Title: Cesium-137 and other gamma radioactivity in the Florida environment
Full Citation
Permanent Link: http://ufdc.ufl.edu/UF00098213/00001
 Material Information
Title: Cesium-137 and other gamma radioactivity in the Florida environment a study of selected media
Physical Description: xi, 169 l. : illus. ; 28 cm.
Language: English
Creator: Roessler, Charles Ervin, 1934-
Publication Date: 1967
Copyright Date: 1967
Subject: Cesium -- Effect of radiation on   ( lcsh )
Radioactive substances -- Florida   ( lcsh )
Nuclear activation analysis   ( lcsh )
Environmental Engineering Sciences thesis (Ph. D.)   ( lcsh )
Dissertations, Academic -- Environmental Engineering Sciences -- UF   ( lcsh )
Genre: bibliography   ( marcgt )
non-fiction   ( marcgt )
Thesis: Thesis - University of Florida.
Bibliography: Bibliography: l. 154-167.
General Note: Manuscript copy.
General Note: Vita.
 Record Information
Bibliographic ID: UF00098213
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: alephbibnum - 000423924
oclc - 11032170
notis - ACH2329


This item has the following downloads:

PDF ( 4 MBs ) ( PDF )

Full Text




December, 1967

6 813,030

ROESSLER, Charles Ervin, 1934-

The University of Florida, Ph.D., 1967
Health Sciences, public health

University Microfihns, Inc., Ann Arbor, Michigan


The author acknowledges with gratitude Dr. Billy G. Dunavant, the

chairmann of his supervisory committee, for direction, encouragement, and

invaluable assistance. He also acknowledges assistance of his committee

co-chairman, Dr. Herbert A. Bevis, especially for making many of the

financial arrangements which were necessary for sample procurement. He

wishes to acknowledge the other members of his committees, Dr. Hiarvey

L. Cromroy and Dr. John A. Wethington, Jr.

He especially wishes to thank Dr. E. G. Williams and others of the

Florida State Board of Health for their assistance and willingness to

provide pertinent data and information. Recognition also is given to

the radiological staff of various county health departments, to various

county agents, and to staff members of the Chemistry Division of the

Florida Department of Agriculture for their assistance in selecting and

locating sampling altes and in collecting samples.

He wishes to thank Dr. John E. Moore for making computer time

available and for providing some of the technical facilities necessary

for the diesertation study and Dr. A. 2. Palmer for hie assistance in

planning and conducting of beef sampling. Appreciation is aleo

extended to Dr. James Montelato for vegetable sampling advice, Mr.

Lawrence Fitzgerald for programming assistance, and Dr. John I. Thornby

for statistical advice.

The author also expresses his thanked to Mary Redrick for typing of

the final manuscript.

Finally, the author is deeply indebted to his wife, Genevieve, who

provided valuable assistance in nearly all phases of this project and he

wishes to acknowledge his children, Terry, Cindy, Mary, Francis, Kay,

and Jean, for their patience and occasional assistance during the

course of his graduate study.

The~ work was supported in part by United States Public Health

Service Training Grants No. 5-T1RH3-07(67) and No. 3-T1RH30-04S1(66).



ACKNOWLEDGMENTS .. .. ... ... .. .. .. .. . ii

LIST OF TABLES .. . .. .. .. ... .... * * vi

LIST OF FIGURES . .. ... . .. . * .* * ** viii

ABSTRACT ... .. ... .. .. .. ... . .. .. ix


I. INTRODUCTION ... .. ... .. .. ... .. 1

II. REVIEW OF THlE LITERATURE . ... ... .. ... 4

Reviews of Environmental Radioactivity .. .. .. 8
Sources of Environmental Radioactivity .. .. ... 10
Ecology of Cesium-137 and Other Environmental
Radioactivity .. .. ... .. .. ... .. 11
Environmental Radioactivity Measurements in Florida . 32
Ceeium-137 and Ot her Radionuclides in Selected
Environmental Media--Mathodology and Findinge . .. 42

III. EXPERIMENTAL. APPROACH .. .. .. . .. .. .. .. 54

Selection of Media for this Study . ... ~. .. .. 54
Selection of Beef Sampling Stations .. . . ... 56
Beef Sampling ... .. . .. ... .. ... . 59
Vegetable Sampling .. .. .. . .. . ... 61
Sequence of Study . . ... . ... .. ... 63


Beef Sampling . .. .. .. ... .. .. 66
Vegetable Sampling . ... ... .. .... . .. 67
Beef Sample Preparation ,. ... .. .. . ... 68
Vegetable sample Preparation . .. ... . . .. 68
Gamma Radioactivity Counting . ... .. .. .. .. 71
Interpretation of Comma Spectra 72


V. RESULTS OF BEEF SAMPLING .. ... ... .. .. 76

Ceeiurm-137 Content of Lean Meat from Grain-Fed Beef,
1967 . ... .. .. .. .. .. . . ... 77
Effect of Year of Collection .. . . ,. .. .. 81
Effect of Feeding Program and Meat Quality on Ceeium-
137 Levels . .. .. .. .. .. . . ... 82


Cesium-137 Content of Florida Vegetables--Average
Levels and Regional Variation .. .. ... .. 86
Effect of Year of Sampling on Ceaium-137 Levels in
Florida Vegetables .. ... ... .. .. .. 90
Cesium-137 Content of Florida Vegetablee--0ther
Observations ,. ... .. .. ... ... .. .. 92
Other Gamma-Emitting Radionuclides .. .. . .. 100

VII. DISCUSSION OF RESULTS . .. .. .. ... ... 103

Geographic Variation of Cesiumr-137 .. .. .. . 103
Other Variations in Ceeium-137 Levels . .. .. . 105
Magnitude of the Observed Ceeium-137 Levels . . 116
Influence of Observed Ceeium-137 Levels on Human
Intake and Exposure .. .. . . .. . .. 122
Possible Mechanisms and Factors Influencing
Cesium-137 Levels in Florida .. .. .. . 130
Other Radionuclides . .. .. . .. . .. 133
Discussion of Sources of Error . ... .. . .. 134
Further Investigations Suggested by the Results of
This Work .. .. .. .. ... .. .. .. .. 137


Conclusions . . .. .. . .. .. . 146

APPENDICES . .. ... .. .. .. .. .,. . .. 48



LIST OF REFERENCES ... ... .. . .. . . .. . 154

BIOGRAPHICAL SKETCH . .. .. ... . . .. . .. .. 168


Table Page

MILE2) 17

2. BEEF SAMPLING STATIONS ... .. .. .. .. .. .. .. 57

3. VEGETABLE SAMPLING .. ... ... .. ... .. .. 62


FLORIDA BEEF, 1967 .. . .. .. . .. .. 78

POTASSIUMIRATIO) ................... .. 79





SAMPLING PERIOD AND) REGION ... ... .. . ... 87


TO CBSIIUK-137 CONTENT, 1967 SAMPLES . .. .. .. 89

1966 AND 1967 SAMPLING PERIODS . ... .. .. . ... 91





SAMPLES .. .. .. .. .. .. .... . .. ... 100

INVESTIGATORS .. .. . .. .. ..... . .. . .. 106

INVESTIGATORS .. .. .. .. .. .. .. .. 111

HEIRT BASIS ................... .... 121


1966-1967 .. ... .. .. .. .. .. .. .. .. 125

23. REPORTED BODY BURDENS 07 CESIUM-137 . .. .. . .. .. 128

24. EQUIPMENT AND INSTRUMENITATION . ... ... .. .. .. 150

MATRIX............. ............ 153


Figure Page

AVERAGES AND RANGES .. . . . . .. . 5


RADIOACTIVITY ANALYSIS ... .. .. .. .. .. .. 40

4. BEEF SAMPLING STATIONS .... .. .. .. .. 58


1967 .. ... .. .. .. .. . ... . .. 77

AND REGION .............,,..., ,... 87

STATIONS WITHIN CATEGORIES, 1966 AND 1967 . .. .. .. 96

WITRIN CROPS, 1966 AND) 1967 . ... . .. ... .. 98

REGIONS, 1966 AND 1967 .. ... ... .. .. .. 99

VEGETABLES ....................... 104

Abstract of Dissertation Presented to the Graduate Council
in Partial Fulfillment of the Requirements for the
Degree of Doctor of Philosophy



Charles Ervin Roesaler

December, 1967

Chairman: Billy G. Dunavant, Ph.D.
Co-Chairmant H. A. Bevis, Ph.D.
Major Department: Bicenvironmental Engineering

A study was performed of the kinds, level, and distribution of

galmma-emitting radionuclides in the Florida environment. The investiga-

tion was initiated because of the unuenal levels and characteristic

geographical patterns of casium~-137 (137Ca) found in Florida milk and

forage in earlier studies. Beef and vegetables were selected as the

most important media for sampling.

Sampling was carried out in May and June, 1966, and during January

through July, 1967. Analyses were performed by gamma spectroscopy on

triturated whole sample. Complex gamrma spectra were interpreted in

terms of the individual contributing components by use of the simultaneous

equations method.

The most signaificant gamma-emitting radionuclide present was 137Ca

and the evaluation of the data was concentrated on this nuclide.

Levels of 13Ca in.both beef and vegetables abowed geographical

patterns of variation similar to those reported earlier by others in

Florida milk. Maximum~ level were found in the central and southern

parts of the State, with intermediate levels in the northeastern and

north central parts of the State, and the lowest levels in the northwestern

part of the State. In addition, vegetable samples showed a marked differ-

ence from southeast to southwest, with average level in the southeastern

part of the State as low as in the northwestern part.

The levels of 1370 in lower-quality meat from animal that had fed

primarily on grass were much higher than those in high-quality meet from

feed-lot animals. These levels were higher than any other level reported

in beef in the conterminous United Statee.

There was no apparent difference in 1370 concentrations between

leafy, fruit, and root categories of vegetables; although certain crope

in each category did exhibit consistently higher levels than the other

crop in the same category.

Ceeium-137 concentrations found in this study were compared to

literature values, both to those reported for year previous to this

study and to 1966-1967 value obtained by extrapolating published values

to that time. Ceeium-137 concentrations in both beef and vegetables in

much of the State were considerably higher than the estimated national

averages for the same period; average concentrations in the regionE

exhibiting the lowest average levels were found to be comparable to the

estimated national averages.

The significance of the observed 13Ca level was evaluated in terms

of the human intake of this nuclide. It was estimated that radioactivity

intakes from locally produced food in Northwestern Florida would be

similar to the national average. Three hypothetical case of assumed

diet composition and source of food.vere set up to evaluate levels in the

remainder of the State. Consideration of diet with various combinations

of Florida milk, beef, and vegetables having the average levels shown

in much of the State resulted in predicted 137s intakes and bddy burdened

ranging from two to 20 times the projected national average. The intakes,

body burdens, and resulting whole-body radiation dose estimated for the

various case considered were all less than the applicable radiation

protection guides, but the guides could be approached by individuals

constantly consuming food at the extreme levels found.

It was concluded that unusual environmental factors or mechanisms

are involved in the levels of 1370 found in the Florida environment.

The possibility of such mechanisms has important implications to waste

disposal, hazard evaluation, and nuclear facility operation. The

mechanisms were not identified but there is increasing evidence that the

role of uptake from the soil is greater than that reported for most areas

of the country.



This research involved a study of the kinds, levels, and distribu-

tion of gamma-emitting radionuclides in selected segments of the Florida

environment with emphasis placed on the nuclide cesiurm-137 (1370s) in

animal and vegetable products.

This study was initiated because of unusual levels and characteris-

tic geographical patterns of 1370s found in Florida milk and forage in

earlier studies described in Chapter II. Milk monitoring program of

the United States Publie Health Service (USPRS) and of the Florida State

Board of Health have indicated that the average levels of 137Ce in milk

in much of Florida have been higher than the national average and among

the highest in the nation for a number of years, while strontium-90

(90Sr) levels have been consistently average or below. Further investiga-

tions have shown a consistent geographical pattern within the Btatep aid

it appears that these levels are primarily related to the intake of

1370s by cows through the medium of locally grown forage.

The major effort in this study was directed toward determining the

extent to which these unusual patterns of 1370s are reflected in other

elements of the human food chain, particularly lean beef and truck


In addition, measurements were made of other gamrma-emitting radio-

nuclides in all eamples.

Significance of this Research

First, it is important to the welfare of the people of Flazida that

the extent to which persons in the State are exposed to radiation from

all sources and by all routes be known.

This study contributes to this knowledge by (1) investigating the

extent to which the elevated levels of 13Cs previonely observed in milk

and forage extend to other media, particularly other elements of man's

food chain, (2) determining whether other gamma-emitting radionnelides

show elevated levels in Florida, and (3) evaluating the potential

influence of these nuclides and their levels on man's radiation exposure.

In Chapter II, it is pointed out that many route of radiation

exposure have been and are being evaluated in Florida. Gross alpha and

beta radioactivity levels are measured in a variety of media in this

state. Detailed studies have been performed in selected localities, and

exposure to specific nuclides through the route of milk is being studied

in great detail. However, this work represents the first extensive,

statewide study of exposure to specific nuclides through food chain

elements other than milk.

Second, the information developed here to estimate exposure levels

should also contribute to the underatending of the mechanisms involved in

the previously mentioned unusual levels of 1370 in Florida.

It also is important to predict and evaluate the impact on the

environment--and, eventually, on human health--of nuclear activities such

as the operations of nuclear power plants, nuclear laboratories, and

nuclear powered vehicles. A thorough, systematic study of radioactivity

in the State provide a baseline of radioactivity levels ~against which

any future increased can be compared. This means determining not only

average levels of various radionaclidae for the purposes of comparison

but also normal variations of these level for use both in designing

future surveys and in determining whether apparent increased are signifi-

cant. In addition, knowledge of the mechanisms and rates of transfer

of radionuclides through the environment provides a baste for predicting

the consequences of accidental released of radioactivity. At the same

time, since intentional waste release is governed by criteria based on

various assumptions concerning the normal behavior of radionuclides in

the environment, it is important to be aware of any unusual behavior such

as suggested by the Florida milk 1370s levels.

Finally, this study is significant in that the data collected will

contribute to other environmental radioactivity study efforts in the

State. For example, these data can be need to complement the study of

13708 in Florida milk being carried out by the Florida State Board of

Health -and to serve as an extension to the environmental monitoring being

carried out in the Cape Kennedy off-site socio-economic impact area by

the United States Air Force, the USPaIs, and the Florida State Board of




This study was undertaken as a direct result of the findings of

earlier studies of radioactivity in Florida milk and dairy feeds carried

out by thin author and other. The USPHS began a program of sampling

raw milk for radioactivity analysis in 1957,and out of these initial

effort grew the present USPRS Pasteurized Milk Network which has 63

stations, at least one in every state of the Union, the Canal Zone, and

Puerto Rico.- A station representing Florida became operative at Tampa

during August, 1960. In this network, a composite sample ie taken from

the pasteurized milk marketed in the community and shipped to a USPHS

laboratory for radionuclide analysis.

