001

Element Concentrations in Soils andOther Surficial Materials of theConterminous United States/B^HANSFORD T. SHACKLETTE and JOSEPHINE G. BOERNGEN/

• c . U . S . GEOLOGICAL S U R V E Y P R O F E S S I O N A L P A P E R 1 2 7 0 , .

An account of the concentrations of50 chemical elements in samples ofsoiL<! and other regnlitfis

R E C E I V E DJU: ) 11 1284

ECOLOGY & ENVIRONMENT

UNITED STATES G O V E R N M E N T P R I N T I N G OFFICE. WASHINGTON 1984

UNITED STATES DEPARTMENT OF THE INTERIOR

WILLIAM P. CLARK, Secretary

GEOLOGICAL SURVEY

Daila.. 1. ? .-':, Director

Library of Congreaa Cataloging In Publication DataShacklette, Hanaford T.Element concentration! in aoila and other turflctal materiala of the conternunoua United Statea.(Geological Survey professional paper; 1270)Bibliography: 106 p.Supt.orOoea.No.: 119.161. Soil*—United SUtet—Composition.L Boerngen, Joaephine G. II. Title. III. Seriea

S598.A1S6 63L4-7-73 8Z-600084AACR2

Fn<t sale by the Distribution Branch, U.S. Geological Survey,604 South Pickett Street, Alexandria, VA 22304

ELEMENT CONCENTRATIONS IN SOILS AND OTHERSURFICIAL MATERIALS OF THE

CONTERMINOUS UNITED STATES

By HANSFORDT. SHACKLETTE and JOSEPHINE G. BOERNGEN

ABSTRACT

Samples of soils or other regoliths, taken at a depth of approxi-mately 20 cm from locations about 80 km apart throughout the conter-minous United Stages <»«•• onalyzed for their content of elements.In this manner, 1,313 sampling sites were chosen, and the resultsof the sample analyses for 50 elements were plotted on maps. Thearithmetic and geometric mean, the geometric deviation, and a histog-ram showing frequencies of analytical values are given for 47 ele-ments.

The lower concentrations of some elements (notably, aluminum,barium, calcium, magnesium, potassium, sodium, and strontium) inmost samples of surficial materials from the Eastern United States,and the greater abundance of heavy metals in the same materialsof the Western United Slate*, Indicates a regional ({economical pat-tern of the largest scale. The low concentrations of many elementsin soils characterize the Atlantic Coastal Plain. Soils of the PacificNorthwest generally have high concentrations of aluminum, cobalt,iron, scandium, and vanadium, but are low in boron. Soils of theRocky Mountain region tend to have high concentrations of copper,lead, and zinc. High mercury concentrations in surficial materials arecharacteristic of Gulf Coast sampling sites and the Atlantic coast sitesof Connecticut, Massachuetts, and Maine. At the State level, Floridaha* the most striking geochemical pattern by having soils that arelow in the concentrations of most elements considered in this study.Some smaller patterns of element abundance can be noted, but thedegree of confidence in the validity of these patterns decreases asthe patterns become less extensive.

INTRODUCTION

The abundance of certain elements in soils and otherBurficial materials is determined not only by the ele-ment content of the bedrock or other deposits fromwhich the materials originated, but also by the effectsof climatic and biological factors as well as by influencesof agricultural and industrial operations that have actedon the materials for various periods of time. The diver-sity of these factors in a large area is expected to resultin a corresponding diversity in the element contentsof the surficial materials.

At the beginning of this study (1961), few data wereavailable on the abundance of elements in surficial ma-terials of the United States as a whole. Most of theearly reports discussed only the elements that were ofeconomic importance to mining or agriculture in a

metallogenic area or State; and the data, for the mostpart, cannot be evaluated with reference to average,or normal, amounts in undisturbed materials becausethey were based on samples of deposits expected tohave anomalous amojnis of certaiifelements, or werebased only on samples from cultivated fields.

We began a sampling program in 1961 that was de-signed to give estimates of the range of element abun-dance in surficial materials that were unaltered or verylittle altered from their natural condition, and in plantsthat grew on these deposits, throughout the contermin-ous United States. We believed that analyses of thesurficial materials would provide a measure of the totalconcentrations of the elements that were present at thesampling sites, and that analysis of the plants wouldgive an estimate of the relative concentrations amongsites of the elements that existed in a chemical formthat was available to plants. Because of the greatamount of travel necessary to complete this sampling,we asked geologists and others of the U.S. GeologicalSurvey to assist by collecting samples when travelingto and from their project areas and to contribute appro-priate data they may have collected for other purposes.The reponse to this request, together with the samplesand data that we had collected, resulted in our obtain-ing samples of surficial materials and plants from 863sites. The analyses of surficial materials sampled in thisphase of the study were published for 35 elements byplotting element concentrations, in two to five fre-quency classes, on maps (Shacklette, Hamilton, andothers, 1971).

Soon after the publication of the results of this study,interest in environmental matters, particularly in theeffects of contamination and industrial pollution, in-creased greatly. At the same time, technological ad-vances in analytical methods and data processing facili-tated measurements of geochemical and other parame-ters of the environment. In response to the need forbackground data for concentrations of certain elementsof particular environmental concern, the samples of sur-ficial materials that were collected for the first study(Shacklette, Hamilton, and others, 1971) (with some ad-

1

ELEMENT CONCENTRATIONS IN SOILS. CONTERMINOUS UNITED STATES

dilional samples) were analyzed fur other elumunU, undthe results were published in U.S. Geological SurveyCirculars: for mercury, Shacklette, Boerngen, andTurner (1971); for lithium nnd cadmium, by Shacklette,and others (1973); and ;'..•* selenium, fluorine, and arse-nic, Shacklette and others U974).

The collection of samples for this study continued,as opportunities arose, until autumn 1975, resulting inthe sampling of an additional 355 sites that wereselected to give •>. more uniform geographical coverageof the contermir as United States. This sampling con-tinuation is reft:.'red to as phase two. These sampleswere analyzed, and the data «""•« "^erged. with_thoaeof the original samples to produce the results given inthe present report. In addition, the availability ofanalytical methods for elements not included in the ear-lier reports permitted data to be given on these ele-ments in the more recently collected samples.

The collection localities and dates, sample descrip-tions, and analytical values for each sample in the pre-sent report were published by Boerngen and Shacklette(1981). The elemental compositions of only the surficialmaterials are given in this report; the data on analysesof the plant samples are held in files of the U.S. Geolog-ical Survey.

