakota: Proc. Internat. Conf. on the
Peaceful Uses of Atomic Energy, Geneva, 1955, v. 6, p. 464-
467; U. S. Geol. Survey Prof. Paper 300, p. 413-418.
%Dietz, R. S., Emery, K. 0., and Shepard, F. P., 1942, Phosphorite
deposits on the sea floor off southern California: Geol. Soc.
America Bull., v. 53, p. 815-848.
Du Toit, A. L., 1917, The phosphates of Suldanha Bay: South
African Geol. Survey Mem. 10, 34 p.
Ellis, D. A., 1952, The recovery of uranium from industrial
phosphoric acids by solvent extraction: U. S. Atomic Energy
Comm., DOW-81, 79 p.
Ellsworth, H. V., 1924, Atomic disintegration as a measure of
geologic time: Pan. Am. Geologist, v. 42, p. 273-280.
1925, Radioactive minerals as geologic age indicators:
Am. Jour. Sci., 5th ser., v. 9, p. 127-144.,
Fairchild, J. G., 1929, Base exchange in artificial autunites:
Am. Mineralogist, v. 14, p. 265-275.
Garrels, R. M., 1955, Some thermodynamic relations among the
uranium oxides and their relation to the oxidation states of
the uranium ores of the Colorado Plateaus: Am. Mineralo-
gist, v. 40, nos. 11 and 12, p. 1004-1021.
Goldschmidt, V. M. (by Alex Muir, ed.), 1954, Geochemistry:
Oxford, Clarendon Press, 730 p.
Graton, L. C., and Fraser, H. J., 1935, Systematic packing of
spheres with particular relation to porosity and permeabil-
ity: Am. Assoc. Petroeum Geologists Bull., v. 43, no. 8, p.


Green, Jack, 1953, Geochemical table of the elements for 1953:
Geol. Soc. America Bull., v. 64, p. 1001-1012.
*Grimaldi, F. S., May, Irving, and Fletcher, M. H., 1952, U. S.
Geological Survey fluorimetric methods of uranium analy.
sis: U. S. Geol. Survey Circ. 199, 20 p.
Grimaldi, F. S., May, Irving, Fletcher, M. H., and Titcomb,
Jane (Compilers), 1954, Collected papers on methods of
analysis for uranium and thorium: U. S. Geol. Survey
Bull. 1006, 184 p.
Gruner, J. W., 1952, New data on syntheses of uranium minerals.
-Part I. Annual Report for July 1, 1951 to June 30, 1952:
U. S. Atomic Energy Comm. RMO-983, 26 p.
Gruner, J. W., and McConnell, Duncan, 1937, The problem
of the carbonate-apatites. The structure of francolite:
Zeitschr. Kristallographie, v. 97, p. 208-215.
Guntz, A. A., 1952, Sur la presence d'uranium dans les
phosphates Nord-Africains: Acad. Sci. Paris Comptes
Rendus, v. 234, p. 868-870.
Hayes, C. W., and Ulrich, E. 0., 1903, Description of the Colum-
bia quadrangle, Tennessee U. S. Geol. Survey. Geol. Atlas,
Columbia folio, 95.
Hebert, Claude, 1947, Contribution a 1'6tude de la chimie des
phosphates de calcium: Annales des mines, Mem., v. 136,
no. 4, 93 p.
Hendricks, S. B., and Hill, W. L., 1950, The nature of bone and
phosphate rock: Natl. Acad. Sci. Proc., v. 36, p. 731-737.
Holmes, Arthur, 1948, The oldest known minerals and rocks:
Edinburgh Geol. Soc. Trans., v. 14, pt. 2, p. 176-194.
Hurley, P. M., and Fairbairn, H. W., 1953, Radiation damage in
zircon-a possible age method: Geol. Soc. America Bull., v.
64, p. 659-673.
Igelsrud, Iver, Chocholak, John, Stephan, E. F., and others, 1949,
Recovery of uranium from phosphate rock by the Battelle
monocalcium phosphate process: U. S. Atomic Energy
Comm. BMI-JDS-201, 99 p.
Igelsrud, Iver, Stephan, E. F., Chocholak, John, and others,
1948: Chemical process to recover uranium from phosphate
rock: U. S. Atomic Energy Comm., BMI-JDS-126, 29 p.
Jacob, K. D., Hill, W. L., Marshall, H. L., and Reynolds, D. S.,
1933, The composition and distribution of phosphate rock
with special reference to the United States: U. S. Dept.
Agriculture Tech. Bull. 364, 90 p.
Jacob, K. D., and Hill, W. L., 1953, Laboratory evaluation of
phosphate fertilizers in Pierre, W. H., and Norman, A. G.,
ed., Soil and fertilizer phosphorus in crop nutrition, v. 4,
chap. 11, p. 299-345, of Agronomy, [a series of monographs
prepared under the auspices of the American Society of
Agronomy] : New York, Academic Press.
* Jaffe, E. B., and Sherwood, A. M., 1951, Physical and chemical
comparison of modern and fossil tooth and bone material:
U. S. Geol. Survey TEM-149, 19 p., issued by U. S. Atomic
Energy Comm. Tech. Inf. Service Extension, Oak Ridge,
Kazakov, A. V., 1937, The phosphorite faces and the genesis of
phosphorites in Geological investigations of agricultural
ores USSR: Sci. Inst. Fertilizers and Insectofungicides
Trans. (USSR), no. 142, p. 95-113. (Special issue in
English published for 17th Internat. Geol. Cong.)
Lane, A. C., 1934, Fivefold check of uraninite age?: Am.
Mineralogist, v. 19, no. 1, p. 1-13.