Average monthly 1370s and 90Sr levels at the Florida station of the

network are compared against the corresponding national ranges and

averages for the period 1961 through February, 1967, in Figure 1.

Consistently, the average levels of 137Ca at the Florida station of thief

network have been well above the national average; aince mid-1964, they

have been the highest in the sampling network. In contract, 90Sr levels

have not been particularly high; with the exception of two months, they

were consistently below the national average. As a result, the 137Ce/90Sr

ratios have been particularly high.

Workers at the Florida State Board of Helalth initiated a program

in 1961 for the statewide collection and regional compositing of raw

milk from a 10 per cent random ample of the dairy farms in the state.2-

CESIUM-137 $



1961 -1196) "g 'l'gs4 1"bs 48





Figure 2 summarizes 137Ca ad 90Sr result for 1963 through 1966.

It shows that the average results reported for pasteurized milk samples

from the Tampa station of the national network are comparable to the

average results for composite raw samples collected from the same general

region. The figure also shows that the Tampa station results for both

nuclidae are reasonably well representative of the reported average of

all milk produced in Florida, particularly the area east and south of

Tallahassee. There does appear to be a gradual transition from lower

137Ca levels in the northwestern part of the State to higher levels in

the eastern, central, and southern parts of the State with the highest in

the central part of the State.

Figure 2 also showe that 137Ca levels in the northwestern part of

the State are below the .State average. These levels fall more rapidly

with time than levels in the remainder of the State and fall from above

the national network average in 1963 through 1965 to below it in 1966.

Statewide average 90Sr results are consistently below the national network

average and remain fairly constant for the eastern, central, and southern

parts of the State. These results show a greater fluctuation in the

Northwest--even rising above the national network average in 1965 and


Porter et al., in a special etudy of the Tampa milkshed during

October, 1963, through February, 1964, observed considerable farm-to-farm

variation within that area and concluded that the elevated milk levels

were due to elevated levels of 137Ca intake by the animals and not to

unusual transfer of the nuclide from the feed to the milk.' They also

identified pangolagrass (Diaitaria dechaboen) as the principal contributor

to the high 137C intake.

-*-Tampa Station National Network
S" National Network Average
*** State Average





i mm,

In 1964, Roeasler and Williame Initiated a study to investigate more

fully the factor influencing radionnelide levels in Florida milr.9

Unpublished reports of their work also show a high farm-to-farm variabili-

ty in milk radionuclide levels. These results enggest that when examining

more farms on a wider geographic basis, equally high or higher intakes

can be attributed to forages other than pangolagrass.7l

Reviews of Environmental Radioactivity

The interest in environmental radioactivity since the early 1950's

is evidenced by the large number of bibliographies, hearings, symposia,

conferences, and books devoted to the subject.

Hoard, Eisenbud, and Harley prepared an extensive bibliography on

radioactive fallout in 1956.10 Because of growing public concern at this

time about the possible effects of fallout from nuclear weapons testing,

the Joint Comrmittee on Atomic Energy of the Congress of the United States

held hearings on radioactive fallout and its effects on man in 1957, and

again in 1959,1,1 The scientific community explored the question of

radioisotopes in the biosphere in a symposium in 1959 and again in the

First National Symposium on Radioecology in 1961.131 The United States

Atomic Energy Commission (USAEC) sponsored a conference on radioactive

fallout from nuclear weapons teast in 1961 and again in 1964.151

Eisenbud authored a text on environmental radioactivity in 1963, and

a collection of paper reviewing the ecology of radioactive fallout, soils,

plants, foods, and man, edited by Fowler, vae published in 1965.171

Natural environmental radioactivity was the subject of a symposium

in 1964 and a 2,000 reference bibliography by Klement in 1965.192

The proceedings of a symposium on radiation and terrestrial eco-

systame in 1965, edited by Hungate, review radioactive fallout phenomena

and mechanisms, primordial and cosmic-ray produced radionuclides, soil-

plant relationship, and radionnelide cycling as well as present results.

of individual research.2 The Second National Symposiurm on Radioecology

was held in 1967, at which time emphasia was placed on reporting .

individual investigations of both effects of radiation on the environment

and behavior of radionuclides in the environment.22

Comar reviewed the literature up to January, 1965, relating to the

movement of fallout radionuclides through the biosphere and man; Ellia

reviewed and appraised the level and biological significance of current

fallout levels in the United Kingdorm in 1965.232 The fallout pattern

in the Nordic countries was a topic at a proceedings of the First Nordic

Radiation Protection Conference in Sweden in 1966.25 About the same time

Leistner, on behalf of Euratom, reviewed the reported research on the

radioactive contamination of foods of animal origin, tabulating reported

distributions, retention, and ambient levels.26

A book edited by Russell in 1966 is devoted entirely to radioactivity

and human diet.27

In August, 1959, the President of the United States directed the

Secretary of Health, Education, and Welfare to collate, analyze, and

interpret data on environmental radiation levels. This assignment result-

ed in the appearance in April, 1960, of the publication, Radiological

Health Data (subsequently changed to Radiological health Data and

ESRepot)* This monthly publication contained regular reports of routine

national and worldwide monitoring programs operated by ~agencies of the

United States as well as frequent special reports of state monitoring

programs, special studies, special evaluation of particular problems, and

topical revieve of current research.28

Sourterr of 'Environlmental Radioactivity

According to the best present knowledge, man has always been exposed

to varying degrees of natural background radiation from cosmic radiation

and from naturally occurring radionnelides.

Eisenbud points out that the world inventory of radioactive materials

prior to World War II, both in the environment and in the laboratory, was

confined to those which occurred in nature, with the exception of a

relatively few millieuries of radioactivity produced in cyclotrone during

the late 1930's.17 Construction of large nuclear reactors during the war

and the associated operations for extracting plutoniumn from irradiated

uranium resulted in the first extensive occasions for contaminating the

environment with radioactive substances. Then, in the late 1940's and

continuing at an accelerated rate throughout the 1950's, there began a

series of nuclear weapons teste that resulted in a worldwide distribution

of radioactive materials in all segments of the environment--the atmos-

phere, the soil, food chain, and man himself.

Eisenbud liste the principal sources of environmental radioactivity

as natural radioactivity, preparation of nuclear fuel through the stages

of mining, concentrating, milling and fabrication, reactor operations and

accidents, conventional radioisotope nee, aerospace applications of

radioisotopes and reactors, fuel reprocessing and radioactive waste

disposal.17 Comar identifies the major source to date as radioactive

debris from the testing of nuclear weapons.23 On the other hand,

important future sources will likely be nuclear reactor operation, radio-

isotope use, and radioactive waste disposal.

Williame ejiJE. pointed out that varying levels of natural radio-

activity do occur in Florida and that these materials do have an

opportunity to become relocated in the environment through the operations

of phosphate mining, phosphate fertilizer production, and subsequent use

of the by-products.29 With regard to the other sources of environmental

radioactivity just mentioned, there is no uraniurm mining, concentrating,

or milling, nor fuel fabricating or reprocessing, nor large acale or

Commercial radioactive waste disposal in the State. Reactors are presently

limited to the University of F1orida Training Reactor, and there have

been no reported contaminating incidents in the State.

The Pinellas Peninsular Pleat of the USAEC is located near St.

Petereburg, Florida, but from the regular published reports of the

contractor's environmental monitoring in the vicinity, it may be inferred

that radionuclides potentially reaching the environment from this plant

are limited to tritium.30 This meman that, with this exception, at the

present time any widespread radioactivity in the Florida environment auet

be naturally, occurring radioactivity or must Come from fallout

injected into the atmosphere. At the same time, the environmental

mechanisms affecting activity from these sources will affect in much the

same manner any future radioactivity from local contamination sources.

With regard to the future, several nuclear power plants are

definitely planned for the State of Florida and it is likely that radio-

isotope or reactor powered electrical generators or propulsion devices

will be used in future vehicles launched from the missile testing and

space flight complex at Cape Kennedy.3

:Ecology of Ceeium-137 and Other Environmental Radioactivity

It was pointed out in the preceding section that, with certain

local exceptions, the major source to date of environmental radioactivity

is radioactive fallout from the nuclear weapons testing. Ceeium-137 and

90Sr are considered biologically to be the most important radionuclides

in long-range fallout because of their long physical half-lives, the high

yield of these nuclidea in fission, and their respective chemical

similarities to the natural body constituents, potassiumn and calcium.233

Indeed, the majority of studies of special nuclides in the environment

appear to be concerned with these two nuclidea.

The remaining examination of the literature emphasizes 13Cs, but

gives attention to other nuclides to the extent that they might also

appear in samples examined for 137s..
In 1959, Langham and Anderson reviewed biosphetic contamination from
Cs. They estimated production by weapons tests and deposition up to

that time, reported important exposure routes to man, and listed an

extensive bibliography.33 They proposed expression of 1370s levels as

13Cs per grea of potassium because of the metabolic similarity of cesium

and potassium. A later exhaustive review of cesium ecology was made by
Davis in 1963.34

In 1965, Comar summarized and updated this information in a concise,

yet comprehensive, review of~ the movement of fallout nuclides through the

biosphere.23 He observes that cesium and potassium are chemically

similar. However, since they are not metabolically interdependent, the

137Ce/potassium ratio is not as meaningful as the 9Sr/calcium ratio.

Nevertheless, 1370 concentrations are often expressed relative to

poasesium for two reaason: (1) naturally radioactive potassium-40 ( OK)
and 1370s ae often measured simultaneously by gamrma spectrometry, with

401 thereby serving as an internal standard, and (2) in man, the cesium/

potassium ratio correlates better with lean body seass and hence with

radiation dose than does the 1370 per kilogram (kg) of body eight.

The worldwide deposition of 1370 is usually estimated by applying

a17 s90r ratio to the values of deposition of 90r since it has been

studied more extensively. The 137C 790Sr ration vary between 1.0 and

3.0 as these nuclides are produced, but this ratio will change markedly

as the two elements pass through metabolic systems.

Radionuclide pathways to man

Fallout radionnelides reach man primarily through plants, either by

his consumption of foods of plant origin or indirectly by his consumption

of animal products.3

Plants become contaminated by radioactive materials from the atmos-

phere by either (1) indirect contamination which occurs when radioactive

materials enter the soil and passes into the plant through the roots as

do soil nutrients or (2) direct contamination in which the material is

deposited on some portion of the plant and passage through the soil is by-

passed.23 Direct contamination reflects events of the recent past, varies

with the rate of deposition, and is designated "rate dependent." Indirect

contamination is governed by the total amount available in the soil and is

designated "cumulative dependent." Indirect contamination has the

characteristics that (1) material deposited before the plant develops can

still enter the plant, (2) short-lived radionuclides that enter the soil

will, have a high probability of decaying to insignificant levels before

reaching the plant roots, and (3) radioactive substances entering the

soil mlay be diluted with soil substances, rendered unavailable to plants

by -fixation to soil materials, or discriminated against in plant uptake.

The concepts of cumulative and rate dependent contributions to radio-

nuclide levels in an environmental component are illustrated in the

generalized expression:

C = Pd d + rFr

where C = concentration of the radionaclide in the sample
on interest,

Fd = cumulative deposition,

PC = current rate of deposition,

and pd and pr are the proportionality factors for cumulative deposition
and deposition rate, respectively, for the particular system and nuclide

under consideration.3356

Once radionaclides are deposited on or incorporated into plants, the

grazing animal effectively collectB CDntaminatioR fTOm plant material and

concentrates it in animal products; however various factors such as

metabolic behavior of the specific nuclide and animal feeding and manage-

ment practices influence the relationship between the amount deposited

in tissues or transferred to secretions that are used for food.23

In order to select important media for further study and to consider

possible movements of radionuclides in the Florida environment and

possible geographic, media, and temporal variations, it is necessary to

review briefly what is known about the passage of radianuclides, particu-

larly 137s, through various segments of the environment.

Origin, transport, and deposition of radioactive fallout

Klement and Langham both reviewed fallout phenomena and mechanisms

in 1965.363 Klement identified the sources of injection of radio-

activity into the atmosphere as nuclear weapons teete, peaceful nuclear

explosives tests or utilization, reactor operations, and space applications

of nuclear energy. He placed weapons tests foremost from the standpoint

of widespread environmental radioactivity. The radioactivity can include

fission products (over 200 different nuclidea of about 36 elements between

atomic numbers 28 and 65) and activation products that come about as a

result of neutron interaction with soil, air, water, or parts of the

nuclear device at the time of the detonation. Radioactive debris is

usually apportioned into three fractions: (1) fallout in the immediate

vicinity, (2) material injected into the troposphere, and (3) material

injected into the stratosphere.

Local fallout is not a concern in this Florida study, but tropospher-

ic fallout can deposit short-lived nuclides in the weeks immediately

following atmospheric weapons tests. Deposition from tropospheric contamil

nation is governed by tropospheric air flow and should be highest at the

latitude of injection with a certain amount of dispersion and displace-

ment and with local low altitude variations due to terrain such as

mountains and land-water contrasts.173 One exception is the downward

and equatorward movement along isoentropic surfaces of debris injected

into the upper polar troposphere. A second exception is the poleward

movement and sinking in subtropical regions of mid- and upper-tropospheric

injections in equatorial regions.