ACKNOWLEDGMENTSThis study was made possible by the cooperation of

many persons in the U.£. Geological Survey. We thankD. F. Davidson, A. T. -Miesch, J. J. Connor, R. J.Ebens, and A. T. Myers lor their interest in, and con-tinued support of, this study. The sampling plan wassuggested by H. L. Cannon, who also contributedanalytical data from her project areas and samples fromher travel routes. Others of the Geological Survey whocollected sami :es, and to whom we express gratitude,are: J. M. Bowies, F. A. Branson, R. A. Cadigan, F.C. Canney, F. W. Cater, Jr., M. A. Chaffey, ToddChurch, J. J. Connor, Dwight Crowder, R. J. Ebens,J. A. Erdman, G. L. Feder, G. B. Gott, W. R. Griffitts,T. P. Hill, E. K. Jenne, M. I. Kaufman, J. R. Keith,Frank Kleinhampl, A. T. Miesch, R. F. Miller, R. C.Pearson, E. V. Post, Douglas Richman, R. C. Sever-son, James Scott, D. A. Seeland, M. H. SUr.tz, T. A.Steven, M. H. Strobell, V. E. Swanson, R. E. Tidball,H. A. Tourtelot, J. D. Vine, and R. W. White. Wethank the following members of the U.S. Departmentof Agriculture Soil Conservation Service for providingsoil samples from areas in Minnesota: D. D. Barron,C. R. Carlson, D. E. DeMartelaire, R. R. Lewis,Charles Sutton, and Paul Nyberg.

We acknowledge the analytical support provided bythe following U.S. Geological Survey chemists: Lowell

ArtiM, Philip AruHcuvuKu, A. J. Uurtcl, S. D. IlotU,L. A. Bradley, J. W. Budinsky, Alice Caemmerer, J.P. Cahill, E. Y. Campbell, G. W. Chloe, Don Cole,E. F. Cooley, N. M. Conklin, W. B. Crandell, MauriceDevalliere, P. L. D. Elmore, E. J. Finlay, JohnnieGardner, J. L. Glenn, T. F. Harms, R. G. Havens,R. H. Heidel, M. B. Hinkle, Claude Huffman, Jr., L.B. Jenkins, R. J. Knight, B. W. Lanthorn, L. M. Lee,K. W, Leong, J. B. McHugh, J. D. Mensik, V. M. Mer-ritt, H. T. MUlard, Jr., Wayne Mountjoy, H. M.Nakagawa, H. G. Neiman, Uteana Oda, C. S. E. Papp,R. L. Rahill, V. E. Shaw, G. D. Shipley, HezekiahSmith, A. J. Sutton, Jr., J. A. Thomas, Barbara Tobin,J. E. Troxel, J. H. Turner, anu G. H. VanSickle.

We were assisted in computer programming for thedata by the following persons of the U.S. GeologicalSurvey: W. A. Buehrer, G. I. Evenden, J. B. Fif*>Alien Popiel, M. R. Roberts, W. C. Schomburg, G.^Seiner, R. C. Terrazas, George VanTrump, Jr., andR. R. Wahl.

REVIEW OF LITERATUREThe literature on the chemical analysis of soils and

other surficial materials in the United States is exten-sive and deals largely* with specific agricultural prob-lems of regional interest. Many of the papers were writ-ten by soil scientists and chemists associated with Stateagricultural experiment stations and colleges of agricul-ture, and most reports considered only elements thatwere known to be nutritive or toxic to plants or ani-mals.

Chemists with the U.S. Department of Agricu eprepared most early reports of element abundance insoils for large areas of the United States. (See Robin-son, 1914; Robinson and others, 1917). The 1938 year-book of agriculture was devoted to reports on soils ofthe United States; in this book, McMurtrey and Robin-son (1938) discussed the importance and abundance oftrace elements in soils. Amounts of the major elementsin soil samples from a few soil profiles distributedthroughout the United States were compiled by the aofldentist ?, Marbut (1935) to illustrate characteris-es of so. itsThe us j of soil analysis in geochemical prospecting

began in this country in the 1940*8, and many reportswere published on the element amounts in soils fromareas where mineral deposits were known or suspectedto occur. Most of these reports included only a few ele-ments in soils from small areas. This early geochemiealwork was discussed by Webb (1963) and by Hawkes(1957). In succeeding years, as soil analyses became anaccepted method of prospecting and as analytical

COLLECTION AND ANALYSTS OF GEOCHEMICAL DATA

methods were improved, many elements in soils wereanalyzed; still, the areas studied were commonly small.

An estimate of the amounts of elements in average,or normal, soils is useful in appraising the amounts ofelements in a soil sample as related to agricultural, min-eral prospecting, environmental quality, and health anddisease investigations. Swaine (1955) gave an extensivebibliography of trace-element reports on soils of theworld, and he also summarized reports of the averageamounts of elements as given by several investigators.The most comprehensive list of average amounts of rareand dispersed elements in soils is that of Vinogradov(1959), who reported the analytical results of extensivestudies of soils in the Union of Soviet Socialist Repub-lics, as well as analyses of soils from other countries.".. ?.id not state the basis upon which he establish-"1

the average values; however, these values are presuma-bly the arithmetic means of element amounts in samplesfrom throughout the world. In their discussions of theprinciples of geochemistry, Goldschmidt (1954) andRankama and Sahama (1955) reported the amounts ofvarious elements present in soils and in other surficialmaterials, Hawks and Webb (1962) and, more recently,Brooks (1972), Siegal (1974), Levinson (197-1), and Roseand others (1979) gave average amounts of certain ele-ments in soils as useful guides in mineral exploration.

A report on the chemical characteristics of soils wasedited by Bear (1964). In this book, the chapter onchemical composition of soils by Jackson (1964) and thechapter on trace elements in soils by Mitchell (1964)gave the ranges in values or the average amounts ofsome soil elements.

Regional geochemical studies conducted by scientistsof the U.S. Geological Survey within the past two de-cades have been largely directed to the establishmentof baseline abundances of elements in surficial mate-rials, including soils. Most of the earlier work investi-gated these materials that occurred in their natural con-dition, having little or no alterations that related tohuman activities, with the objective of establishing nor-mal element concentrations in the materials by whichanomalous concentrations, both natural or man induced,could be judged. Some of these studies were conductedin cooperation with medical investigators who weresearching for possible relationships of epidemiologicalpatterns to characteristics of the environment. In onestudy, the geochemical characteristics of both naturaland cultivated soils were determined in two areas ofGeorgia that had contrasting rates of cardiovascular dis-eases (Shacklette and others, 1970). In an extensivegeochemical study of Missouri, also conducted coopera-tively with medical researchers, both cultivated andnatural soils were sampled. The results were presentedfor the State as a whole, and for physiographic regions

or other subdivisions and smaller areas, as follows:Erdman and others (1976a, 1976b); Tidball (1976, 1983a,1983b); and Ebens and others (1973). The results ofthese studies, and of other regional geochemical investi-gations, were summarized and tabulated by Connor andShacklette (1975).