Larsen, E. S., Jr., Keevil, N. B., and Harrison, H. C., 1952,
Method for determining the age of igneous rocks using the
accessory minerals: Geol. Soc. America Bull., v. 63, p. 1045-
Larsen, E. S., Jr., and Phair, George, 1954, The distribution of
uranium and thorium in igneous rocks, in Faul, Henry, ed.,
Nuclear geology, p. 75-89: New York, John Wiley and Sons.
Larsen, E. S., Jr., Phair, George, Gottfried, David, and Smith,
W. L., 1956, Uranium in magmatic differentiation: Proc.
Internat. Conf. on Peaceful Uses of Atomic Energy, Geneva,
1955, v. 6, p. 240-247; U. S. Geol. Survey Prof. Paper 300,
p. 65-74.
LeCornec, J., 1951, Estimate of world supplies of the principal
plant nutrients by cost range: U. N. Sci. Conf. Conserv. and
Utiliz. Res. Proc., v. 2, Mineral resources, p. 270-274.
Lenoble, Andre, Salcan, Henri, and Ziegler, Valery, 1952, De-
couverte de l'uranium dans les niveaux phosphates du
Maroc: Acad. Sci. Paris Comptes rendus, v. 234; no. 9, p.
'McConnell, Duncan, 1938, A structural investigation of iso-
morphism of the apatite group: Am. Mineralogist ,v. 23, no.
1, p. 1-19.
McKelvey, V. E., and Carswell, L. D., 1956, Uranium in Phos-
phoria formation: Proc. Internat. Conf. in the Peaceful
Uses of Atomic Energy, Geneva, 1955, v. 6, p. 503-506; U. S.
Geol. Survey Prof. Paper 300, p. 483-487.
McKelvey, V. E., Cathcart, J. B., Altschuler, Z. S., and others,
1953, Domestic phosphate deposits, in Pierre, W. H., and
Norman, A. G., eds., Soil and fertilizer phosphorus in crop
nutrition, v. 4, chap. 11, p. 347-376, of Agronomy, a series
of monographs, prepared under the auspices American
Society of Agronomy: New York, Academic Press.
a McKelvey, V. E., Everhart, D. L., and Garrels, R. M., 1955,
Origin of uranium deposits: Econ. Geology, 50th Anni-
versary volume, 1905-1955, pt. 1, p. 464-533.
a McKelvey, V. E., and Nelson, J. M., 1950, Characteristics of
marine uranium-bearing sedimentary rocks: Econ. Geology,
v. 45, p. 35-53.
McKeown, F. A., and Klemic, Harry, 1956, Rare-earth-bearing
apatite at Mineville, Essex County, New York: U. S. Geol.
Survey Bull. 1046-B, p. 9-23.
MacNeil, F. S., 1947, Correlation chart for the outcropping Ter-
tiary formations of the eastern Gulf region: U. S. GeoL
Survey Oil and Gas Inv. Prelim. Chart 29, with text.
Malde, H. E., Geology of the Charleston phosphate area, South
Carolina: U. S. Geol. Survey Bull. (in presss.
Miller, L. J., and Kerr, P. F., 1954, Progress report on the
chemical environment of pitchblende in Annual Report for
June 30, 1953 to April 1, 1954, pt. 2: U. S. Atomic Energy
Comm. RME-3096, 98 p.; see p. 72-92.
Moore, G. W., 1954, Extraction of uranium from aqueous solu-
tion by coal and other materials: Econ. Geology, v. 49, P.
, Neuman, M. W., and Neuman, W. F., 1948, The deposition of
uranium in bone, II-Radioautograph studies: Jour. Biol.
Chemistry, v. 175, no. 2, p. 711-714.
*Neuman, W. F., Neuman, M. W., Main, E. R., and Mulryan,
B. J., 1949a, The deposition of uranium in bone, V-Ion
exchange studies: Jour. Biol. Chemistry, v. 179, p. 335-340.
1949b, The deposition of uranium in bone, VI--lo0
competition studies: Jour. Biol. Chemistry, v. 179, P.