Material injected into the stratosphere may remain there from several

months.to several years and provides the source of long-lived materials

currently being deposited on a worldwide basis. Stratospheric materials

are transferred to the troposphere, from which deposition is primarily

due to precipintiato with a small increment due to dry deposition.36

List et al. described the generalized worldwide deposition of 9Sr up

to late 1963.3 The mean latitudinal distribution in the Northern

Rlemisphere showed a very low deposition in the vicinity of the north polar

region, a rapid increase to a maximum between 40. and 50. N, and then a

gradual decrease in the vicinity of the equator. Davie prepared a similar

latitudinal deposition curve for average 1370 content of precipitation

for 1955-1958.3 Florida lies between 25. and 31. N, a location whose

deposition on the mean distribution curve lies midway. between the

Northern Hemaisphere peak and the equatorial minimum. Within latitude

bands, differences in rainfall sad possibly other meterological phenomena

canae variations in deposition. It has been suggested that fallout over

ocean might be higher than over land but List et al. report that comparison

of soil from eites with maritime exposure to soil with sites from continen-

tal exposure shows no systematic difference either in total deposition or

in amount of 908r deposited per inch of rainfall.37

Two major techniques have been employed for the direct quantitative

assessment of fallout deposition on the earth's surface: (1) deposition

sampling (dry and precipitation) and (2) soil sampling. Based on soil

sampling in 1963-1964, maps of 90Sr deposition were prepared for the

world and for the United States. The latter shows Florida lying between

the 100-millicurie per square mile contour at the southern tip of the

State and the 125-millieurie per square mile contour along the northern


Stratospheric deposition is neually described by a simple exponential

function; howeverMachta has described shortcomings of such a simplified

model .38 Comar quotes an overall mean stratospheric residence time of

two years, while data presented by Klement imply a value of one year.233

As an aid to predicting general fallout levels, the Federal Radiation

Council divides the United States into "wet" and "dry" areas depending

upon average annual rainfall (roughly separated by the 20-inch rainfall

isoline).394 It further denotes an intermediate area with lightly less

than 20 inches of annual rainfall and, by superimposing the latitudinal

effect on the precipitation effect, assigns most of the State of Florida

to another imrmediate area with "an expected lesser fallout compared with

the 'wet' eastern United States because of its subtropical location."

Estimates of 9Sr deposition in the United States through 1966,

derived from several sources, are shown in Table 1. Values for years

beyond 1966 can be obtained by an extrapolation based on the assumptions

that deposition is described by a single exponential function and that

there will be no signifiesnt additions to the stratospheric content.



Increment Accumulated
Wat Area Dry Area Net Area Dry Area

Based on Federal Radiation Council Report No. 6:40

1963 - 150 65

1964 30 12 180 77

1965 15 6 195 83

Total after 1965 25 10 220 93

1966 estimate36 7.5 3 203 86

Although air is the medium by which fallout is delivered to terree-

trial environment, Klement emphasized that it is difficult to obtain

explainable relationships between concentrations in surface air and other

environmental media each as pr~eipitation, soil, sad biological systems.36

Some qualitative information is being developed; for example, Pelletter,

Whipple, and Wedlick have attempted empirically to fit measurements in

the Lake Erie area with a model of the form:

D t H(a-a-br)

where D = deposition, picocuries per square meter, (pCi/m2)
CA = air concentration, pCi/m
H = height of fall constant, m,

a = dimensionless factor to account for dry deposition
into the precipitation collector,

b = elimination constant, (inches of rain)-1, and

r = rainfall amount, inlches.41

Further study is necessary before the various relationships can be clearly

defined. Air concentrations and deposition rates show a peak during

the spring of~ each year which is related to entrance of debris from the

stratosphere during late winter, spring, or early summer. After new

tropospheric injections, peaks in concentrations are seen along the path

of the cloud, and air measurements are particularly useful to determine

arrival time of new debris at specific locations and thus anticipate the

rise in levels in other media. Atmospheric contamination of fallout

1370 is not considered an inhalation hazard.3

Radionuclidea in water

In his 1965 review, Klement also summnarized some of the observations

of fallout radionuclides in water reported up to that time.36 He stated

that short-term variations in deposition are probably not as detectable

in fresh surface waters as they are in surface air, but that the same

seasonal variations are seen. Relative concentrations of nuclidea in

surface water vary with local deposition rates and characteristics of

the local aquatic environment. Large differences can be expected within

a major river or lake because of runoff and flow.

Langham and Anderson concluded that drinking water contarmination

with fallout 1370 does not constitute a health hazard.3 However, water

may have an importance in the food chain in that 13Ca is more available

to organisms living in aquatic environments than in terrestrial


Cesium-137 and other radionuclides in the soil

Schulz reviewed soil chemistry in 1965.3 He listed fission

products likely to be found in the soil, classified elements according

to their mobility in the soil, and discussed the use of soil chemistry

in predicting. plant uptake of radionuclides.

Cumurlative levels of radionnelides in the soil and their distribution

in depth, important factor in plant uptake from the soil, are affected

by weathering and other disturbing influences.36 In most soils in the

United States, 90Sr does not appear to leach very rapidly (supporting

the validity of soil sampling as a means of determining accumulated

fallout deposition). Virtually all of the nuclide is found in the upper

8 inches of these soils. In medium to fine textured soils, about 75 to

80 per cent of the deposited 90Sr is found in the upper 2 inches and

even in medium textured soiled similar proportions are found in the upper

4 inches.

A number of author have cited the fact that cesium is readily

retained by soils.343,34 Its sorption is characterized by high

adsorption onto mineral particles (the ion exchange properties of soil

being largely associated with clay minerals in the soil complex), but

there is a certain amount of disagreement over the mechanism of this

fixation. Nearly all of the 1370 is found near the surface (the top

2.5 centimatera) of undisturbed soiled in the United States.343

Downward movement of cesium in undisturbed soiled is even less than

strontium. After five year, 50 to 80 per cent of added 13Ca remained

in the upper 5 centimeters with little penetration below 10 centimeters.23

It is pertinent to the Florida soil ecene that, in batch experimental,

sandy loans and clays absorbed 90 per cent of the available cashum,
while sand absorbed only 50 per cent.34

Ranges of ph commonly encountered have little effect on cesium

sorption since it is approximately constant above a value of 3.5 (though
decreasing rapidly below p 3.5).3

While easium is desorbed by cations of neutral ealts, more readily

by K+ than by Na+ or Ca +, it effectively compete with potassium in
interactions in the soil.3 Trace amounts of cesium adeorbed by the soil

are held more strongly than strontium. Russell etatee that the extent of
Cs fixation in soil depends upon the clay content and is inversely

related to the organic matter in which ceeiumr remained less strongly

bound.3 Schulz states the chelation of nuclides by soil organic matter

may cause them to keep some mobility in the soil.43

Complementary ions have a strong influence on the redistribution of

nuclides in the soil profile.43 For example, it is expected that the

amount of soluble calcium will have a direct effect on the mobility of

Srfl and RaC in soils and consequently on the eaee of uptake by plants.

Alk~ali cations present a more complicated case.) With regard to
Cs, complementary tone can be classed into two group: (1) those
having little effect on ceeiurm, oneh as dilute If the alkali cation Na ,

and the alkaline earth Ca >@;**, and Ba+ which solubilize or release

Lase than 10 per cent of added carrier-free 13Cs and (2) those liberating

moderate amounts of bound ceeiumr, such as Ce NRq K ~, and Rb which

free 20 to 80 per cent of the cesium to a labile form, Typical agricul-
++ ++I
tural soil in the neutral pR range possesses primarily Ca and Mlg as

labile ions. These have little effect in releasing added 1370 for

plant uptake.

Movement of cash~m-137 and other radionaclidea from the soil to plants

A number of experiments in nutrient solution indicate that cations

apparently compete against cseium in uptake by plant rooted. In one =

experiment at low external centum concentrations, potaeeina, rubidium,

emrmonia, and cesiumn were markedly effective in depressing 1370 uptake by

barley roots, while sodium, lithium, calcium, and magnesium were not; at

high concentrations, all inhibited 13Co absorption. Millet, oats, buck-

wheat, sweet clover, and sunflower discriminated slightly against rubidium

and strongly against ceeium in favor of potassium. In another experiment,

ani incraeae in the potassiumn of the solution decreased 137Ca uptake by

bean' plants but resulted in an increased potaeeiurm content of the plants.

At low potassium levels, barley showed high selectivity for potassium

relative to cesium; ~however at high concentrations, uptake was lees

selective' in favor of potaesium. On the other hand, other investigators

reported that adding potassium had only a mall effect--a 100-fold increase

in potassiurm reduced 1370 uptake by only a factor of two and non-

radioactive cesium enhanced 137Ca urptakre.45

Uptake by plants from soil is influenced by both the availability

of the element in the soil and the inherent behavior of the plant with

regard to available elements. According to Mensel, plant concentrations

Sof various radioactive elements, after they have been added to the soil

in water-soluble formsB, may be several orders of magnitude higher ~or

lower than the Concentration in the soil. He classifies elements into

five classes of uptake, ranging from strongly concentrated to strongly

excluded by assuming that the behavior of radioactive elements in the

soil-plant system is identical with that of stable nuclides of the same

element. In general, those nuclidea that are most readily absorbed are

soluble in the soil or are isotopic with elements that have metabolic

functions in the plant. Conversely, those that are least absorbed are .

quite insoluble in the soil. Notable examples are strontiurm, lightly

concentrated; radium, not concentrated; and easium, lightly excluded

in uptake from the soil.

Removal of 1370 from the soil is inhibited by the strong fixation

in the soil but uptake varies with soil type.2343

In typical soils of the temperate regions, 1370 was taken up about

one-tenth as much as 90Sr, with this ratio decreasing with time to about

one twenty-fifth or less after about three years.23 This sugg~ets more

than one cesiumn storage compartment in the soil. Lees strong binding in

organic matter has already been mentioned, and observations suggest that

a high level of organic matter in the upper layer of some permanent

pastures might reduce binding of cesium by clays, thereby enhancing and

extending its availability to plant rooted to one or two yeare.233 In

certain tropical arene Ca was found to be more available to plants

than had been expected.23

Schulz reasons that complementary ions, in strongly affecting

mobility of ions in the soil as outlined in the preceding section, will

in turn influence uptake by plants. In modern agriculture, nitrogen,

potassium, and phosphorus are commonly added to soils. From the

complementary ion effect, it would be expected that addition in the forms

K+ and NR{+ would liberate considerable amounts of 137Ca into a labile

form for subsequent plant uptake,.while the application of nitrogen in

the 803~ form would have little effect on the fixed Co. In an

erelen ySfUE bt + sa 1+ 137
exprimnt y chuz, ot K nd R4increased the Co uptake by

romaine lettuce but the NH{+ had a much greater effect, than K This

was explained by the hypothesis that after the NH4hdreesd ie

1370, apicrobial oxidation of the oxcess N1 oN3~rmvd tfo

competition with the now labile 13Ca for plant uptake.

Davia agrees that uptake of Ca from the soil is increased by the

addition of cesium and cites this as evidence that the capacity to fix

cesium in the unavailable form is actually very small. He cites a

number of experiments, however, where the uptake of Ca by lettuce,

greas, alfalfa, corn, millet, and wild plante was decreased by the

addition of potassiumr or was inversely related to the exchangeable level

present. In another experiment, added potassium resulted in a decrease

in the 1370 uptake by Ladino clover and, as a corollary, the uptake

increased as soil potassium was reduced by continued cropping.5 In this

latter experiment, stable cesium added to the soil, even to toxic levels,

caused increased 1370 uptake in this crop.

The collected evidence from the studies cited in this section seems

to indicate that complementary ions play a role in releasing fixed cesium

from the soil, and that stable cesium added to the soil enhances the

overall uptake of Co by plants, but that the effect of stable cesiurm

on plant uptake from available 137s and the effect of potassium on the

uptake from the soil are not clearly understood.

The effect of soil moisture content is apparently not clear. In

studies with beans in soil, limited soil moisture resulted in greater

1370s uptake than writh ample moisture.45 On the other hand, Pendleton

speculated in 1962 and ~again in 1967 that because of the high availability

of the nuclide to aquatic and emergent plants, milk and meat from grazing

animals utilizing forage from wetlands may contain much more of the

nuclide from fallout than such products from drylands, and that the

geographic variations observed in 13Cs content of ailk may be, in part,

a result of different levels of uptake by plants of vet sad dry lands.424

Behavior of ceeium-137 and other radionuclides in plante

Ceeiurm and potassium follow similar paths in physiological processes

but are not used in an equally competitive manner by plants.3 Trans-

location of cesium is more rapid than that of other fiesian products.

Tissue distribution of this element varies with species. When available

during active vegetative growth, it goes to the leaves and flowers with

smaller amounts going to the seeds. Accumulation of 13Ca by plants under

natural conditions varies considerably among apecies and with environ-

mental conditions.

In 1959, Langham and Anderson speculated that 13Ca was entering the

biosphere via direct contamination of the vegetation rather than via

soil contamination and plant uptake.3 It was soon determined that trace

amounts of ceesium can be readily assimilated by direct foliar deposition.34

Russell cites,as well documented, and other author agree, that direct

contamination of plants rather than absorption from the soil is the

dominant uptakre process for major nuclidea, particularly during periods

of high fallout.35,23,34 Evidence for this includes the fact that

vegetation 13Ca levels follow seasonal variations rather than being

accumulative, with amounts in plants being comparable to deposition

during the period of foliage existence.3 This observation is also

consistent with a large discrimination against soil to plant transfer.

Comar refers to foliar absorption of 90Sr.23 Russell states that

it is necessary to consider both direct foliar retention and "plant base

uptake."3 He cites evidence that much of this nuelide found on edible

leaves is actually absorbed at the basal region in the mat of prostrate

stems and surface roots and then transported upwards. In 1964 and 1965,

Wykee observed a reservoir of radioactivity in the vicinity of the plant

base in Upper Midwest grassland when he observed more 137Co and 90Sr per

unit masse in the root-mat traction than in the soil or grees.4

According to Russell, the amount of foliar deposition is comparable

for 1370 and 9Sr, but absorption from the soil is responsible for a

smaller portion of the 1370 reaching edible tissues than for 0Sr.3

He also states that in times of relatively low fallout, the ratio of

137Co to 90Sr in herbage and milk has decreased. This is evidence for a

greater rate dependence relative to the cumulative dependence for cesium

than for strontium.

Comar suggests that ainee milk concentrations of 13Ce (related to

levels in pasture grsas) have not followed the fallout rate in some areas,

it must be assumed that there are also mechanisms for accumulation of

previonely deposited 137Ce.23 This would be consistent with the earlier

mentioned less tightly binding compartment in the organic matter in the

soil. Davie predicts that when the stratospheric load of 1370 becomes

nearly depleted, soil will than become the primary route of uptake.34

Direct floral contamination is reported as a dominant process for

90Sr absorption in cereals but is negligible with storage orgman which

grow beneath the soil since there 18 little downward translocation of

strontium and other divalent catione in plants.35 Comar reviewed reported

that 0Sr accumulates in cabbages mainly by the lodging of rain in

cavities at leaf axile and in potato tubers by the downward leaching

by rain over the stem surfaces. The latter represents a means by which

underground tissues can become contaminated from deposition.2 Neither

author makes reference to mechanisms of 13Ca accumulation in these

crops .

Johnson and ward reported the 13Ca concentration in grain to be

much lower than that in hay.8 Elder and Moore and also Porter et al.

reported that pangolagraea (Dgitri documbene) and Spanish moas

(Tillandata ueneoides) had 13Ca concentrations much higher than any

other plants in the Tampa, Florida, area.98 They concluded that the

nuclide was surface contamination or weakly bound on the grass but was

actively bound on the mose.