Recent regional studies of soil geochemistry by theU.S. Geological Survey related to the development ofenergy resources in the western part of the UnitedStates, including North Dakota, South Dakota, Mon-tana, Wyoming, Colorado, Utah, and New Mexico.These studies established regional geochemicalbaselines for soils, both in undisturbed areas and inareas that had been altered by mining and related ac-tivities. Some of these studies considered the elements;" ""'.Is both as total concentrations and as concentra-tions that were available to plants of the region. Theresults of these studies were published in annual prog-ress reports (U.S. Geological Survey, 1974, 1975, 1976,1977, and 1978). The data on soils, as well as on othernatural materials, in these reports were summarizedand tabulated by Ebens and Shacklette (1981). In astudy of the elements in fruits and vegetables from 11areas of commercial production in the United States,and in the soils on which this produce grew, soils wereanalyzed for 39 elements, as reported by Boerngen andShacklette (1980) and Shacklette (1980).

The average amounts of elements in soils and othersurficial materials of the United States, as determinedin the present study, are given in table 1, with theaverage values or ranges in values that were reportedby Vinogradov (1959), Rose and others (1979), Jackson(1964), Mitchell (1964), and Brooks (1972). The averagesfrom the present study given in table 1 are the arithme-tic means. Although the averages were computed bythe methods described by Miesch (1967), the values ob-tained are directly comparable with the arithmeticmeans derived by common computational procedures.

COLLECTION AND ANALYSIS OFGEOCHEMICAL DATA

SAMPLING PLAN

The sampling plan was designed with the emphasison practicality, in keeping with the expenditures of timeand funds available, and its variance from an ideal planhas been. recognized from the beginning. Because thecollection of most samples was, by necessity, incidentalto other duties of the samplers, the instructions forsampling were simplified as much as possible, so thatsampling methods would be consistent within the widerange of kinds of sites to be sampled. The samples were

ELEMENT CONCENTRATIONS IN SOILS. CONTERMINOUS UNITED STATES

TADLE 1.—Average or median content*, and range in content*, reported for element* in toil* and other ntrficial material*

Thla report

Element

C, total

Nd ————

3, totalSb ————Sc ————

z=

Average

72,0007.2

33580

.92

23.00024,000

759.1

5425

43026,000

17

1.2.09

1.215,000

37

249,000

350.97

12,000

114619

430It

671.6OO

.663.9

.39

310,0001.3

2402.900

9.4

2.780

253.1

60230

Range

700 - C I O , 000<0.1 - 97

<20 - 30010 - 5,000<1 - 15

TO. 5— M-600 - 370,000100 - 320.000

<I50 - 300<3 - 70

1 - 2,000<1 - 700

<IO - 3,700100 - MOO, 000<3 - 70

<0. 1 - 2.5<0.01 - 4.6<0.3 - 9.6

50 - 63,000<30 - 200

<3 - 14050 - MOO, 000<2 - 7,000<3 - 15

<500 - 100,000

<10 - 100<70 - 300<3 - 700

<20 - 6,800<10 - 700

<20 - 210<«OO - *«,ooo

<l - 8.8<T5 - 50

<0.1 - 4.3

16,000 - 450.000<0.l - 10

<3 - 3.00070 - 20,000

2.2 - 31

0.29 - 11<7 - 500

<10 - 200<1 - 50<J - 2.900

<20 - 2,000

>, grmlM* thaa)

Roae, and othcre(1979) (eleaenti

uaeful Ingeochenlcal

proapectlng)

7.529

3000.5

———————

10

6.315

30021.000

0.036

11,000

6.2

3202.5

13

17300

17

33100

2

0.31

1067

137

36270

(M)(M)

- 4

(M)( M )(M)( M )

( M )

(M)

(M)

(M)(A)

(A)

(M)(M)

(M)- 2.000

(A)

(M)

(A)

(A)(M)

(M)(M)

"'""jV)0" Jackaon (1964) Mltchall (1964)

averagea f r o m Typical".' Range Invorldvide average. contenta inaaapllng) or range Scottiah eur—

In valuea face aoila

71,300 10.000 - 60.000 ————————————————j —————————————————————————————————

10 30 ———————————————————————————————————————————— 400 - 3,000

6 —————————————— <3 - 5

1 3 , 700 7 ,000 —————————————————

g ——————————————— <2 - 80

20O ——————————————— 5 - 3.00020 20 <10 - 100

200 ——————————————————————————————————38,000 7,000 - 42.000

13.600 400 - 28,000 —————————————————————————————————————————— <30 - 200

)0 ——————————————————————————————6,300 <6.000 ———————————————

850 —————————————— 200 - 5.0002 —————————————— <l - 5

40 —————————————— 10 - 800800 500 ———————————————

—————————————————————————— <20 - 80

nso ————————————————————————7 —————————————— <3 - 15

.001 —— ——— - ———————————————————————

130.000 ———————————————————————————————

JOO —————————————— 60 - 7004,600 1.200 - 6.000 ———————————————

50 —————————————— 25 - 100

50 ——————————————————————————————

irooka (1972)

Average orraage

310

5006

10

20020

10,000 - 50.00C20

3.01

30

8502.3

IS

40

10

.3

•

10300

13

1100

50

'Author'i ueaga: generally ueed Co indicate th« mott

collected by U.S. Geological Survey personnel alongtheir routes of travel to areas of other types of fieldstudies or within their project areas.

The locations of the routes that were sampled de-pended on both the network of roads that existed andthe destinations of the samplers. Sampling intensitywas kept at a minimum by selecting only one samplingsite every 80 km (about 50 miles; selected for conveni-ence because vehicle odometers were calibrated inmiles) along the routes. The specific sampling sites

ily occurring value.

were selected, insofar as possible, that had surficial ma-terials that were very little altered from their naturalcondition and that supported native plants suitable forsampling. In practice, this site selection necessitatedsampling away from roadcuts and fills. In some areas,only cultivated fields and plants were available for sam-pling.