Oakley, K. P., 1955a, Analytical methods of dating bones, in
Techniques in Archaeology: Advancement of Science, v. 11,
no. 45; p. 3-8.
.---- 1955b, The composition of the Piltdown hominoid re-
mains: British Mus. Nat. History Bull., Geology, v. 2, no. 6,
p. 254-265.
palache, Charles, Berman, Harry, and Frondel, Clifford, 1951,
Dana's system of mineralogy, v. 2, 7th ed.: New York, John
Wiley and Sons, 1224 p.
rpannell, J. H., 1950, Aid oxidation of uranous solutions: U. S.
Atomic Energy Comm. MITG-261, 18 p.
PiLnkowski, S., 1953, Radioactive d'un group de phosphorites
polonais I: Acad. Polonaise Sci. Bull., C1. III, v. 1, p. 39-41.
piggot, C. S., and Urry, W. D., 1941, Radioactive relations in
ocean water and bottom sediments: Am. Jour. Sci., v. 239,
p. 81-91.
Rogers, G. S., 1913, The phosphate deposits of South Carolina:
U. S. Geol. Survey Bull. 580-J, p. 183-220. [1915].
Runnels, R. T., Schleicher, J. A., and Van Nortwick, H. S.,
1953, Composition of some uranium-bearing phosphate
nodules from Kansas shales: Kansas Geol. Survey Bull. 102,
pt. 3, p. 93-104.
Rusakov, V. P., 1933, 0 soderzhanii radiya i toriya v fosforitakh
[The radium and thorium content of phosphorites] : Akad.
Nauk SSSR, Leningrad, Doklady (Compte rendus) ser. A,
no. 3, p. 25-33.
SW., 1924, The phosphate deposits of Abeokuta Province:
Nigeria Geol. Survey Bull. 7, p. 9-38.
eyer, J. M., 1954, The chemistry of uranium (IV) ortho-
phosphate solutions. Part, I. The solubility of uranium
(IV) orthaphosphates in phosphoric acid solutions: U. S.
Atomic Energy Comm. ORNL-1747, 34 p.