A number of author refer to the use of the equation:

C I pd'd Pr r

with regard to radionuclide levels in plants. Element cautions that: it

is an empirical formula and does not necessarily imply understanding of

the mechanisms of transfer into and within a system.3 He makes note of

the fact that pd approaches zero for some plants such as Spanish mosa and

some lichen eand that pr is near zero for plants sheltered from appreci-

able deposition on foliage. A modified form has been presented for 137Ca

in which F2c and p2c are substituted for Fd and pd, respectively, thus

relating concentration to deposition accumulated over the two previous

year rather than to the entire history of accumulated deposition.23

This form presumlably relates to 13Ca released by decay of organic

material and to the less tightly bound deposition in the organic layer

and considers as negligible any uptake of the nuclide tightly bound by

the mineral fraction of the soil. Russell also cautions about~the empirical

nature of the expression and emphasizes that aside from the large errors

normally inseparable from field investigation, deduced relations relate

only to climatic conditions and patterns of deposition which prevailed

during the surveyed or field experiments from which the proportionality

factors were derived.3

Cesium-137 and other radionuclides in animals

Pendleton et al. have briefly summarized literature relating to cesium

and potassium metabolism by aninals.5( Absorption of both cesium and

potassium by mammals is large and takes place through the digestive

tract. These elements enter the body cellsw'heze cesium is more tena-

ciously retained; excretion from the body is urinary and in varying

degrees, depending on species and diet, fecal. Increasing potassiurm

intake has only slight effect in decreasing body cesium content.

According to Davie, the discrimation factor from foodstuffs to

animals (or man) should favor cesium due to preferential assimilation of

trace amounts of cesium relative to potassium.3 Pendleton et al., citing

a number of studies and authors, report that in many animal species the

1370s/potassium ratio in the body is about two to three times higher than

the 137s/potassium ratio in their normal diets (unlike the values

observed for the 90Sr/calcium ratio in animals which are less than one).50

Comar reiterates that 1370 is efficiently absorbed from the gastro-

intestinal tract for transport to musacle tissue and milk.23 .He cites

literature values for concentration in the muscle tissue of cattle of

about 4 per cent of the daily intake per kilogram and values ranging from

10 to 30 per' cent for sheep and pigs.

Hard and Johnson quote literature values for cattle which indicate

that 1 to 2 per cent of the daily ingested dose appears in each kilogram

of muscle tissue,5 Their own work showed 137s levels per kilogram of

edible meat leveling off at less than 1 per cent. of tue daily intake in

mature dairy cattle, at 3 per cent in feed-lot cattle, and at 15 per

cent in calves. These differences may have been due either to age or

feed type, Kreuzer reported a close correlation between 13Cs in meat of

cattle and the methods of feeding and management in Southern Bavaria.52Z

The muscle of grazing cattle vae more heavily contaminated than that of

housed cattle during the grazing season. Change from housing to grazing

was reflected in a rapid increase in levels.

K~ahn et al., in a study of the effect of feeding practices on 137Ca

concentrations in cow's milk, and porter et al., in a study of the Tampa

milkshed, reported that 12 per cent of the daily intake of 13Ca was

tranafarred! to the daily milk production when the animal is in equilibrium

with its diet.58

Eatimatee of the biological half-life of ceeium in the cow, the goat,

the pig, and the hen fall in the range of 20 to 30 days although some

workers report a value as short as 2 to 3 days for the goat.23 In any

event, the turnover in these animal is rapid enough for tissue levels to

show little lag in following dietary levels. Retention of 1370 in the

body is affected by potassium and sodium levels as well as by diuretic

action, but these effects appear to be variable and relatively small.

Russell noted that an error in the eatisate of expected milk levels

of 908r vae systematically related to the particular fallout history of

the preceding year.35 Taking into consideration the use of stored hay as

feed early in the year, he introduced an additional set of terms into the

concentration equation:

C = prF r PdFd 1 P1F

where F1 is the strontium deposition during the second half of the

previous year and pl the "lag rate" factor. With this three-term

expression, he obtained a much better correspondence of predicted values
to observed values.3

Pendleton et al. etate that 1370s is even more available to organisms

living in aquatic environments than in terrestrial environments, and,

because of the trophic level effect already cited, predacious fish may

accumulate this nuclide up to 10,000 times the level in the water.5

Comar reports that 1370 is concentrated in the flesh of aquatic organisms

by a factor of 5 to 20 over that of the surrounding water and Gustafoon

reports high concentrations of 13Ca.in fresh water fish.2545

Casium-137 in man

What has been said about cesina metabolism and distribution and about

the increase in the 1370s/potassium ratio with the trophic level in animal

applies generally to man as well.345 The data of Pendleton t al. indicate

that the increase ratio is larger in adult humans than in children.50

Cesium~-137 retention in humans has been described as multi-

compartmental function, but many workers have found or believe that a

single exponential is a satisfactory expression for most biological

purposes.5,78 Reported half-lives for the longer-lived component in

adults are on the order of -80 to 140O days.

In 1962, Richmond gg 41 measured whole-body retention in four

subjects and reviewed the literature.56 Hardy et al. reported in 1965

on a study of retention of 13Ca and90rolwngcueneso.7

In this study, contaminated foods (1370s and 90Sr, 20 and 60 times higher,

respectively, that in the normal diet) brought back from Rongelap were

consumed by a single individual over a seven-day period. In 1965, McGraw

reviewed the collected measurements of the half-time of this nuclide in

man and suggested a model for expreeaing this parameter as a function of

age.58 Eberhardt reported a different approach in 1967.59 Using

published data from a variety of sources, he derived an expression for

137Cs biological half-life in humane as a function of body weight.

The maximum permissible body burden for 1370 for the occupationally

exposed, as recommended by the International Commission on Radiation

Protection, is 30 microcuries (pWi).60 If one: (1) applies a factor

of one-tenth for members of the general population, (2) assumes that cesiumn

retention can be described with a single exponential function with~half-

life on the order of 100 to 150 days, (3) aeaumes an equilibrium situation

of constant intake, body burden, and excretion, and (4) follows the

Federal Radiation Council reco~mmendation of aeausing that the maximum

exposure by an individual is no greater than three times the average

measured for a population group, then one can derive a maximum permissible

daily intake on the order of 4,000 to 7,000 poi/day. A value of 4,400

pCi/day has been used in the literature.616

Equations have been presented relating dose to 1370 intake and a

number of authors have attempted empirically to fit models relating

measured body burdens to measured levels of 1370 intake and thus predict

future body burdens from intake data.9636

Arctic ecosystems

In 1961, Baarli et al. and Ilden. found unusually high 13Ca body

burdens in people in Norway and Sweden, respectively.676 These initial

findings led to further studies which enggested that unique ecological

situations exist with regard to radionuclides in the fallout-food-man

chain in arctic regions.69-72 Although these regions have lees

fallout than many other areas of the world, levels of 137s in arctic

inhabitants are among the highest documented.

High human body burdened have been traced primarily to the influence

of ecological factor upon fallout accumulation in the lichen-caribou(or

reindeer)-man chain.737 Lichens represent a most important reservoir

of 1370 and other fallout radionuclides because of their longevity

(decades, even one century), persistence of aerial parts, and their

dependence upon nutrients dissolved in precipitation. This effective

retention of 1370 by lichensr, the importance of lichens as a winter food

for caribou, and the dependence upon caribou and reindeer for food by

many northern peoples has resulted in this important 13Ca concentration

In the United States, studies of radionuclides in the arctic food

chain were addressed initially to questions associated with the proposed

Project Chariot site located near Ogot~oruk Creekr in northwseern Alaska.l

Since 1963, surveillance activities have continued and expanded in

Alaska.798 The Battelle-Northwest Laboratory has studied the body

burdens of 13Ca in northern Alaskan residents especiallyy in Anaktuvuk

Pass) since the summer of 1962,85

In April and May of 1965, the USPHS expanded the geographic area

investigated by measuring 3Ca body burdens throughout Arlaska.7

Radiation dosage received by the average adult Anaktavuk Pass Eskimo

from 1370 body burdens during 1964, the year of highest values, was

estimated at 135 to 150 moillirem (marem).73 This value is about 30 per

cent less than that amount serving as a Radiation Protection Guide for

population groups during normal peacetime.

`Envirdnaerltal 'IRadioactivlity `Measurements in Florida

Environmental radioactivity sampling and measurement within the

State of Florida began with the collection of sample in 1957 by the

Florida State Board of Health, Bureau of Sanitary Engineering.8 This

soon led to a statewide program of sampling and grose alpha and beta

radioactivity analyses of surface and underground water resources, sedi-

ments, raw and finished water from water supplies, and sewage plant

influents and effluents.2- Some soil, aquatic organisms, and rainfall

vere also included in this early sampling. Gilcreas, Mbrgan, and Vreeland

reported the establishment in 1957 of a routine program involving regular

monitoring of grose alpha and total radioactivity of air and rainfall

at the University of Florida and of surface and underground waters in

Alachua county.8889 The results of these sampling program are

reported in a number of publicatione.2-,69 The University of Florida

program also included twice yearly sampling and grose alpha and total-

activity analyses of plante, animal, and soil.8888

As qa cnsequence of an agreement with the USAEC, the USPHS establish-

ed the Radiation Surveillance Network in 1956 with nationwide stations

for sampling of dry deposition (later discontinued) and air.919 In

1957, the network was expanded and precipitation sampling was added. One

station was designated at Jackeonville under the operation of the Florida

State Board of Health with five to seven continuous precipitation and 24-

hour air sample collected per week. In 1961, a second station was

established at Miani.

In 1960, the Florida State Board of Health estabilished~ similar air

and precipitation easpling stations at Orlando, St. Petereburg,

Tallahassee, and Peneacola; the six stations became known collectively as

the Florida Radiation Surveillance Network.2 Florida data from the

state and national networks are reported in the various iaanes of Re221

o~f Florida Radiological Data.2- In 1963, air sampling station were

aleo established at Tituaville, Cocoa, and Melbourne in the vicinity of

Cape Kennedy.3

In 1953,the USPHS established the National Air Sampling Network--

a survey of airborne particulate pollution, includinB gross beta radio-

activity, in urban and non-urban areas in the United States.9 One

24-hour sample is ordinarily taken every two weeks; sampling frequency

was increased at many station during time of anticipated increased in

airborne radioactivity due to fallout. Stations are designated for

operation either yearly or during alternate years on a staggered schedule.

During the period 1959-1964, stations operated every year both at Tampa

and in the Florida Keye and on more selective schedules in Jacksonville,

Niasi, Orlando, and St. Peteraburg. Reenlts were regularly reported in

Radiological Rlealth ji.E28 Precipitation sarmpling was added at designated

stations with the cooperation of the United Statae Heather Bureau in 1959.

These 'included a station at Tampa beginning July, 1960.

A station of, the 80th Meridian Air Sampling Program was established

at Miami in June, 1957. This network.~of stations near the 80th meridian

(west) was established in 1956 by the United States Naval Research

Laboratory with the assistance of the USAEC, the United States Heather

Bureau, and interested groups in Canada and South America. The stations

sampled ground-level air for measurement of radioactivity neing a variety

of asperiOmenal collecting devices. In 1957, all stations were supplied

with filter samples and blower units, and the network was expanded to

Include locations of interest.to the United Statee Air Force Cambridge

Research Center. It was operated in 1958 and 1959 as part of the

International Geophysical Year Program on Atmospheric Nuclear Radiation

and continued from 1960 through 1962.vith a reduced number of sampling

aites.949 Individual eamples were analyzed for grose beta radio-

activity and samples were composite monthly for analysis for a variety

of specific nuclides. Reports of results and 80th meridian concentration

profiles from 1959 an appeared regularly in Radiolog~ical Health Data.95
Lockhart y). al. summarized the results for 1957-1962 and included a

complete list of reports (26 references) resulting from this program

during that time period.9,9

The USAEC Health and Safety Laboratory (HASL) deposition sampling

program, in operation since 1958, employee two methods of collection for

90Sr analysis.97 In the "pot" method, precipitation and dry fallout are

collected for a period of one month in an exposed etainlese steel pot and

transferred to a bottle for shipping, while in the columnm" method pre-

cipitation and dry fallout are collected in a polyethylene funnel

connected to an ion exchange column which is capped for shipment after a

one-month collection. A station utilizing a pot collector is in operation

at Coral Gables.

Operational responsibility for the 80th Meridian Air Sampling Progrea

was transferred from the Naval Researdb Laboratory to RASL as of

January 1, 1963, to be run in conjunction with the fallout sampling

program.98 Caneurrent with this transfer, gamma malaees, both total-

gamma activity and activity above 1 million electron volta (MaV) energy, were

adbetituted for gross beta analysis as the initial means of sample

evaluation. Ion exchange collectors for deposition sampling were added

to the stations not having them. This resulted in the initiation of a

column collection station at Miami in July, 1963, in addition to the

Coral Gables pot station. Results of the 80th Meridian Program since

1963 and of the depoettion sampling are reported in RASL techateal

reported, and summaries have been published periodically in Radiological

Health Data.2

The USAEC and the United States Department of Agriculture have been

conducting soil sampling,first individually and then jointly,ainee 1955

to determine world-wide distribution of 90Sr.99,100 During this course

of sampling, collections have been made at both Mand. and Jacksonville.

The Appalachicola River at Chattahoochee and the Eacambia River at

Century have both been sampled under the USPHS National Water Quality

Network (cuzzently the Federal Water Pollution Control Administration's

Water Pollation Surveillance System).10 This program was established in

1957 for sampling surface waters of major United Statse river basin for

physical, chemical, biological, and radiological analyses. Weekly grab

samples are collected for analysis at a central laboratory for gross

alpha and gross beta activity in suspended and dissolved solide. Analyses

are performed on either weekly samples or monthly composites, depending

upon the amount of radioactivity expected. Stron~tium-90 analyses have

been performed on selected quarterly composities from each station since

1959. Reenita have been reported for the Appalachicola River For 47

one-month period between April, 1960, and June, 1965, and for the

Escambia River for 39 such periods between May, 1961, and Mhy, 1965.

In 1961, the Water Supply Activity, Interstate Carrier Branch,

Division of Bagineering and Food Protection of the USPHS, established a

Drinking Water Analyasi Program to gather extensive data on the radia-

activity content of water enpplies on interstate carrierel02t Results

have been reported for two samplings each at eva water plants in Fort

Lauderdale, two water plants in Miami, and a plant in Tampa during the

period, 1960-1963.