Contamination of the sampling sites by vehicularemissions was seemingly insignificant, even thoughmany sites were within 100 m or less of the roads. Col-

DATA PRESENTATION

lecting samples at about 20 cm depth, rather than atthe upper soil horizons, may have avoided the effectsof surface contamination on the samples. However, wehad no adequate way of measuring any contaminationthat may have occurred. (See Cannon and Bowles,1962.) Many of the sampled routes had only light veh-icular traffic, and some were new interstate highways.Routes through congested areas generally were notsampled; therefore, no gross contamination of the sam-ples was expected.

The study areas that were sampled follow: Wisconsinand parts of contiguous States, southeastern Missouri,Georgia, and Kentucky, sampled by Shacklette; Ken-tucky, sampled by J. J. Connor and R. R. Tidball;Nevada, New Mexico, and Maryland, sampled by H.L. Cannon; various locations in Arizona, Colorado, Mon-tana, New Mexico, Utah, and t^yo.ning, sampled byF. A. Branson and R. F. Miller; Missouri, sampled byShacklette, J. A. Erdman, J. R. Keith, and R. R. Tid-ball; and various locations in Colorado, Idaho, Montana,South Dakota, Utah, and Wyoming, sampled by A. T.Miesch and J. J. Connor. Sampling techniques used inthese areas varied according to the primary objectivesof the studies being conducted, but generally thesetechniques were closely similar to the methods used insampling along the roads.

In general, the sampling within study areas was moreintensive than that along the travel routes. To makethe sampling intensity of the two sampling programsmore nearly equal, only the samples from selected sitesin the study areas were used for this report. Theselected sites were approximately 80 km apart. Wheretwo or more samples were collected from one site, theywere assigned numbers, and one of these samples wasrandomly chosen for evaluation in this study.

SAMPLING MEDIA

The material sampled at most sites could be termed"soil" because it wa« a mixture of comminuted rock andorganic matter, it supported ordinary land plants, andit doubtless contained a rich microbiota. Some of thesampled deposits, however, were not soils as definedabove, but were other kinds of regoliths. The regolithsincluded desert sands, sand dunes, some loess deposits,and beach and alluvial deposits that contained little orno visible organic matter. In some places the distinc-tions between soils and other regoliths are vague be-cause the materials of the deposits are transitional be-tween the two. Samples were collected from a few de-posits consisting mostly of organic materials that wouldordinarily be classified as peat, rather than soil.

To unify sampling techniques, the samplers wereasked to collect the samples at a depth of approximately20 cm below the surface of the deposits. This depth

was chosen as our estimate of a depth below the plowzone that would include parts of the zone of illuviationin most well-developed zonal soils, and as a convenientdepth for sampling other surficial materials. Where thethickness of the material was less than 20 cm, as inshallow soils over bedrock or in lithosois over large rockfragments, samples were taken of the material that layiust above the rock deposits. About 0.25 liter of thismaterial was collected, put in a kraft paper envelope,and shipped to the U.S. Geological Survey laboratoriesin Denver, Colo.

CHEMICAL-ANALYSIS PROCEDURES

The soil samples were oven dried in the laboratoryand then sifted through a 2-mm sieve. If the soil mate-rial would not pass this sieve, the sample was pul-verized in a ceramic mill oeuit sewing. Finally, thesifted, minus 2-mm fraction of the sample was used foranalysis.

The methods of analysis used for some elements werechanged during the course of this study, as new tech-niques and instruments became available. For most ele-ments, the results published in the first report(Shacklette, Hamilton, and others, 1971) were obtainedby use of a semiquantitative six-step emission spec-trographic method (Meyers and others, 1961). Themethods used for other elements were: EDTA titrationfor calcium; colorimetric (Ward and others, 1963) forphosphorus and zinc; and flame photometry for potassi-um. Many of the elements analyzed in the 355 samplescollected in phase two of the study were also analyzedby the emission spectrographic method (Neiman, 1976).Other methods were used for the following elements:flame atomic absorption (Huffman and Dinnin, 1976) formercury, lithium, magnesium, sodium, rubidium, andzinc; flameless atomic absorption (Vaughn, 1967) formercury; X-ray fluorescence spectrcrr.etry (Wahlberg,1976) for calcium, germanium, iron, potassium, seleni-um, silver, sulfur, and titanium; combustion (Huffmanand Dinnin, 1976) for totul carbon; and neutron uctivn-tion (Millard, 1975, 1976) for thorium and uranium.

DATA PRESENTATIONSummary data for 46 elements are reported in tables

1 and 2. In table 1, the element concentrations foundin samples of soil and other surficial materials of thisstudy are compared with those in soils reported in otherstudies. Arithmetic means are used for the data of thisstudy to make them more readily compared with thedata generally reported in the literature. These arith-metic means were derived from the estimated geomet-ric means by using a technique described by Miesch(1967), which is based on methods devised by Cohen(1959) and Sichel (1952). The arithmetic means in table

ELEMENT CONCENTRATIONS IN SOILS, CONTERMINOUS UNITED STATES

1, unlike the geometric means shown in table 2, areestimates of geochemical abundance (Miesch, 1967).Arithmetic means are always larger than correspondinggeometric means (Miesch, 1967, p. Bl) and are esti-mates of the fractional part of a single specimen thatconsists of the element of concern rather than of thetypical concentration of the element in a suite of sam-ples.

Concentrationn of 46 elements in samples of thistudy are presented in table 2, which gives the determination ratios, geometric-mean concentrations and deviations, and observed ranges in concentrations. 1Tnanalytical data for most elements as received from thilaboratories were transformed into logarithms becausiof the tendency for elements in natural materials, particularly the trace elements, to have positively skewa

TABLE 2.—Mean concentrations, deviation*, and ranges of element* in sample* of soil* and other surficial material* in the conUratmumUnited State*

(Horn tnd rangn an reported In pan* per million luK'H). and meana and deviation* an (tonwtrk ncept a* indicated. Ratio, number of aample* in which In* atcmeat em **••In meamirable eonemlraliani u number of aamplea analyied. <. Ira* than: >, rreaur than)

Klemrnt

Al ,Ae-8—Ba-Se-

»r-C,Ca.C«-Co-

Cr-Cu-F—F«.

Ce-Ha-1 —K.l.a-

Ll-H*.

Na.