I Silverman, S. R., Fuyat, R. K., and Weiser, Jeanne D., 1952,
Quantitative determination of calcite associated with carbo-
nate-bearing apatites: Am. Mineralogists, v. 37, p. 211-222.
Smith, K. G., and Bradley, D. A., 1952, Radioactive dinosaur
bones from the Camp Davis region, western Wyoming:
Michigan Acad. Sci. Papers 1951, v. 37, p. 257-263.
Strutt, R. J., 1906, On the distribution of radium in the earth's
crust: Royal Soc. London Proc. Ser. A, v. 78, p. 150-153.
1908, On the accumulation of helium in geologic time:
Royal Soc. London Proc. Ser. A, v. 81, p. 272-277.
Thompson, M. E., 1953, Distribution of uranium in rich phos-
phate beds of the Phosphoria formation: U. S. Geol. Survey
Bull. 988-D, p. 45-67.
1954, Further studies of the distribution of uranium In
rich phosphate beds of the Phosphoria formation: U. S.
Geol. Survey Bull. 1009-D, p. 107-123.
( Tomkeieff, S. I., 1946, The geochemistry of uranium: Sci. Prog-
ress, v. 34, no. 136, p. 696-712.
Volodchenkova, A. I., and Melentiev, B. N., 1943, Apatites of two
textural types from apatite-nepheline rocks of Chibiny:
Acad. Sci. U.R.S.S., Comptes rendus (Doklady) v. 39, no. 1,
p. 34-35.
Weiner, J. S., Oakley, K. P., and Le Gros Clark, W. E., 1953,
The solution of the Piltdown problem: British Museum
(Natural History) Bull., Geology, v. 2, no. 3, p. 141-146.
Wells, A. F., 1950, Structural inorganic chemistry, 2d ed., 727
p.; Oxford, Clarendon Press.
Wickman, Frans E., 1939, Some graphs on the calculation of
geological age: Stockholm, Sveriges geol. Undersakning,
ser. C, no. 427, Arsbok 33, no. 7, 8 p.


Abstract ..-..................----------------- ------- ---------------- 45
Acknowledgments .....................--------------------...--------------------------- 47
Akin mine, Tennessee............--------------------------------------------- 65
Aluminum phosphate zone, distribution of chemical elements-------------............... 60-61
field relations ......--....--------------------------------------------------- 57-58
origin..........................---------------------------------------------------- 57,62
petrography.... ............. ..------------------------------------------------ 6-61
Aluminum phosphates-------...............----------...----......------------------- 57-61,65
Analyses, chemical..................... ..........------------------------------- 49,51,56,62,63,65,69,77
methods of .................----------------------------------------------- 51,69-71
Angaur Island, Pacific Ocean ---------------------------------------- 51
Apatite, complete chemical analyses-- ------------------------------ 49
mineralogy................------------..................----------------....----------------- 47-50
spectrographic analyses -------.............----------------------------------- 49
structure of---...............-------- --------------- ----------------- 47-48,74
uranium content....................-------------------------------------------- 50-57
Ashley River .......................--------------------------------------------------- 62
Ashley River flood plain....................................------------------------------------------ 63
Autoradiograph .....---........-------------.....................------------------------------ 66
Auto-oxidation-----------..... .........---------------- -------------------.. 81-82
Autunite................------------------------------------------------------ 52,74

Bigby-Cannon limestone..............------------..-------------------------------. 65
Bigby formation.............---------- ----------------------------------- 65,78
Bone Valley------................----- ------------ --------------------------- 62
uranium content in fossils from ------------------------------------ 67
Bone Valley formation.............................--------------------------------- 52,53, 54,57,67
stratigraphy................---------..............-----------------....------------------- 57-58
weathering........................------------------------------------------------ 57-60
Bones, uranium content ---..................---------...----------------------------- 66
Bonny Lake mine, Florida --------------..................-------------------------- 69,72
Bonny Lake pebbles -------------------- -------------------- 73
Boulder Creek batholith .-- -.------------------------........ 51,80,82
Boyette mine, Florida............---------.....---------------------------------- 53
Brown-rock phosphate...--................------------------------ 65,72,78