In addition to the studies of radionuclides in milk already

mentioned, a special study conducted by Consurmere Union in July through

August, 1958, included Miami among 50 North American stations from which

milk samplee were collected for 9Sr and stable calcium analysis.10

In describing shellfish radioactivity research, Lackey reports

the total background activity of oysters collected at two locations in

Florida in March, 1958.10 Gross alpha and grose beta determinations and

gamma apectra vere reported for analyses performed by the USPHlS on oveters,

fish, shrimp, crab, and other biota, as well as on salt water and silt

collected periodically during January, 1962, through January, 1963, from

several locations in Brevard County, Florida.

In summarising results for a variety of food samples, Straub et al.

reported 90r and table calcium levels for one sample each of radishes,

turnips, and trnaip greens from Florida, all collected from Cincinnati,

Ohio, market in 1959.105

In 1959, 1961, and 1962, the Miami area was included in the USPHS-

eponsored Consumers UI~on stuies of S0r, 13Cs, and other radionaclidee

in a typical diet of teenagera.10,0 These studies were carried out

periodically between 1959 and 1964 in a number of cities in the United

Florida easples have been included in the United States Food and Drug

Administration's studies of radioactivity in domestic and imported food

and animal fodder.2,,0

Fifty-six Florida vegetable and citrus samples, collected January

through March, 1964, by the Florida Department of Agriculture, were

analyzed for specific nuclidea by the Florida State Board of Blealth.4

The USPHS Institutional Diet Sampling Network, established in

January, 1961, with 8 locations, was extended to 21 cities, including

Tampa, in October, 1961.10,1 Twenty-one meals plus snacks (three

meals per day for seven days) are collected each month at a participating

school or other institution in the respective cities. Solid food, dairy

products, and wastes are analyzed as separate portions at USPRS regional

laboratories. Reenlts are reported in terms of weight consumed per day

and concentration and daily consumption of calcium, potassium, phosphate,

strontium-89 (89Sr), 90Sr, iodine-131 (1311) 1370s, barium-140 (140Ba)

and either total radium or radium-226 (226Ra). These results appear

periodically in Radiological Health Data.2
In the previously mentioned study of background radiation in Florida,

Willians eg al. reported levels of natural radioactivity in the air from

three Florida cities during 1963-1965, levels of grose alpha and grose

beta radioactivity in food collected in 1964 at various places in the

State, and'groes alpha activity in ash from the 1963 samples of the state-

wide milkr sampling network.29

In addition to the special studies in Florida milkcsheds already

mentioned, Cromroy at al. evaluated the exchange characterietice of soil

from a dairy farm near Tampa and measured 137s uptake from this soil by

diferet rasee.11They reported: (1) a higher 137s level.in

pangolagrase growing at the Teapa farm than in samples from the

Agricultural Experiment Station at Gaineeville, (2) that the clay fraction

of this soil was kaolinite, which exhibited ion exchange and absorption but

has a layer spacing not favorable for "structural absorption," (3) that

the uptake of 137C by this soil in a slurry test was much less than that

reported for other soils, (4) significant uptake from the soil by grasses

transplanted to this soil, (5) different rates of uptake with Bahia-

grass, reaching an equilibrium level more rapidly than pangola, and (6)

that the more slowly equilibrating pangola showed an indication of

progressing toward an equilibrium level about 30 per cent higher than that

of Behia. The thast sampling of a third grass, Bermuda, showed levels

that indicated either a much lower equilibrium level than the other

grasses or a slow rate of uptake. However, retarded growth prevented

further sampling. Unfortunately these authors did not report radioactivity

levels in these graseae prior to treanplanting.

Reported of monitoring sever effluent, air, surface water, and milk

for tritina in the environs of the USAEC's Pinellas Peninsula Plant near

St. Paeereburg have appeared periodically since 1960.30 In 1965, tritium

level were also reported for precipitation collected at Ocala during 1961

and 1962 as part of a global tritium distribution atndy.11

Hartgering described a worldwide sampling program in 1955 and 1956

to estimate human exposure by analysis of 24-hour urine specimens for

isotopes of iodine, strontium, and eesium.11 Participants were five to

ten personnel at each of a number of military installations, which included

McDill Air Force Base, Tampa.

Radioactivity of humans has been measured at a number of installa-

tions. In some of the published reports of 13Ca measurements in humans

at the walter Reed Army Institute of Research, average concentrations have

been tabulated by etates.11,1 The average of six detenninations on

Florida individuals during 1958-1959 was not noticeably different frost

averages in.other parts of the country at that time.11 In addition,

the average 13Ca levels and number of subjects were reported by quarter

for Florida subjects for eix quarters during 1960 and 1961 (a total of

eight Florida subjects).11 Seven Florida subjects were also identified

as haviing been included in the United States averages for July, 1963,

through Auguet, 1964.116

Ceeiump-137 measurements have been made in humana at the Univeretty

of Florida, but results have not yet been published.11 It vae announced

to the press that a whole-body counter survey wee to be made of residents

in the Tampa institution participating in the Institutional Total Diet

Sampling Program, but no published reports were found to date.11

Undoubtedlyraubjects from Florida have been counted at other installations

even though specific Florida results have not been reported in the litera-


Current environmental radioactivity measurements in Florida

Current sampling in Florida for radioactivity analyses is shown in

Figure 3.

At the present time, grose alpha and total radioactivity measure-

ments continue to be made regularly at the Univeretty -of. Floride on

semples of air, precipitation, sewage, water supplies, and surface water.119

~Moitoring also continues routinely on a statewide basei.114 The Tampa

Pasteurized Milk Network Station collects weekly milk easples; the two

Radiation Surveillance Network Stations at Jacksonville and Miami continue

regular operations. The Florida Radiation Surveillance Network Stations

continue to collect air and precipitation samples, and the State labora-

tory analyzes these samples for grose beta activity. The statewide grab

sampling of enrface and underground waters and of water supplies continues

AlachuB County Surveillance:
(University of Florida)
Air and precipitation (one station),
surface water, water supplies, sewage

Radiation Serveillance Networks
SAir and precipitation

Florida Radiation Surveillance
O Air only
Air and precipitation

AI Cape Kennedy Vicinity: Air

SPaeteurized Milk Network

X RASI. Fallout Program:
Deposition, pot collector
Deposition, column collector
Air (80th Meridian Station)

Regions indicated are for Florida
raw milk sampling

Not shown
Statewide water sampling
Shellfish sampling

Turkey Point Vicinity:
Air, water, soil,


for grose alpha and beta analysis. Grab samples of shellfish are

submitted to gamma-spectrum analysis, and then they are ashed and analyzed

for grose alpha and beta activity. Raw milk samples are currently being

collected on a statewide basis for combining into six regional composites

for garmmeaspectrum and 90Sr radiohepemica analyses.

In addition, several special monitoring projects are underway in

the State.12 A cooperative plan has been drawn up by the United States

Air Force, the USPHS, and the Florida Stata Board of Health to collect

environmental samples in the vicinity of Cape Kennedy for radioactivity

analysis. Air is being sampled for gross alpha and beta analysis at Fort

Pierce, Orlando Air Force Base, McCoy Air Force Base, and New Smyrna

Beach; the three previously mentioned state-operated stations at Titusville,

Cocoazand Melbourne function as part of this program. Soil, edible

vegetation (including citrus), and non-edible vegetation are being sampled

regularly and submitted to gamma spectral analysis, while water is collect-

ed and submitted to gross alpha and gross beta analyses of both the

dissolved and suspended solids. Seaples from the 12 eites within the

first 10-mrile radius of the area are analyzed by federal laboratories;

samples from the 10 stations in each of the two 10-mile annuli beyond

this are analyzed by the Florida State Board of Health Radiological

Laboratory .

A special sampling program is also being carried out by the

Florida State Board of Health near the site of the planned nuclear power

pleat at Turkey Point, a facility due to go into operation in the early

1970's.120 Eighteen sites ear sampled on a rotating basia; aix are

sampled per month so that each eite is sampled quarterly. Analyses are

similar to those for the Cape Kennedy off-site area.

Results from the Miami and Coral Gables stations of RASL's air

and fallout sampling program were included in the latest reports of

these program, and results were reported for the environment of the

Pinellas Peninsular Plant of the USAEC as recently as February, 1967,

implying that both these program are currently in operation.0957

Cesiurm-137 and Other Radionuclides in Selected Environmental
Media Methodology and Findings

As will be discussed in more detail in Chapter III, the earlier

observations on radionuclides in Florida milk and feeds and the preceding

review of cesium ecology and Florida radioactivity measurements led to

the selection of meat and vegetables as the major items of study with

nuclide emphasis on 137Cs.

Although thief study is intended to investigate levels and patterns

of radioactivity in specific media, the ultimate significance of the

findings lies in the influence of the observed levels on human radio-

activity intake. Methods of evaluating dietary intake as reported in the

literature can be grouped into three general categories: (1) evaluation

of individual food items end, if all components of a total diet are

evaluated, subsequent computation of total dietary intake, (2) analysts

of the composite total diet of a real or hypothetical individual, and

(3) analysis of urine and feces.1122230

Casium-137 in individual food items

Meat sampling nonally involves selection of some particular muscle,

portion, or product for analysis. It has been reported that the concen-

tration of 1370 can vary from muscle to muscle.12 Fredrikeaon et OL.

reported considerable variation in the rate of cesium uptake between

different muscles with the meet active muscles taking up cesium most

readily and the inactive muscles ultimately reaching a higher content,12

Ekrman reported that the half-time of 13Cs varies between tissues, being

longer, for example, in skeletal muscle than in most other organs or


On the other hand, in a study of uptake and fate of 137s in rats,

cattle, sheep, evine, and poultry, Hood and Comar report that the most

striking feature of cesium distribution in tissues vae the constancy of

the pattern, both between individual and species and between the various

tissues of the body.12

Data reporting environmentally occurring levels of 1370 in meat

or animals in the United States are limited; some of the first data was

from animals maintained in the vicinity of nuclear testing sites. Ceeium-

137 levels were reported for unspecified muscle from cattle sacrificed

in 1957 and 1958 from herds maintained in the vicinity of the Nevada Test

Site, and another reference vae made to cattle sacrificed before and after

the 1961 "Gnome" underground nuclear test in New Mexico.12,2

The United States Department of Agriculture has reported 1370 and

908r levels of beef rib meat surveyed in 1960 and beef soup stock surveyed

in 1961.13 The following year bologna and frankfurters were enrveyed

for 0Sr only, and the results were not significantly different from

those reported earlier in rib meat.

Plummer recently reported very little difference in concentrations

of naturally accumulated 1370 between various muscles of the Georgia

while-tailed deer; this prompted him to use the tongue as an estimator of

general body levels.13 Wettinen reported 1960 and 1961 13Ca levels in

reindeer, potatoes, and certain other food items of Finnish Lapps.72t Re

reports that there is an error of only a few per cent when analyzing the

shoulder of the reindeer as representative of all consumned tissues. He

estimates losses of activity in meal preparation to be about 10 per cent,

but he does not make a corresponding correction in his calculations.

In a' highly detailed study in the United States, Ward and Johnson

reported 13Ca levels in beef from dairy and feed lot cattle at Fort

Collins, Colorado, in 1963.51 They found no differences, wPithin limits

of counting statistics, in content between 12 retail cuts from the same

animal as measured an four different beef animals. They concluded that,

while greater precision in counting may have shown statistically signifi-

cant differences between areas of the carcass, these would be small

compared to the effect of differences in the 1370 contamination of the

various animals' diets.

Straub reported surveys in which individual vegetable crop samples
90 137
collected from market in 1958 were analyzed for Sr, Cs, and other

gamma-emitting radionuclides.10 Food categories reported include fruit,

fruit juices, meat, leafy vegetables, root vegetables, legumes and corn,

and rice.

Setter et al. reported an elaborately planned survey of individual

food items.13,3 Market sampling was performed simultaneously with a

food consumption survey for six quarters beginning July, 1962. Selection

of samples was based on varieties and amounts of food available at the

time of eampling as indicated by the United States Department of

Agriculture marketing reports and local marketing data. Categories

included meat, -vegetables, fruit, and potatoes. During the six calendar

quarters,~ ground chuck beef was sampled for four quarters and lean pork

for two, lettuce for four and cabbage for two, and apples for four and

oranges' for two; potatoes were sampled for four quarters. RLesulta for

137Cs and 90Sr were reported for each of seven regions and the Whole of

the United States. The lowest 1370 levels were found in lettuce,

cabbages, potatoes, apples, and ~egge. These authors stress that .identifi-

cation of food from production areas is difficult because of factors such

as: (1) centralized marketing, (2) raising of the majority of poultry

feed in one region, and (3) raising of beef in one region for fattening

in another on feed from a third.

Highlights of the United States Food and Drug Administration's

prognre of measuring radioactivity in individual domestic and taported

human and animal food items included: (1) a call in 1954 for food

packed prior to the nuclear era, (2) a subsequent program of grose radio-

activity analysts, (3) grose beta activity measurements on a broad basis

sines 1958, and (4) emphasis on 90Sr, 137Cs, and 1311 since 1960.108,134-138

Under this program, results of one 13Ca analysis and about 100 9Sr

analyses were reported for Florida samples collected in 1962 and 1963.2,,0

Laug reported that in this program samples were mostly raw agricul-

tural products, usually un~washed and unpeeled, and generally in the condi-

tion in which they were found in the warehouse or store.13 For the purpose

of data summarization, samples were assigned to classifications such as

vegetables, brassicae, root vegetables, white potatoes. corn, and fruit

(including tomatoes, cucumbbers, pumpkins, and squash, as well as the

bervice, citrus, and other tree fruits). Ceeiumn-137I level were generally

five times as high 'as 90Sr levels. Vegetables ranked in 137Cs content

from highest to lowest as follows: leafy vegetables, dairy products,

brassicae, root vegetables, fruit, corn, and white potatoes. Lang also

states that, although the first year following the resumption of atmoe-

pheric tests vae not marked by dramatic increases of 90Sr in foods,

sporadic higher concentrations were noted in leafy crops. Fallout con-

centrations were reported to be higher in the eastern and central United

States than in the West, an effect believed to be associated with rain-

fall. In an experiment with a leafy vegetable and one kind of brassicae,

commercal processing washingg and canning or freezing) reduced 1370s

by 20 per cent.