Hb-Hd-

*—Pb-

Rb- —————

pprernc

percent-percent

percent

percent

percent

percent

—————

S, percent-Sb ——————Sc-S«-

81,So-Sr-Tl.Th-

U —

Y —Tb-Zn-

percent1

percent

ConteralnoueUnited Statea

Mean

4 .75.2

26440

.63

.561.6.92

636.7

3717

2101.8

13

1.2.038.73

1.330

20.44

330.39.59

9.34013

26016

58.12.48

7.5.26

31.8*

120.24

8.6

2.358212.6

48inn

Devia-tion

2.482.231.972.142.38

2.502.374.001.782.19

2.372.443.342.382.03

1.372.522.H)

.791.92

1.853.282.772.723.27

1.751.682.312.671.86

1.722.042.271.822.46

6.682.363.301.891.53

1.7)2.231.781.791.931 01

EatlMtedarithmetic

•can

7.27.2

33580

.92

.852.32.4

759.1

5425

4302.6

17

1.2.089

1.2None

37

24.90

550.97

1.2

1146

19430

19

67.16.67

8.9.39

Move1.3

240.29

9.4

2.780233.1

60310

Ratio

K 6 I : 77n728: 730506:778778:778310:778

113:220250:250777:77781:683

698:778

778:778778:778398:610776:777767:776

224:224729:733169:2467 7 7 l 7 7 7462:777

731:731777:778777:777

57:774744i 744

418:771120:538747:778524:524712:778

221:22434:22435:223

685:778390)733

250:250218:224778:778777:7771931195

224:224778:778739:778734t764766:76677T, 7711

Ueitern United Statee(veil of 96th meridian)

Mean

5.05.5

23580

.68

.321.71.8

657.1

4121

2802.1

16

1.2.046.79

1.63.0

22.74

380.85.97

8.73615

32017

69.13.47

8.2.23

30.90

200.22

9.1

2.570222.6

35IAQ

Devla* Obaervedtlon range

2.001 .981.991.722.30

2.742.373.031.711.97

2.192.072.321.951.68

1.322.332.33

.711.89

1.582.211.982.171.95

1.821.762.102.3)1.80

1.502.372.151.742.43

3.702.112.161.781.49

1.431.931.661.631.791.77

0.3 - >IO<0. 10 - 97

<20 - 30070 - 5,000<l - IS

<0.3 - II0.16 - 100.06 - 32<150 - 300

<) - 50

3 - 2,0002 - 300

<IO - 1,900O.I - >IO

<5 - 70

0.58 - 2.5<0.01 - 4.6

<O.S - ».60.19 - 6.3

<30 - 200

5 - 1 3 00.03 - >IO

30 - 5.000<3 - 7

0.03 - 10

<IO - 100<70 - 300

<3 - 70040 - 4,500

<IO - 700

<20 - 210<0.08 - 4.8

<1 - 2.6<5 - 50

<0.1 - 4.3

1 5 - 4 4<0.1 - 7.4

10 - 3,0000.03 - 2.0

2.4 - 31

0.4* - 7.97 - 500

<10 - 150<l - 7010 - 2,100

<ta - i voo

Kit 1 matedarithmetic

•can

7.47.0

29670

.97

.862.53.3

739.0

5627

4402.6

19

1.2.063

1.2None37

251.0

4801.11.2

104319

46020

74.19.62

9.6.34

NOM1.2

270.26

9.8

2.788253.0

A3I9O

Ratio

4V1:477521: 527423:541541:541169:525

78:128162)162514:514

70:68940): 533

541:541523:533390:435539: 5404)1:540

130:131534i534»0t 1 51

33715372941516

479:527528:528537)54032:324

36)1449

322:498109:33244315403*Or3fl2422:541

107:1)120:1)131:1)1

3891526449t5)4

1 56i 1 56123:131501:540540: 540102:102

130:130516:541477:541432:486473:4823Jti S4I

Eaatern United State*(eaat of 96th ••rldlan)

•M̂ean

3.3'!*.8

31290

.55

.621.5.34

633.9

3313

1301.49.)

1.1.081.68

1.229

17.21

260.32.23

104611

2 no14

4).10.52

6.5.30

34.86

53.28

7.7

2.143202.6

40?JO

Devia-tion

2.*72. 561.882.352.53

2.182.883.081.852.37

2.602.804.192.872.38

1.432.522.81

.731.98

2.163.553.823.934.55

1.651.582.642.951.93

1.941.342.381.902.44

6.642.813.612.001.58

2.122.511.972.062. II3.OI

KittenObaerved •• <•*ranee m

0.7 - >IO<0. 1 - 71

<20 - 15010 - 1,500<1 - 7

<0.) - 5.30.06 - 370.01 - 28<150 - 300<0.) - 70

1 - 1,000<1 - 700

<10 - 3,7000.01 - >IO

<3 - 70

<0.1 - 2.00.01 - 3.4<0.) - 7.0

0.003 - 3.7<30 - 200

<3 - 1400.003 - 5

<2 - 7,000<3 - 15

<0.05 - 5

<IO - 50<70 - 300<5 - 700

<20 - 6.800<IO - 300

<20 - 160<0.08 - 0.31

<l - 8.8<5 - 30

<0.l - 3.9

1.7 - 45<0.l - 10

<5 - 7000.007 - 1.5

2.2 - 2)

0.29 - 11<7 - 300

<10 - 200<l - 50<5 - 2.900

no > i.ooo

3.;7.4

IB428)

J

JZJ.1

7t9,:a22

3«82.:

14

I.J

i.;37

a.^

64t

12M11

36*)17

aj«.«.4

i.:120

1.12.:

6621).:

52rt»

'Heine are arithmetic, deviation* are atandard.

DISCUSSION OF RESULTS

frequency distributions. For this reason, the geometricmean is the more proper measure of central tendencyfor these elements. The frequency distributions for po-tassium and silicon, on thextther hand, are more nearlynormal if the data are not transformed to logarithmsand the mean is expressed as the arithmetic average.

In geochemical background studies, the magnitude ofscatter to be expected around the mean is as importantas the mean. In lognormal distributions, the geometricdeviation measures this scatter, and this deviation maybe used to estimate the range of variation expected foran element in the material being studied. About 68 per-cent of the samples in a randomly selected suite shouldfall within the limits MID and M-D, where M repre-sents the geometric mean and D the geometric devia-tion, .i-^^ * -..> percent "should fall between MID2 andM-D2, and about 99.7 percent between M/D3 and M-D3.