Calcite-..................--- ---------------------------------------------------62,64
Calcium, leaching and replacement of ---------.. -------------------------..... 60
substitution for....----------------------------------------- 47,52,62,69
Calcium aluminum phosphates-----.............---------------------------------- 57-60
Carbonate-apatite---.........--------------------------------------------... 48-49
Cave phosphates, partial analyses ............----------------------------------- 51-52
Cerium, in apatite ................------------..----------------------------------- 69,80
oxidation or reduction of uranium by----.........-----------------........-------..... 69,82
Cerium phosphates---..................----------..------------------------------- 47
Oloverly formation, uranium content of bones------......... -------------------- 66
Conclusions .....-----...........---------------------.....................--------------------------........ 82-85
Cooper marl..............................----------------------...--------------------- 62,63,64,65
Coupled diodochy, in phosphate minerals ---------........................---------------........---- 47-48
randallite..........-------.--.........------------------------------------------- 57, 60

ZBnrlchment, factors for PsOs and U---.........--- -------------------------- 63, 65
marine-.................................--------------------------------------------------- 67-68
postdepositional...................-------------..----------------------------- 59
residual.......................................------------------------------------------------------. 62
supergene ---...............-------........------..............----------..........------------------ 62
uranium................-----------------------------------------------------.. 68
lulvalent uranium..----------...................------------------------------------- 50

rapatite.........-----.......-------...........---..---------------.........------------... 48, 49, 74
ce, in apatite nodules-.............----------..------------------------- 53
rine,excess in apatite.......-----------------------------------......... 48-49, 75, 77
in apatite structure-....---------------------- --- -------------- 47, 75, 77
v-eplacement of OH.-------------------------------------------- 47,49, 51
Union formation, uranium content ------..................-------------------------... 66
ont County, Wyo., apatite deposit.........-------------------------------... 68

term defined......................---------------.....----------------------------.. 49
erivedphosphorites------ ------------------------------------- 51
SCalifornia---------------------...........................................------------------------- 5
of Mexico, phosphate and uranium contents of bottom samples--------......... 56

Hawthorn formation-----.............---- ------- ----------------------- 55, 56, 67, 83
Hells Creek formation, uranium content--...........------------------------------ 66

Idaho batholith -------.................-------------- -------------........ 79
age of apatites of...............--------- ------------------------------------- 82
Introduction.........................---------------------------------------------- 45-47
Island phosphorite-------.....................------------------------------------------ 51

Khito Island, age of uraninites of...------.......------------------------------- 82
Khouribga, French Morocco---.........---------------------------------- 54, 56

Land Pebble field, Florida -------------....... ---------------- 52, 54, 56, 57, 58
Lateritic weathering---.......-- --------------------------------------------- 59
Leaching..............................------------------------------------------------------ 59, 62
Lignites, uraniferous -...........---.............-----------------------------------------...... 66

Magmatic oxidation---....-------------------------------------------- 81
Marine phosphate deposits ----..............---------------------------49,52
Marl ..............---------------------------------------------------- 62, 64, 83
Millisite................---------- ---------------------- --------------------- 57, 60
Mona Island, Puerto Rico, analyses of cave phosphorite....-----------------51-52

Nodule, term defined-.......---------------------------------------------. 49
Nodules, phosphorite, from Pacific Ocean---..........------ ------------------- 73, 83
Nuclear emulsion studies----------.................---------------------------------... 68

Old Red sandstone....................----------------------------------------------- 66
Oolite, term defined..................------------..----------------------------------- 49
Organic content, correlation with uranium ----------..........------------------ 54-55
relationship to reducing conditions -----------.... -----------------63, 72, 84
Oulad-Abdoun, French Morocco, phosphate beds ----------- 53-54