The Florida State Board of Health determined and reported 1370s

and zirconiurm/niobium-95 (95Zr/N~b) contents of 56 leafy, non-leaty, root,

brassicae, and miscellaneous vegetable samples and fruit samples that had

been collected by the Florida Department of Agriculture in 1964.4

Bruce reported the levels of 13Cs and 9Sr found in milk, meat,

root vegetables, and leafy vegetables in the United Kingdom during 1962-

1964 and in milk and meat during 1965.13,4 Cesium-137 levels in meat

from cows, sheep, and reindeer in Norway were reported by Hyinden for the

period 1959-1965 and by Madshus for 1964-1965.14,4 Madshus took the

precaution to collect all samples from the semi-tendinous muscle when

sampling meat.14 He also reported average 137C levels by zone for

potatoes and carrots for the fall of 1964. These two items were chosen

for sampling because of universal production. DeRuyter and Aten estimated

intake by humane in the Netherlands by sampling urine and faces.12 Be-

tween November, 1964, and Mlarch, 1965, they also sampled some individual

food items, including milk, grain products, kale, and brussela sprouts.

Uptake of 1370 by some leafy vegetables should be similar to grasses

and legumes. Porter et al, and Cromroy et al., in the studies reviewed

earlier, reported 13Ca levels in grasses and animal feeder collected near

Tampa, Florida.8,1 Caupka published 1370 levels for grass and alfalfa

in Western Slovakia during 1;962-1965r and Nykes reported levels of this

nuclide in grass in Minnesota in 1965.137

The highest 137s levels in lean tissue samples in the United

States are those reported by the National Center for Radiological Health

in a program of monitoring unspecified muscle from Alaskan caribou and
renerduig 18216.2 137
reidee duing196-196. Pendleton et al. reported Ca levels

that were measured in mule deer meat by gamrma counting the hip to knee

portion of skinned hind lege.5
Cesium-137 in the total diet

A number of total diet studies conducted in the United States are

pertinent to this research because they report nuclida levels in

individual food items or categories, illustrate how food items have been

classified into categories, or report effect of food on total intake for

the area and time represented.

In the Consumere Union total diet study, diet representing the total

diet of teenagers for a week in the particular sampling region were

prepared by home economists from produce purchased in local market, and

the composites were submitted for analysis.14

Reference has already been made to the Tampa station of the National

Institutional Diet Sampling Network.10 Ceaium-137 intake levels at this

station have been among the highest in the network, a condition which is

not inconsistent with high milk levels in that area and the fact tbat

dairy products are one of the major contributors of this nuclide. Reports

are available for the January through June, 1966, portion of this study.

Baratta and Williams have made comparisons between these two studies.14

They traced differences in observed level of intake to differences in

amounts of food consumed in each study.

In addition to the program of sampling raw tood items, the Food and

Drug Administration has a total diet study.146,147,148 Sampling in thia

program includes 82 food items in the 11 categories of the food plan

recommended by the United States Department of Agriculture as nutr-ition-

ally adequate at a moderate cost level for boys 16 to 19 years of age

(the highest intake group of the United States population). Items are

sampled in amounts proportional to consumption as reported in the 1955

Household Food Consumption Survey. Samples are collected at designated

cites (none in Florida) in the United States. The categories include

leafy vegetables, smooth vegetables, root vegetables, potatoes, dried beans,
fruits, and meat and eggs. The most recent report includes Ca levels

in the total diet at nine cities for February through November, 1965.

In March of 1960, the USAEC instituted a diet study in a single

city.14 This program was later expanded to three stations and it

became known as the Tri-City Diet Study. In this study of the typical

adult diet, foods eae grouped and analyzed in 19 categories with 9Sr and
Ca levels reported in each category for each city. Categories include

meat, poultry, fresh fish, shellfish, eggs, fresh vegetables, canned

vegetables, root vegetables, potatoes, dried beans, fresh fruit, canned

fruit, and fruit juides. At Chicago, analyses of these samples for Ca70

have been performed quarterly since 1961. At the other two stations,

these analyses were performed at the and of each year through 1964 and

then quarterly in 1965. Reports for each city include yearly quantity

intake, nuclide concentration, and yearly nuclide intake for each food

category and daily and percentage of total 1370 intake attributable to

the five overall categories of milk, meats, cereals, fruit, and


Thompean and Lengeman draw, attention to the fact that estimates

of quantity of food consumed do not match the precision of the laboratory

radianuclide measurements.12 By use of different available consumption

estimates, they show how estimates of intake based on the same laboratory

data can vary as much as 75 per cent.

In spite of this potential difficulty. Rivers reported that 1370s

body burdens calculated from intake data were in relatively close algree-

ment to measured body burdene.bl He suggested that diet analysis vae a

useful means of predicting body burden. He also stressed the need for

better analytical techniques and more frequent food measurements.

Using 13Ce/potassium ratios as a basis, Gustafoon examined the

relation between diet and human body burdens.5 Re concluded that diat

corresponded to in vive data when the former vae transformed by a factor

of 3.0 and advanced in time by four months. In the same report, he

extrapolated the human body burden to the end of 1966.

The factor of 3.0 used by Gustafoon to transform diet 1370,/

potaeeium ratios to correspond with body burden levels is consistent with

the trophic level effect discussed earlier.0 The four-month time

advance used in this transformation is probably related to time required

by the human body to equilibrate with changes in the diet and may also be

related to storage of food before consumption.

The Federal Radiation Council developed predictions of the expected

radionuclide levels in the diet in 1963-1965.394 Since then, measure-

ments of radionuclides in milk, the total diet, and hurmans have been

reviewed for the years 1963-1965 and compared to the Council prediction

estimates for these years.616 In addition, levels for 1966 were

predicted using the approach proposed by the Council. The observed values

were in ~agreement with the predictions; the peak of intake by population

groups appears to have occurred in 1964.

Other garmna-emitting radionneclides
In addition to 137Co and 408, food and vegetable samples are

commonly analyzed in the various surveillance program in the United

States by gapmma spectroscopy for any or all of the nnelides 1311

certurm-14r4 (144Ce), manganese-54 (54Mn), ruthenium/rhodium-106 (106Rou/RIh),

sine-65 (65Zn), ad 95Er/Nlb; by gamma spectroscopy or radiochemical

method for 140Ba or barianm/1anthanurm-140 (140Ba/La); and by chemical or
emanation method for total radium or 226Ra. 78,105,106,109,110

The Institutional Diet Sampling Network samplee are analyzed for

poteasium, 13Cs, total radium or R26a, Ba08, and 31.10 However,
in reports for 1966 the latter two were reported to be below detectable

levels. Zirconium/niobium-95 was reported for reindeer and caribou
muscle in 1964.82 Manganese-54, 65Zn, 106R~u/Rh, and ceriumr/praaeodynium-

144 (144Ge/Pr) were rported in lichens, moases, sedges, and other plants
in Alaska during 1959-1961.78 As was noted earlier, the Florida State

Board of H~ealth reported levels of 95Zr/Nb found in Florida vegetable

samples in 1964.4

The presence of 65Za,_a neutron activation product, was reported
in a variety of foods collected in 1958-1959 and attributed to high

altitude fallout by Murthy et al.1 It was reported by Van Dilla in
1960 in beef liver, beef muscle, hamburger, and milk.15

Manganese-54 was reported in human and bovine liver in 1965 and in
freshwater clama in 1966 by Sax and Gabay, and in the Netherlande diet
in 1964-1965.15,5,2

Antimony-125 (125Sb) was reported by Sveneson and Liden in forest

moss in 1965 and by Johnson et .al. in Colorado forage in the years 1962-
1965.15,5 De~uyter and Aten tentatively identified trace amounts of

both 125Sb and beryllium-7 in leafy vegetables in the Netherlands in


Cesium-134 was identified in 1960 by Krieger and Groche as a

neutron activation component of fallout from some weapons tests.157

During the period 1960-1964, it was found in air samples, elk, milk,

wheat, and beef in the State of Washington and in caribou and reindeer

meat, fish, and human inhabitants in the arctic.728,5,5 This

nuclide was thought to have wide distribution and to have been produced

prior to 1961. It was present at the level of about 1 to 2 per cent of

the 1370 levels.

Traces of sodium-22 (22Na) hav been reported in vegetables in the

conterminone part of the United States in 1964; in fallout, herbase, milk,

and total diet samples in Italy in 1962-1964; in elk, bass, milk, wheat,

beef, and human urine collected in the State of Washington during 1960-

1964;' and in moose, caribou, and reindeer meat as well as human urine and

Eakimoe in Alaska in 1963-1964.1810618 Perkins and Niaelen

suggest that a considerable amount of 22Npa va generaed during 1961-

1962.15 DeBortolli et al. ehow evidence that this nuclide is taken up

from the soil rather than being deposition rate dependent.16

The activation product, iron-55, has been reported in arctic and

subarctic food chains and residents, and in the blood of residents of the

State of Washington.768 The low energy X-ray, 5.9 thousand electron

volta (keV), emitted by this nuelide does not interfere with the spectra of

other nuclidea previonely discussed and probably would not be detected by

a conventional counting system used to analyse for these othere.85

Nuclear testing will add fresh fission products to those observed

in environmental samples. Klemoent reported that tranetent increases in

air and precipitation radioactivity were observed following the French

nuclear tests of February and April, 1960, but that debris from these

events added not more than0s.1-0.2per cent to the total worldwide fallout

of long-lived fission products.16 lodine-131 and 14Ba were found in

air, precipitation, forage, and, in some cases, milk following the

Chinese nuclear tests of October, 1964, May, 1965, and May,

1966.16,6,6,5,6 The Florida State Board of Health reported a

very pronounced increase of several days duration in the grose beta

activity levels of air in seven Florida cities following the 1964 test.4

Following the 1965 test, gamma analysis of air samples in the United

State showed 131I, tellurium-132, molybdenum-99, and neptuniurm-239 in

addition to 131I and 140Ba/La.165

Naturally occurring radionuclides of the uranium and thorium series

may also be present in environmental samples. In 1960, Muth et al.

reported the radium content of various food in Germany.16 Either total

radium or 22Ra is also reported for several of the total diet sampling

program in the United States.11,6 In reporting a study of gamrma

activity in a variety of plants in the states of Kerala and Madras in

India, Minstry et al. published vegetable ash gamma spectra that very

clearly showed activity from thorium series nuclides as the principal

component.169 Klament reviewed the literature in 1965 and reported

sources and typical levels of natural radionuclides in foods.17

Analytical method

Analytical methods vary from laboratory to laboratory and are

characterized by a combination of innovation and standard physical,

chemical, and radiochemical procedures.


Two of the laboratory systems handling large numbers of environ-

mental examples, the Health and Safety Laboratory of the USAEC and the

laboratories of the National Center for Radiological Health of the

USPHS have issued analytical reports and manuals in a readily available

Boni has developed a number of procedures for analyeis of 1370s

and other radionuclides in environmental samples.18,8 Other papers

on the analysis of environmental samples by gamma spectroscopy have been

published by Ward, Johnson, and their associates.18,8 In 1966,

Hercer reviewed analytical methods and corresponding references for

measuring radioactivity in food and other biological materials.187



The programs of monitoring radionuclides in Florida milk have

indicated that 1370 levels in Florida milk have characteristic geographic

patterns with unusual levels in much of the State. These patterns and

levels are particularly striking when compared to those of 9Sr.

Reports of additional investigations of this situation indicated that

the levels of 13Ca in Florida milk are directly related to levels of

intake of this nuclide by the animals and that high levels in milk are

due to elevated levels in one or more types of locally grown forage.

The findings reported for milk suggested that an unusual radio-

ecological situation exists in Florida. This suggestion in turn prompted

a study of (1) the extent to which these unusual levels and characteristic

patterns of 13Cs extend to other media and involve other nuclides, (2)

the influence of environmental radioactivity on human radiation exposure

in Florida, and (3) some of the reasons for unusual environmental radio-

nuclide levels in tais state.

Selection of Media for this Study

Various media were examined with respect to the following characteria-

ties: (1) their role in the food chain, (2) their distribution in the

State, (3) the importance of their economic role, (4) their capacity for

concentration of radionuclides, and (5) their likelihood of becoming

contaminated. As a result, lean beef and vegetables grown for human

consumption were selected for sampling.


.Since cesiumn follows metabolic pathways in animals similar to those

of potassium and is preferentially concentrated by animal, animals and

animal products are important sources of exposure to 13Ca. Beef and

milk are among the most important animal products produced in Florida.

Mlk is an important source of human exposure to radioactivity; dairies

are distributed throughout this state, and the dairy animal receive a

large portion of their feed from local pastures and locally grown forage.

However, milk was not considered in this investigation because it is

already being routinely sampled on a statewide basis and is currently

being studied in detail at, the farm level.

In contrast, there have been no reported studies of specific radio-

nuclidae in Florida meats. In fact, Ward and Johnson referred to the

paucity of any information on 1370 levels in beef.5 Since high levels

of 137C have been reported in forage in parts of the State, beef animaels

deriving a portion of their: diet from these forages will have correspond-

ingly high intakeas of this nuclide. Since ingested-137Ca to distributed

throughout the tissues, particularly the muscle, and because beef is an

important part of the human diet, this commodity is potentially a signifi-

cant source of 137C intake in this htate.

Other animal products produced in Florida for human consumption include

poultry products and pork. However, in this state chickens and other

poultry are commonly raised ~in batteries off the ground and are fed grained

and supplemental feeds largely brought in from outside the production

areaa Consequently these -animals receive little feed from the immediate

environment. Roga receive a variety of prepared feeder as well as refuse

and scraps, but, since Florida is a "grain-poor" atate, much of the hog

feed is also imported. Because of the feed sources and management

practices, poultry, egg, and pork radioactivity levels would probably show

little relationship to the factor influencing the milk levels that

originally instigated this study and therefore were not studied in this


From the fact that certain animal forages grown in Florida appear to

have unusual level and distinctive patterns of 1370, it might be

predicted that other vegetation grown under abmilar conditions of

exposure to depoettion and uptake from the soil would have similar

patterns of this nuclide. Vegetables are an important commercial crop in

thia State (400,000 acres, production valued at over $250,000,000 in the

1965-1966 production year), they are consumed directly by man, and some

Vegetable crop or another is produced statewide.18,8

Although occasional 13Ca analyses of Florida vegetable samples have

been reported in the literature, there have been no eystenatic, statewide


Selection of Beef Sampling Stations

A program was drawn up to sample animals raised at the Florida

Agricultural Experiment Stations with supplementation where necessary by

samples purchased directly from laughter houses or butcher shops. Beef

sampling stations are described in Table 2 and shown in Figure 4.

The beef sampling program was based on information obtained by

consultation with the director of the University of Florida Meat

Laboratory.19 Within the Florida Agricultural Experiment Station

system, beef animals are fed at seven sites within the State. Five

of these eites are identified in Table 2 as Stationa 1, 2, and 3;

Station 4--Source 1; and Station 7. Beef animals are also fed at the

Animal Nutrition Laboratory in Gaineaville but none were killed during

the course of sampling in this study. In addition, beef research is

getting underway at the West Florida Station at Jay, but no animiala were

available for laughter during the course of thief study.