The analytical data for some elements include valuesthat are below, or above, the limits of numerical deter-mination, and these values are expressed as less than(<) or greater than (>) a stated value. These data aresaid to be censored, and for these the mean was com-puted by using a technique described by Cohen (1959)and applied to geochemicul studies by Miesch (1967).This technique requires an adjustment of the summarystatistics computed for the noncensored part of thedata. The censoring may be so severe in certain setsof data that a reliable adjustment cannot be made; withthe data sets used in the present study, however, nosuch circumstances were encountered. The use of theseprocedures in censored data to quantify the central ten-dency may result in estimates of the mean that arelower than the limit of determination. For example, intable 2 the geometric-mean molybdenum concentrationin soils from the Eastern United States is estimatedto be 0.32 ppm, although the lower limit of determina-tion of the analytical method that was used is 3 ppm.Use of this procedure permits inclusion of the censoredvalues in the calculation of expected mean concentra-tions.

The determination ratios in table 2—that is, the ratioof the number of samples in which the element wasfound in measurable concentrations to the total numberof samples—permit the number of censored values, ifany, to be found that were used in calculating the mean.This number is found by subtracting the left value inthe ratio from the right.

The distribution of the sampling sites and the concen-trations of elements determined for samples from thesites are presented on maps of the conterminous UnitedStates (figs. 1-47). Figure 1 shows the locations of siteswhere four elements, bismuth, cadmium, praseodymi-um, and silver, were found in the samples. These ele-ments were determined too uncommonly for reliable

mean concentrations to be calculated. Each of the re-maining maps (figs. 2—47) gives the locations where anelement was found in a sample from a site and the con-centration of the element, shown by a symbol that rep-resents a class of values. By examining the tables offrequency for concentration values of the elements, wewere able to divide the ranges of reported values formany elements into five classes so that approximately20 percent of the values fell into each class. The limitedrange in values for some elements, however, prohibitedthe use of more than two or three classes to representthe total distribution. Symbols representing the classeswere drawn on the maps by an automatic plotter thatwas guided by computer classification of the data, in-cluding the latitude and longitude of the sampling sites.A histogram on each map gives the frequency distribu-tion 01 cfio analytical values, and the assignment ofanalytical values to each class as represented by sym-bols.

We were able to obtain analyses of 11 more elementsfor the 355 samples of phase two of this study thanfor the 963 samples of phase one because of improvedanalytical methods and services. These elements are an-timony, bromine, carbon, germanium, iodine, rubidium,silicon, sulfur, thorium, tin, and uranium. The con-straints of resources and time prohibited analysis of the963 samples of the first phase for these additional ele-ments. Results of analysis of the plant samples thatwere collected at all soil-sampling sites are not pre-sented in this report.

Some elements were looked for in all samples butwere not found. These elements, analyzed by thesemiquantitative spectrographic method, and their ap-proximate lower detection limits, in parts per million,are as follows: gold, 20; hafnium, 100; indium, 10; plati-num, 30; palladium, 1; rhenium, 30; tantalum, 200; tellu-rium, 2,000; and iliaiiium, 50. If lanthanum or ceriumwere found in a sample, the following elements, withtheir stated lower detection limits, were looked for inthe same sample but were not found: dysprosium, 50;erbium, 50; gadolinium, 50; holmium, 20; lutetium, 30;terbium, 300; and thulium, 20.

DISCUSSION OF RESULTSThe data presented in this report may reveal evi-

dence of regional variations in abundances of elementsin soils or other regoliths; single values or small clustersof values on the maps may have little significance ifconsidered alone. Apparent differences in values shownbetween certain sampling routes, such as some of thoseacross the Great Plains and the North Central Stateswhere high values for cerium, cobalt, gallium, and leadpredominate, suggest the possibility of systematic er-

8 ELEMENT CONCENTRATIONS IN SOILS, CONTERMINOUS UNITED STATES

rors in sampling or in laboratory analysis. Some grosspatterns and some of lesser scale, nevertheless, are evi-dent in the compositional variation of regoiiths, asshown in figures 2-47.

The lower abundances of some elements (notably alu-minum, barium, calcium, magnesium, potassium, sodi-um, and strontium) in regoiiths of the Eastern UnitedStates, and the greater abundances of the heavy metalsin the same materials of the Western United Statesindicate a regional pattern of the largest scale. Thisvisual observation of the maps can be substantiated byexamining the mean concentrations for these two re-gions given in table 2. The abundances of these ele-ments differ markedly o.. ^..r.ar side 6Ta line extendingfrom western Minnesota southward through east-cen-tral Texas. This line is generally from the 96th to 97thmeridian, and corresponds to the boundary proposedby Marbut (1935, p. 14), which divides soils of theUnited States into two major groups—the pedalfersthat lie to the east, and the pedocals to the west. Mar-but (1928) attributed the major differences in chemicaland physical qualities of these two major groups to theeffects of climate on soils. A line approximating the 96thmeridian also separates the Orders, Suborders, andGreat Groups of moist-to-wet soils in the EasternUnited States from the same categories of dry soils thatlie to the west, as mapped by the [U.S.] Soil Conserva-tion Service (1969). As shown in table 2, soils of theWestern United States have the highest mean valuesfor all elements considered in this report except for an-timony, boron, bromine, mercury, neodymium, seleni-um, titanium, and zirconium. The differences, however,probably are not significant for these latter elements,except for zirconium.

Superimposed upon this large-scale compositionalvariation pattern are several features of intermediatescale. Perhaps the most notable of these are the lowconcentrations of many elements in soils of the AtlanticCoastal Plain. Soils of the Pacific Northwest are highin concentrations of aluminum, cobalt, iron, scandium,and vanadium, but low in boron, and soils of the RockyMountain region tend to be high in copper, lead, andzinc.

Several small-scale patterns of compositional varia-tion can be noted, among them the high mercury con-centrations in surficial materials from the Gulf Coastof eastern Texas, Louisiana, Mississippi, Alabama, andnorthwest Florida, and a similar pattern on the AtlanticCoast in Connecticut, Massachusetts, and Maine. Highphosphorus values occur in soils along a line extendingwest across Utah and Nevada to the coast of California,then south-east in California and Arizona. At the Statelevel, Florida shows the most striking pattern by hav-

ing low soil concentrations of most of the elements con-sidered in this study.

The concentrations of certain elements do not showwell-defined patterns of distribution, and the regionalconcentrations of some other elements cannot beevaluated because they were not present in detectableamounts in most of the samples, or because the sam-pling density was insufficient. The degree of confidencein regional patterns of element abundance is expectedto be in direct proportion to the number of samplesanalyzed from the region. As the observed patterns be-come smaller, the probability increases that the charac-teristics that form the patterns are the results 01chance.