Peace Valley mine, Florida----............-------------------------------------..... 69
Pebble apatite-----...............----- ----------------- ---------------------- 67, 68
Phosphate deposits, Florida.....--..-----...------------------....... 51, 53, 56, 57, 65, 68
French Morocco--.--........---------- -------- ----------- 53, 54, 56, 67, 68
Tennessee---......----------------------------------------------------- 78
terrace gravel near Charleston, S. C-----------.........................---------------------..... 77
Phosphate fertilizer, production in United States -------.......------------------46
Phosphate rock, term defined-----..............-----------------------------------... 50
Phosphoria formation......................----------------------------------------- 53,54,78
Phosphorite, spectrographic analyses -------.....................---------------------------- 64
term defined ...........----------------------- -------------------------- 50
Phosphorite pebble, term defined--..........---- ------------------------------- 50
Photomicrograph.....................------------------------------------------------- 68
Polish phosphorite--......------ ------------------------------------- 55
Purpose of report..--.......... ---- ------------------------- ----- 46

Radioactive decay-----...........-- ------------ --------------------- 82
Rantowles Creek--.................----------.....------------------------------- 62
Recovery methods.....................---------------------------------------------- 69-71
References, selected--.........-------------------- ------------------------ 85-87
Replacement, subaerial...---...............------------------------------------- 51
Reserves, PsOs, apatite, and U ---.......----------- ------------------------- 46
Reservoir Mountain-----..................----------------------------------------.... 55
Reworking, marine......................-------------------------------------- 57,67-68,71,83
subaerial----.......------------------------ -- -----------------... 59

Sedimentary apatite--....................-------------------------------------------. 57
Sedimentary carbonate-apatites, composition.--------------..........------------------ 48-49
Senonian Moroccan apatite ------.- ------------------------------------- 55
Spectrographic analyses-------...........------------ ----------------------- 49,64

Terminology, petrographic ............--------------------------------------- 49-50
Thorium, content in Florida pebble phosphorite --------------------------50
in igneous rocks and minerals....----------------------------------- 47,80
source of radioactivity in fossils..---.....------------------------------- 66
Titanium.......---- --......-------------- -------------------------------.... 69


Uraninite, oxides and auto-oxidation in---.........-- ------- --------------- 81-82
Uranium, as byproduct of phosphate production ..........................---------------------- 46
correlation with other elements......---------------- ---------------- 54-55
crystallo-chemlcal nature in apatite.....-- ....-- ------------------------- 68
distribution in phosphate pebbles--............----------------------------........... 60
experimental oxidation and reduction of---.........-- ------------------ 70-71
form and manner of occurrence in phosphorites ---.....--------------- 52-55
in Igneous apatite----..........--------------------------------------- 79-82
in sea water.........---- ----------------- -------------------- 8384
marine sources of .........------------------------------------------ 55-56,83-84
marine versus ground-water origin in phosphorite----------------- 55-57
oxidation or reduction by other elements----.....--.....--------- 69-71,73,82,84
oxidation state in apatite-...........------------------------------------- 69-74
reserves in phosphorite..................--------- ------ --------------------- 46
secondary emplacement in bones ......-- ------------ ------------ 66-67
substitution for calcium in apatite---------.............................------------------..... 69-75

Uranium contents, Igneous apatite ------------------------------------ 50-51,79
sedimentary apatite......-------------------------------------------.. 50-51,72
Uranyl minerals ---.............------------------------------------------------- 52,74
Uranyl pyrophosphates .....------------------------------------------------ 74
Vanadates---..............------------------------------------------------------ 52
Varn mine, Florida .........-------------------------------------------------- 69

Watson mine, Florida .................------------------------------------------ 609
Wavellite .............-- ---------------------------------------------------- 57,60,69
Weathering--............------------------------------------------------------- 64,65
laterite --.............------------------------------------------------------ 65
redistribution of elements during-.....---......-- ------------------... 60-61,63-4
uranium oxidation during........-----.....------- ----------------- 72-73,82,84
Weathering patterns---..........------------------------------------------- 60,64-66

X-ray powder diffraction analysis.....................----------------------------...............--- 6.......6

University of Florida Home Page
© 2004 - 2010 University of Florida George A. Smathers Libraries.
All rights reserved.

Acceptable Use, Copyright, and Disclaimer Statement
Last updated October 10, 2010 - - mvs