Number Location Source of Animals Collection Point

1 Belle Glade Everglades Station Meat Laboratory

2 Ona R~ange Cattle Station Meat Laboratory

3 Brooksville West Central Florida Mhat Laboratory

4 Gaineaville 1. Beef Research Unit Meat Laboratory
2. Local Farmsx Local Packing Plant

5 Pensacola Local Farms, Becambia Local Butcher Shops
and Santa Rosa Counties and Slaughter Housee

6" Graceville Local Farm Farm where Raised

7 Quincy North Florida Station Local Packing Plant
with Assistance of
Heat Laboratory

*Special sampling.
"Special sampling, one sample only.

All of the beef feeding experiments culminate in alaughter and

Trading to determine the marketability (quantity and quality) of the meat

raised under the various feeding programs.19 Experiment station animals

are usually alaughtered at the Meat Laboratory in Gaineaville. In the


i)Experiment Station Sources

SOther Sources


case of the station near Quincy, animal are slaughtered at a local

packing plant in that city, but even these are graded by the director of

the Meat Laboratory.

The sampling of experiment station animals has the advantages that:

the majority of the eamples are available fram..a single source, the

identity of the carease is maintained through the slaughter house, and

detailed recorded are available if needed. These records identify the

history and description of the animal, the feeding program, and the

weight and grade of the carcaes. In contrast, slaughter house animal

are frequently bought at auction and mixed together in holding pens and

carcaes identity is not maintained, both of which make it difficult to

identify the source and history of any sample collected. This is not of

concern if the sole object of the study is to determine the average

radionuclide exposure to the public from meat marketed at a particular

point, but it does present a problem when it also is desired to relate

maet activity levels to specific locations and feeding practices.

Beef Sampling

Since random sampling from certain feeding experiments could have

resulted in the sampling of bullet far in excess of the small number appear-

ing in retail cute on the market, the restriction vae placed that bullet

were to be eliminated from sampling. A further restriction was placed

that anaples were to be from "control" animal or from .those on the

experimental diete most closely resembling the feeding practice of the

particular area.

In sampling from butcher shape or commercial slaughter houses,

samples were taken from meat on hahd at theu time, and special care had to

be exercised to receive eaamples from animals that had been locally grown.

This was no problem in the several cases where samples were collected

from country butcher who fattened their own beef.

Sex, breed, and age were not caneidered as variables in this

experiment, but these and other factors were recorded where available for

possible future use in investigation of any unusual or particularly

interesting results. It was assumed that the distribution of different

easplings from a station (often different experiments in the case of

experiment station samples) and the method of selection would produce

sample representing the average and range of radioactivity levels in

fattened beef raised under various "typical" Florida conditions.

Sampling in May, 1966, of two careassee from each of two experiment

stations confirmed the feasibility of the mechanism of sample procurement,

indicated the sample size necessary for analysis, and tested the sample

preparation, the analytical, and the computational procedurest.

The main beef sampling took place during the period January through

June, 1967. After part of the samples were collected, the 13Ce/potassium

ratio were examined to help determine the total number of samples to

collect. Inspection of the data suggested a trend with highest values in

the southern and central parts of the State and lowest in the northwestern

part. It was assumed that the "within stations" variance for all 1967

sample would not be greatly different from that computed for all samples

collected up to that time (1966 and 1967) and that the final 1967 station

means would not be greatly different from those observed up to that time.

Estimation were then made of sample sizes necessary to show selected

statistically significant differences.

Although inspection of the available data suggested differences

between stations, it was found that, with the assumed variance, at least

eix samples per station would be necessary to conclude that any of these

differences is statistically significant.

Vegetable Sampling

To minimize possible bias introduced by crop differences, stratified

sampling was employed with samples collected within each of three

categories (strata): (1) leaf and stem, (2) fruit, seed, and pod, and

(3) root and tuber. This categorization is generally consistent with the

one used by the United States Food and Drug Administration to su~mmarise

results of analyses of individual vegetables and with the food groupings

used in the total diet studies described in the literature review. If

sufficient numbers of samples are collected in each category, an examina-

tion of category effect is also possible.

Sampling was further stratified within these categories by specifying

a total of 12 crop classifications to be sampled wherever available.

These classifications, shown in Table 3, were set up by selecting and

combining from the approximately 20 different vegetable crope grown in

significant quantities in Florida.18,8

Selection was based on the amount of production and availability

in the State, following consultation with a vegetable crope specialist

from the Florida Agricultural Extension Service. Acreage figures taken

from published reports are also included in Table 3.18 The only high-

acreage crop not included in the sampling vere sweet corn and waterselone.

The corn presented technical problems in the collection and preparation of

sufficiently large easples of the edible portion, and the watermelons are

seasonal in nature and represent only a mall portion of a person's

yearly intake.

The principal harvest saeaon for most vegetables is between October

and June.188,189 Preliminary sampling was carried out in May and June,

1966, to test the feasibility of the sampling mechanism, to develop and

test sample preparation, analytical, and computational procedures, and

to determine the size of portion and counting time necessary to detect

existing levels of radioactivity. Full scale sampling took place between

January and July, 1967.




Category Crop

In the case of vegetable sampling, sampling "stations" were identified

by vegetable production areas as adopted from the designation by the

vegetable crop epecialists.18 All samples from fields within one of

these production areas were identified with that particular station

number. In order to perform statistical analyses of the data and keep

in 1965-16

not reported


not reported
not reported

No. oflB88
Where Gro~n

Leaf, Stem Lettuce, endive, escarole, romaine 8
Celery 5
Cabbage 11
Greens many
All Leaf, Stem 13 without greene

Fruit, Beans, peae 12
Seed, Pepper 10
Pod Tomato 10
Cucumber 16
Squash 13
All Fruit, Seed, Pod 19

Root, Potato 0
Tuber Turnip, Rutabaga many
Other (such as carrots and radishes) 2
All Root, Tuber 10 without turnips

the number of samples to a reasonable level, the State was divided into

regions and each station was assigned to an appropriate region. Initially

the Stata was arbitrarily divided into four regions of approximately equal

area and designated as Southern, Central, Northeast, and Northwest. The

data itself, however, began to suggest that more homogenous regions would

result if the lower east and west coasts were assigned to separate regions.

This resulted in the regional designation shown in Table 4 and Figure 5

and identified as Southeast, Southwest, Central, Northeast, and Northwest.

Sequence of Study

Following sampling, the work was pursued in five stagee:

1. Determination of kinds and levels of gamma radioactivity in the

selected media,

2. Determination of variations in media, geographic location, and

points in time,

3. Identification of unusual levels of gamrma radioactivity in these

media and development of hypotheses concerning causes and mechanisms

involved ,

4. Investigation of the interrelationships of the kinds and levels

of radioactivity in various media, and the relationships with other

available parameters,

5. Estimation of the effect of these nuclide levels cmn the radio-

activity in food products, and in turn, on human radionnelide intake an~d

radiation exposure.

Number Region Area Represented (county or part of county) Soil Type

1 Southeast Dade Mrl, Rockland
2 Southeast Broward, E. Palm Beach Sande
3 Southeast W. Palm Beach, N. E. Hendry, E. Glades, Eighlands Mucks, Peate
4 Southeast Martin, St. Lueie, Okeechobee, Indian River, Brevard Sands
5 Southwest Collier, Lee, Hendry, Charlotte, 01ades Sande
6 Southwest Hardee, S. Polk, DeSota, E. Manatee Sands
7 Southwest Sarasota Mucks
8 Southwest Manatee, S. Hillsborough, Sarasota Sands
9 Southwest E. Eillsborough, W. Polk Sands
10 Central N. H. Orange, S. E. Seminole, E. Lake Mucks
11 Central N. Seminole, S. Volusia Sands
12 Central Sumnter, Lake Sands
13 Northeast Marion (less Weirsdale) Mucks
14 Northeast S. Marion, S. E. Alachua Mucks
15 Northeast Flagler, St. Johns, Putnam, Clay Sands
16 Northeast Alachua, Union Sands
17 Northeast Bradford Sands
18 Central Citrus, Hernando, Pasco Sands
19 Northeast Gilchrist, Levy Sands
20 Northeast Taylor, Lafayette, Dizie Sands
21 Northeast Colurmbia, Hamlilton, Surwannee Sands
22 Northeast Jefferson, Madison Sands
23 Northwest Gadsden, Leon Sands
24 Northwest Calhoun, Washington, Robmes, Jackson Sands
25 Northwest Eacambia, Santa Ross Sandy Loans
26 Northeast Dural County Sands

a 188
Adapted from University of Florida Vegetable Crop Specialists.
*Corresponds to station numbers as shown in Figure 6.




Stations adapted from major
production areas designated
by University of Florida 18
Vegetable Crops Specialists.





Beef Sampling

The standard procedure for beef sampling was to sample two animals

from each group of animals. The most notable exception was the sampling.

of six animals at one time at Quincy, where there was only one slaughter

of experimental animals during the course of this study. Three pounds of

lean muscle were requested per sample; a total of 5 pounds was requested

when the sample contained untrimmed bone and fat.

In sampling experiment station animals, samples were selected from

the first two carcaeaes hanging in the cooler which met the restrictions

of the sampling deslign (Chapter III). Samples in this case consisted of

the portion associated with the 13th rib (3-4 inches of short loin) from

either side of the carcass. This portion could be collected without

seriously affecting the marketability of a hanging carcass.

In sampling other animals, it was not always possible to sample this

portion or identify the portion sampled, particularly when samples were

obtained through cooperating individuals.

Special samples from a packing plant at Station 4 (discussed in

Chapter V) were collected in the cutting room from animals selected by

packing plant personnel. When picked' up in the shipping department, the

samples were not identified as to the portion of the animal.

Other samples were cut from unspecified lean portions of carcaseae

of locally fed animal on hand.at the butcher shop or packing plant at

the time of easpling.

Vegetalble Sampling

Samples were collected both directly from farms and from markets.

A large proportion of the samples were obtained through the cooperation

of the Chemistry Division of the Florida Department of Agriculture, which

has an extensive program of sampling and analyzing agricultural products

for pesticide residue. The remainder of the samples was obtained directly

by the investigator, who visited both farms and market, and through the

assistance of volunteers, who collected and shipped in samples.

Vegetable samples were requested to be market eamples that had gone

through a washer or were in a condition suitable for washing in the

laboratory. Portions of at least 3.5 kg of edible portion (8 pounds)

were requested. If a large enough single portion were not available, a

sufficient number of smaller portions from the same area vere composited

to make up a sample of the required size. All aamples were identified by

area where grown and, if possible, by farm or nearest community.

Vegetable samples received from the Florida Department, of Agriculture

laboratories were collected according to standard procedures of that depart-

ment.19,9 Briefly, field sample were composite from five portions,

each collected from different parts of the field. Market or packing ehed

samples were composite from portions taken from a number of crater in

order to represent the lot or originating field. These samples were then

sub-sampled at the Department laboratory for analytical purposes, and the

excess portions after sub-eampling were made available for this study.

Some of the Department procedures call for quartering heads of leafy

samples such as cabbage and lettuce with analysis of opposite quarters.

when the original ample did no't include sufficient heads to provide

unquartered heads for this study, unused quarters were provided. In the

case of a small number of samples, the entire sample was sliced before

sub-sampling and the sliced excess provided for this study. In these

cases, the possible effect of unwashed samples had to be considered in

evaluating the data.

Field samplesl collected directly by the investigator or by volunteers

were collected in essentially the same manner as those received from the

Florida Department of Agriculture. Market samples collected by these

individuals consisted of 8 to 10 pounds of produce selected from that on

display at the market at the time of sampling. Market samples were

collected only if the area where grown could be identified.

Beef Sample Preparation

Beef samples were prepared for analysis by boning, trinning the

exceass.fat, and grinding the lean portion twice through a 3/16 inch (in.)

grinder plate.

Vegetable Sample Preparation

In general, analysis of vegetable samples was performed on the whole

of the portion corresponding to the particular category being examined

(leaf and stem; fruit, seed, and pod; or fleshy roat and tuber); samples

were not peeled, cored, or shelled. Tips were not removed from bean pods,

and seeds were not removed from squash and peppers. The entire portion

analyzed was edible although not always eaten.

Unless already washed when received, vegetable samples veza washed

under running tap water, neing a stiff bristled brush where necessary to

remove clinging dirt. They were then drained dry or blotted with a

turkish towel to remove excess moisture. After washing, samples were

chopped into smaller pieces if necessary and triturated with either a

meat grinder or a blender. About 1 milliliter (al) of 40 per cent

formaldehyde per 100 grame of finished sample was added as a preservative.

The tap water used for washing was checked periodically for radioactivity

to assure that washing did not introduce significant radioactivity.

Mbre specifically, tops were cut off turnips, rutabagas, and other

root vegetables at the top of the fleshy root, and stems were removed

from hard skinned squash. These vegetables were washed and cut where

necessary and then ground in two paease through a meat grinder with a

plate with 1/8 in. holes.

Stems were removed from cucumber, peppers, soft skinned squash, and

tomatoes; they were then washed, dried, chopped, and blenderized.

The outer leaves were removed from the cabbage head, the base of the

head was cut off, and the remaining portion was then chopped and blend-

erized. Some cabbage was received as quartered heads; these quarters were

trimmed along the cut faces before chopping to remove dirt and dehydrated


On the few occasions when untrimmed bunches of celery were received,

they were topped near~the center of the leaf cluster, and the upper

portion was discarded; a.portion of the base of all bunches was cut

away. Individual stalks of celery, collards, and turnip, mustard, and

other greens were separated from the bunches and washed, drained, blotted,

chopped, and blended. It was necessary to scrub with a brush to~remove

clinging dirt at the base of the stalk. Collards presented a particular

problem in that they were very tough and required considerably more

distilled water for blending (up to 1,000 at per 3.5-liter (1.) sample)

than did other vegetables.

Reads or bunches of salad greens such as lettuce, romaine, escarole,

and endive were out in halves or quarters which were then further

saparated into clusters of a few leaves for washing, blotting, chopping,

and blending.

The equipment need in sample preparation is identified in Appendix

A. Several different pieces of power equipment were tested for first

stages of sample reduction, but, since these devices had to be borrowed

and returned for each batch of samples, it was deemed more convenient to

prechop by hand with a butcher knife and cutting board.

In grinding, samples were preent only to a size that would paea into

the grinder, ground material was collected in tared containers, the

required amount of preservative was added and hand mixed, and the amount

of preservative and the net weight of the sample were recorded.