Some features of element-abundance patterns probably reflect geologic characteristics of the areas that thesoils overlie. Samples from most of the regoiiths overly-ing basic volcanic rocks of Washington and Oregon \-tained higher than average concentrations of iron^Sncother elements, as mentioned earlier. A few soil sam-ples with high phosphorus content are associated withphosphate deposits in Florida, and a single sample irMichigan with high copper content is known to be o.soil that occurs over a copper deposit.

These data do not provide obvious evidences of northsouth trends in elemental compositions that might bexpected to relate to differences in temperature regimes under which the surficial materials developedThere is, moreover, no consistent evidence of significant differences in element abundances betwee;glaciated and nonglaciated areas (the general area ccontinental glaciation includes the northern tier cStates from Montana to Maine and south in places tabout lat 40°N.; see fig. 1\

The world averages of abundance for some elt. .atin soils, as given by Vinogradov (1959) and by other(table 1), do not correspond to the averages of abundance for these elements in the soils of the Unite<States, according to the data presented in this reportThe world averages are too low for the concentration,of boron, calcium, cerium, lead, magnesium, potassiumand sodium in United States soils and other surfickmaterials, and too high for beryllium, chromium, gallium, manganese, nickel, phosphorus, titanium, vanadiurn, and yttrium.

The stability of values for concentrations of most elements seems to be satisfactory because the addition canalytical values for 355 samples of phase two of thstudy to values for 963 samples of the first phase dinot significantly change the geometric means and devutions of element abundance that were reported earlie(Shacklette, Boerngen, and Turner, 1971; ShackletUHamilton, and others, 1971; Shacklette and other

REFERENCES CITED

1973, 1974). Although additional sampling of the sametype as reported here might give a clearer picture ofsmall-to-intermediate element-abundance patterns,mean values reported herein most likely would notchange significantly.

REFERENCES CITEDBear, F. E.. ed., 1964, Chemistry of the soil [2d ed.]: New York,

Reinhold Publishing Corp., 615 p.Boerngen, J. G., and Shacklette, H. T., 1980, Chemical analyses of

fruits, vegetables, and their associated soils from areas of convmercial production in the conterminous United States: U.S.Geological Survey Open-File Report 80-84, 134 p.

———1981, Chemical analysis of soils and other surficial materialsof the conterminous United States: U.S. Geological Survey Open-File Report 81-197, 143 p.

Brooks, R. R., 1972, Geobotany and biogeochemistry in m-n.ml ex-ploration: New York, Harper and Row, 290 p.

Cannon, H. L., and Bowles, J. M., 1962, Contamination of vegetationby tetraethyl lead: Science, v. 137, no. 3532, p. 765-766.

Cohen, A. C., Jr., 1959, Simplified estimators for the normal distribu-tion when samples are singly censored or truncated: Technomet-riM, v. 1, no. 3, p. 217-237.

Connor, J. J., and Shacklette, H. T., 1975, Background geochemistryof some rocks, soils, plants, and vegetables in the conterminousUnited States, with lection* on Field studies by R. J. Ebens,J. A. Krdmnn, A. T. Mlowh, It. It. Tldbull. and II. A. Tourtclot:U.S. Geological Survey Professional Paper 574-F, 168 p.

Ebens, R. J., Erdman, J. A., Feder, G. L., Case, A. A., and Selby,L. A., 1973, Geochemical anomalies of a claypit area, CallawayCounty, Missouri, and related metabolic imbalance in beef cattle:U.S. Geological Survey Professional Paper 807, 24 p.

Ebens, R. J., and Shacklette, H. T., 1981, Geochemistry of somerocks, mine spoils, stream sediments, soils, plants, and watersin the western energy region of the conterminous United States,with tectvmt on Field studies by B. M. Andersen, J. G.Boerngen, J. J. Connor, W. E. Dean, J. A. Erdman, G. L.Feder, L. P. Cough, J. R. Herring, T. K. Hinkley, J. R. Keith,R. W. Klusman, J. M. McNeal, C. D. Ringrose, R. C. Sevenon,and R. R. Tidball: U.S. Geological Survey Professional Paper1237, 173 p.

Erdman, J. A., Shacklette, H. T., and Keith, J. R., 1976a, Element!)composition of selected native plants and associated soils frommajor vegetation-type areas in Missouri: U.S. Geological SurveyProfessional Paper 954-C, p. C1-C87.

———1976b, Elemental composition of corn grains, soybean seeds,pasture grasses, and associated soils from selected areas in Mis-souri: U.S. Geological Survey Professional Paper 954-D, p. Dl-D23.

Goldschmidt, V. M., 1964, Geochemistry: Oxford, Clarendon Press,730 p.

Hawkes, H. E., 1957, Principles of geochenu'cal prospecting: U.S.Geological Survey Bulletin 1000-F, p. 225-355.

Hawkes, H. E., and Webb, J. S., 1962, Geochemistry in mineralexploration: New York, N. Y., and Evanston, I1L, Harper andRow, 416 p.

Huffman, Claude, Jr., and Dinnin, J. L, 1976, Analysis of racks andsoil by atomic absorption spectrometry and other methods, tnMiesch, A. T., Geochemical survey of Missouri—Methods of sam-pling, laboratory analysis, and statistical reduction of data: U.S.Geological Survey Professional Paper 954-A, p. 12-14.

Jackson, M. L., 1964, Chemical composition of soils, tn Bear, F. E.,ed.. Chemistry of the soil [2d ed.]: New York, Reinhold Publish-ing Corp., p. 71-141.

Levinaon, A. A., 1974, Introduction to exploration geochemistry: Cal-gary, Applied Publishing, Ltd., 612 p.

Marbut, C. F., 1928, Classification, nomenclature, and mapping ofsoils in the United States — The American point of view: Soil Sci-ence, v. 25, p. 61-70.

——— 1935, Soils of the United States, pt. 3 of Atlas of Americanagriculture: Washington, D.C., U.S. Government Printing Office,98 p.

McMurtrey, J. E., Jr., and Robinson, W. 0., 1938, Neglected soilconstituents that affect plant and animal development, in Soilsand men — Yearbook of Agriculture 1938: Washington, D.C., U.S.Government Printing Office, p. 807-829.

Miesch, A. T., 1967, Methods of computation for estimating geochemi-cal abundance: U.S. Geological Survey Professional Paper 574-B,15 p.

Millard, H. T., Jr., 1975, Determination of uranium and thorium inrocks and soils by the delayed neutron technique, . '.""'ical Survey, Geochemical survey of the western coal region, sec-ond annual progress report, July 1975: U.S. Geological SurveyOpen-File Report 75-436, p. 79-81.

——— 1976, Determination of uranium and thorium in U.S.G.S. stan-dard rocks by the delayed neutron technique, in Flanagan, F.J., ed. and compiler, Description and analyses of eight newU.S.G.S. rock standards: U.S. Geological Survey ProfessionalPaper 840. p. 61-66.

Mltchull, R. I... 10R4, Trace eli-rm-ntx In m>il«, in Roar, F. E., ml..Chemistry of the noil (2d ed.]: New York, Keinhold PublishingCorp., p. 320-368.

Myers, A. T., Havens, R. G., and Dunton, P. J., 1961, A spec-trochemical method for the semiquantitative analysis of rocks,minerals, and ores: U.S. Geological Survey Bulletin 1084-1, p.207-229.

Neiman, H. G., 1976, Analysis of rocks, soils, and plant ashes byemission spectroscopy, tn Miesch, A. T., Geochemical survey ofMissouri — Methods of sampling, laboratory analysis, and statisti-cal reduction of data: U.S. Geological Survey Professional Paper594- A, p. 14-15.

Rankama, K. K., and Sahama, T. G., 1955, Geochemistry: Chicago,Chicago University Press, 912 p.

Robinson, W. 0., 1914, The inorganic composition of some importantAmerican soils: U.S. Department of Agriculture Bulletin 122.

Robinson, W. O., Steinkoenig, L. A., and Fry, W. H., 1917, Variationin the chemical composition of soils: U.S. Department of Agricul-ture Bulletin 651, 16 p.

Rose, A. W.. Hawkes, H. E., and Webb, J. S., 1979, Geochemistryin mineral exploration [2d ed.]: London, Academic Press,668 p.

Shacklette, H. T., 1980, Elements in fruits and vegetables from areasof commercial production in the conterminous United States:U.S. Geological Survey Professional Paper 1178, 149 p.

Shacklette, H. T.. Boerngen, J. G., Cahill, J. P., and Rahfll, R. L.,1973, Lithium in surficial materials of the conterminous UnitedStates and partial data on cadmium: U.S. Geological Survey Cir-cular 673, 8 p.

Shacklette, H. T., Boerngen, J. G., and Keith, J. R., 1974, Selenium,fluorine, and arsenic in surficial materials of the conterminousUnited States: U.S. Geological Survey Circular 692, 14 p.

Shacklette, H. T., Boerngen, J. G., and Turner, R. L., 1971, Mercuryin the environment — Surficial materials of the conterminous

10 ELEMENT CONCENTRATIONS IN SOILS, CONTERMINOUS UNITED STATES

United States: U.S. Geological Survey Circular 644, 6 p.Shacklette, H. T., Hamilton, J. C., Boerngen, J. G., and Bowles,

J. M., 1971, Elemental composition of surfirial materials in theconterminous United States: U.S. Geological Survey ProfessionalPaper 574-D, 71 p.

Shacklette, H. T., Sauer, H. I., and Mieach. A. T., 1970. Geochemicalenvironments and cardiovascular mortality rates in Georgia: U.S.Geological Survey Professional Paper 674-C, 39 p.

Sichel, H. S., 1952, New methods in the statistical ev&juation of minesampling data: Institute of Mining and Metallurgy Transactions,v. 61, p. 261-288.

Siegel, F. R., 1974, Applied geochemistry: New York, John Wileyand Sons, 353 p.

Swain, D. J., 1955, The trace-clement content of soils: England Com-monwealth Agricultural Bureau, Commonwealth Bureau of Soil" >nee Technical Communication 48, 157 p.

Tiuball, R. R., 1976, Chemical variation of soils in Missouri associateawith selected levels .f the Soil Classification System: U.S.Geological Survey Pi- i^aional Paper 954-B, 16 p.

1983a. Geography >:. -'oil geochemistry of Missouri rrr.culturalsoils, in Gcochemiou survey of Missouri: U.S. Geologvcu PurveyProfessional Paper 964-H, in press.

———1983b, Geochemical classification by factor analysis of Missouriagricultural soils, in Geochemical survey of Missouri: U.S.Geological Survey Professional Paper 954-1, in press.

U.S. Geological Survey, 1974, Geochemical survey of the westernccra ft-••• era, first annual progress report, July 1974: U.S. Geolog-ic*. . -rve/ Open-File Report 74-250,38 p.

———1976, Geochemical survey of the western coal regions, secondannual progress report, July 1975: U.S. Geological Survey Open-File Report 75-436,132 p.

———1976, Geochemical survey of the western energy regions, theannual progress report, July 1976: U.S. Geological Survey OpeFile Report 76-729,138 p. + appendix, 44 p.

———1977, Geochemical survey of the western energy regiorfourth annual progress report, July 1977: U.S. Geological SarrOpen-File Report 77-872, 207 p.

———1978, Geochemical survey of the western energy regions, fifannual progress report, July 1978: U.S. Geological Survey OpeFile Report 78-1106,194 p.

[U.S.] Soil Conservation Service, 1969, Distribution of principal Ionof soils—Orders, Suborders, and Great Groups, in National Atiof the United States of America: U.S. Geological Survey, She86,2p.

Vaughn, W. W., 1967, A simple mercury vapor detector for geocfaeacal prospecting; U.S. Geological Survey Circular 640,8 p.

"" _ '-ov, A. P.. 1969, The geochemistry of rare and dispern-chemical element* in soils [2d ed., revised and enlarged}: MeYork, Consultants Bureau Enterprises, 209 p.

Wahlberg, J. S., 1976, Analysis of rocks and sous by X-ray fiaoncence, in Miesch, A. T., Geochemical survey of Mim»~m-Methods of sampling, laboratory analysis, and statistic*.'tion of data: U.S. Geological Survey Professional Paperp. 11-12.

Ward. F. N., Latin, H. W., Canney, F. C., and others, 1963, Analjncal methods used in geochemical exploration by the U.S. Geoiogcal Survey: U.S. Geological Survey Bulletin 1162,100 p.

Webb, J. S., 1953, A review of American progress in geoeheabtprospecting and recommendations for future British work in tfcfield: Institute of Mining and Metallurgy Transactions, v. 62, p7, p. 321-348.