The first step in blending was to weigh the washed, dried, and

prechopped samples in a suitable container. A portion of the weighed,

prechopped ample was then~ introduced into the blender along with the

least amount of distilled water necessary to get blending action (none in

the case of tomatoes and cucumbers; about 200-500 al in the case of other

vegetables). Once blending action started, larger pieces of the sample

were introduced and additional distilled water was added only if necessary.

The net weight of sample added to the blender was computed and recorded

after taking the tare weight on the container. Any added water was

measured and recorded by volume; unit density weeas seaned. The weight of

blended material (sample plue water) was calculated and the required

amount of preservative was added to and blended with the sample.

Preserved samples were transferred to labeled wide mouth plastic jars

and held for analyeia. Dilution factor due to added water and preservative

was calculated and recorded for future use in calculating nuelide concen-


Ga~aa Rladioactivity CunatinR.

Individual pieces of equipment are described in Appendix A. Geamma-

radioactivity analyses were performed with a multichannel scintillation

spectrometry system. The detector was a stainless steel clad 4-in. by 4-in.

right cylindrical sodium iodide (thallium activated) crystal coupled to a

phototube. The detector was located in a shield 20 in, by 20 in. by 24 -;ba,

high (inside dimrensions) with 2-in, thick lead walls, floor, and cover and

with a 30 mil cadmiurm and 5 mil copper lining. Signals are aniklyzed by a

400-channel pulse height analyzer calibrated to 10.0 keV per channel.

Samples were loaded to the full mark in preweighed polyethylene

Marinelli beakers, and the loaded beakers were weighed and placed over the

detector in the shield. Two counting configurations were used; a 3.5-liter

beaker was used for vegetable samples,and a 1.0-liter beaker was used for

meat samples and vegetables of less than 2 liters in quantity. Samples

between 2 liters and 3.5 liters in quantity were diluted to 3.5 liters.

Normally, samples were counted for 50.0 minutes; some of the samples

at the beginning of the study were counted for 100 minutes.

At least two analyses, usually on the same or consecutive days, were

performed on each lean beef sample. All the vegetable samples were given

an initial count, and about two-thirds of these samples were recounted

after 50 to 100 days. The delayed second count helped confirm the tentative

identification of any short-lived activity and permitted recounting of

the longer-lived activity in the presence of much reduced levels of the

short-lived activity.

Gross (background plus sample) spectra for individual samples and

a background spectrum for the particular counting period were digitally

recorded for computer analysis and future reference. Backgrounds were

also subtracted from the individual counts in the analyzer memory, and the

oscilloscope displays of net spectra were inspected visually for the

presence and relative size of peaked. The net spectra of all original

counts and selected recounts were also recorded with the x-y plotter for

future visual reference.

Interpretation of Gamma Spectra

A garmma spectrum, as developed on a multichannel gamrma spectrometer,

is a frequency distribution of counts registered in particular energy

increments. Gamrma rays are emitted by radioactive atoms at one or more

discrete energies characteristic of the particular nuclide. However,

becanee of the nature of the interaction process and the limitations of

the instrumentation, gamma-ray photons of a single energy do not show a

"line" spectrum at the output of the spectrometer system but rather show

a continuous spectrum with an approximately Gaussian-shaped distribution

(photoelectric peak) corresponding to absorption of the total energy of

the photon plus a lower energy portion (Compton continuum) corresponding

to smaller fractions of the total photon energy. In addition, the spectrum

may show effects related to the configuration of detector and shield, such

as backscatter peaks or escape peaked peakede lower than the principal peak

by the amount of energy carried 'wvay by X-rays escaping from the detector)

and may also show secondary peak~ corresponding to additional photon

energies emitted by the nuclide.) The resolution or width of the photo-

peakg the shape of, and ar under, the Compton portion of the apactrum

relative to the photopeak) and the importance of backscatter and escape

peaked are functions of configuration and composition of the particular

sample, detector, and surroundings.

The number and energy of specific photopeake are utilized in identi-

fying specific nuclides. The quantity of nuclide in a particular sample

is determined from the area under some portion of the spectrumn

(such as the photopeak portion). The counts in a selected region are

summed, and a conversion or "efficiency" factor, appropriate to the

nuclide, ample configuration and counting eyetem, is applied to convert

this sum or area to the amount of nuclide in the sample.

In the case of a mixture of more than one gamma-emitting nuclide in

a sample, the spectrum reflects all of the contributors and is assumed

to be a linear combination of the individual contributorse Because of

the limits of resolution (width of the photopeaka), additional peake at i

other energies, and the Compton portion of the individual spectra, some

unabers of the mixture will contribute counts in the region of, or in

some case even obscure, the principal peake of various other members of

the mixture.

In thief study, gamma spectra were first inspected vianally for the

presence, location (energy), and relative size of peaks and then were

evaluated by a computational procedure.

Computational procedures for interpreting gamnma spectra vere reviewed

in the manual recently issued by the National Center for Radiological

Realth.19 The method need in this study is the one described as the

"Simiultaneous Equatione Method." This method vae chosen for a nahrber of

reasons. It was the method most familiar to the investigator at the time

of the beginning of the study and it vae being used an a large number of

public health laboratories at that time. This method in less subjective

than the leesselaborate stepwise nulide-by-nuclide trippingn" method,

and errors in the' estimation are not compounded successively in one

direction as severely as when stripping is performed. In addition, it is

less sensitive to instrument instability than some of the more sophisti-

cated method and once the initial simultaneous calibration equations

have been solved, computations can be performed with a calculator or

even by hand if necessary.

Briefly, in this method the interference between some number (N) of

contributors to a composite spectrumn are corrected for by setting up and

solving a system of N simultaneous equations involving the unknown

contributions of the N nnelides in terms of counts in the photopeak

regions. The original equations are developed from counting a standard

of each of the nuclides separately with the eyetes and in the configure-

tion to be used for the unknown samples. Once the simultaneous equations

have been solved to yield explicit equation for each nuclide, these

same latter equations can then be used for all subsequent sample computa-

tions until such time as the system needs recalibration.

A computer program for calibration computations was written in

FORTRAN II and later converted to FORTRAN IV for use with the Univeretty

of Florida IBM 709 computer. This program reads the conditions of the

calibration and the counting data for the various standard. It then

averages any replicate counts for the same nuclide, computes the matrix

corresponding to the coefficients of the simultaneous equations, inverts

the matrix in order to obtain the coefficients for the explicit equations,

and computer counting efficiency (counts per disintegration) in the

photopeak region for each of the nuclidea. Identifying information,

the computer coefficienta, and computer efficiency are printed out and

all the necessary calibration information is also punched on card

for subsequent nee in computations. For performing the subsequent


sample calculations, a program was written in FORTRAN II, later converted

to FORTRAN IV for the IBM 709, and also converted to FO)RTRAN IV for use

with the University of Florida IBM System 360 computer. This latter

program reader in the calibration coefficients and efficiencies and reads

the background counts and sample groes counts for each ample. It then

computes and writes out the concentration and the associated evo-standard

deviation counting error of each specified radionuclide in terms of

disintegrations per minute and pCi per unit (vteight or volume).



Individual samples were collected from 45 animal at seven different

stations.* Duplicate analyses vere performed on the lean portion of each

of these samples with repeat duplicate analyses on one sample for a total

of 92 lean beef analyses. Four of the samples were collected from two

stations during the preliminary work in May, 1966, and the balance were

collected between January and June, 1967. A departure from the original

plan took place when temples were collected at Station 4 from several

groups of animals that were obviously not fed to the same degree of

finish as the majority of the animals. The data from Station 4 are given

special Coneideration in this presentation of result.

The major gamma-emitting nuclides observed in the beef samples were

137Ce, 40K, and 226Ra and its decey products, bismuthh-214 (214Bi) and

lead-214 (214Pb). Quatitative comlputations were made for 137Cs, 40K,

and 226Ra in equilibrium with its daughters. It was necessary to include

the latter activity in order to correct for the interference with the

determination of 1370s and potassium by the 214Bi that occurred in a

large number of the beef samples. No particular significance is placed

on the quantitative values for radium and daughters since there was no

assurance that equilibrium actually existed and, indeed, no attempt was

*The single sample collected at Station 6 consisted of lean stew
meat from the farm's retail outlet and although it vae counted here as
one animal, it may actually have represented more.

made to prevent the lose of the gaseous radon-222 intermediate. Ceei~um-

137 is reported as p~ilkg of vet weight and as picocuries per gram a

potassium (pCi/g K). Since eeaentially none of the potassiurm is

contained in the fat, the latter method of reporting compensates for

variations in the completeness of trimming of the fat. On the other hand,

it obscures the effect on 13Ca of fat distributed through the lean.

Cesium-137 Content of Lean Meat from Grain-Fed Beefs 1967

A summary of 137Cs concentration, potassiurm content, and 137Ce/

potassium ratios of lean meat from grain-fed Florida beef sampled in 1967

is shown in Table 5 and in Bigure 6. There is an apparnt: gradual decrase

in 1370 content from south to northwest. The highest station average was

296 pCi/kg (78.8 pCi/8 K) at Ona and the lowest station average was 105

pCi/kg (31.8 pCi/g K) at Peneacola. The highest individual value, 539

pCilkg (133.6 pCi/g K), vae also found at Ona and the lowest individual

value, 55 pCi/kg (16.3 pCi/g K) occurred at Pennacola. The statewide

average of all samples was 214 pCilkg (54.4~ pCi/g K).

... Average of 8 samples
~ __...-1 from 1966 and 1967


No. of Samples: 16 161 6 21 16 111 1

Station No. 1 2 3 4 7 6 5


1370 K 1370s/potassiurm

Location dAimals Ave. Range~i~ Av.Ragv~Te. Ri~ange~



1 Belle Glade

2 Ona

3 Brooksville

4 Gainesvirlle

7 Quincy

6 Jackson County

5 Pensacola

Statewide Average
of Samples

Average of 7

263 j; 10

296 + 13

211 j 11

173 + 19

214 + 10

109 403

163 539

153 283

142 -203

182 277

3.65 j; .10

3.66 + .12

3.57 ]; .11

3.53 1 .19

3.44 + .11

3.26 4.22

3.30 4.04

3.09 3.97

3.31 3.75

3.27 3.58

2.97 3.82

71.0 j: 3.4

78.8 + 4.4

58.4 + 3.7

48.4 + 5.9

62.1 + 3.6

51.1 1 8.7

31.8 + 3.3

30.4 95.3

49.5 133.6

49.5 72.5

42.5 54.3

53.0 77.4

16.3 54.0

1* 175 + 26

6 105 + 10

--- 3.43 + .26

55 174 3.37 + .11

33 214 + 25 55 539

3.53 + .05 3.09 4.22

54.6 + 1.6 16.3 133.6

--- 3.52 + .06

57.4 + 1.9

-265 + 6

*Confidence intervals represent the two-standard deviation counting; error only
variance from all sources.

**This single sample was a composite of lean beef from meat in the show case at
have represented more than one animal.

and are not based on the total

the farm cormmissary and may

The effect of geographic location on 1370 content of lean meat

from grain-fed Florida beef was tested by performing an analysis of

variance on the 1370/potassium ratio reported for the samples at the

five stations from which the full complement of six samples were collected

(Stations 1, 2, 3, 5, and 7). This analyeia, presented in Table 6, shows

that, at the a = 0.01 level, there is sufficient evidence to indicate

a location effect.




Degrees Sum
of of Mean
Source Freedom Sqiuaree Square F

Station 4 7,526.77 1,881.69 4.56

Error 25 10,312.11 412.48 --

Total 29 17,838.88 -- --

Significant at the ar = 0.01 level.

The data were then examined to determine where significant differences

occurred. The same five stations were arranged in order of 137Ce/potassium

ratios and the Tukey procedure for simultaneously testing all paired of

meeanswas applied.194 A~s shown in Table 7, the "eastern" tatitona (1, 2,

7, and 3) constitute a group for which there is insufficient evidence to

indicate a significant difference between stations, but all stations in

the Broup are significantly different from Station 5. The "northern"

stations (5, 3, and 7) constitute a similar group, differing significantly

from Stations 1 and 2. The inability to establish the significance of

any smaller differences in station means is due to the high variation

within stations. It is possible that by increasing the number of samples

per station and/or by reducing the within station variance (such as

blocking on some variance contributing characteristic not identified

here) more of the apparent station differences could be demonstrated to

be statistically significant.




Station 2 1 7 3 5

Average pCi/g K 78.8 71.0 62.1 58.4 31.8

Solid lines indicate stations showing no significant difference at the
ar = .0 level.

Simultaneoualy testing of all possible paired of means according to the
procedure of Tukey.194

Inspection of the data suggests a trend of highest 1370 levels in

the southern part of the State, intermediate in the central and north-

eastern part of the State, and lowest in the far western part of the

State. Statistical testing of the data supported the hypothesis of a

location effect and supports the hypothesis of a systematic geographical

trend to the extent that the data can be grouped into two overlapping

groups of adjacent stations.

Effect ~of fPear 6f Collection

'twJo samples each were collected at Station 1 and 2 during May, 1966,

and six samples each were collected at the same stations during 1967.

As shown in Table 8, the two stations appear to show temporal trends in

opposite directions, so that any overall year-to-year difference is

insignificant, at least for the amount of data available.



No. of Samples 1370s pCifk 37et nf
Station 1966 1967 1966 1967 Ave. 1966 1967 Ave.

1 2 6 189.0 263.0 226.0 55.0 71.0 63.0

2 2 6 374.0 296.0 335.0 100.1 78.8 89.5

Ave. 281.5 279.5 280.5 77.5 74.9 76.2

EffeCt of Feeding Progreer and Meet gOality on casine-137 Level.

After cancellation of a scheduled kill of experiment station animals

at Station 4, an attempt was made to obtain alternative samples from

animals grown in the same general area. It was not possible in the time

available to obtain samples that were of the same quality of meat as those

obtained from experiment station' animal a 399 that were positively

identified as locally fed. It was possible to obtain four samples from

a local packing pleat consisting of unidentified cute from four locally

fed animals. This meat was intended for ground beef and meat products

(frankfurters, bologna, etc.) and was of a mudh lower quality than meat

which is usually sold as retail cute. These animals, slaughtered in

April, 1967, were identified only as "two had some grain, two fed

primarly grass."*

As shown in Table 9, two U)SDA GOOD samples collected in January,

1967, from two steers fed corn and eitrue pulp on an experiment station

dry lot in this same general area had 137C levels of 204 and 142 pCi/kg

(54.3 and 42.6 pCi/g K), respectively. Reference to Table 5 shows that

these values fell within the general range of all the "eastern" samples.

By contrast, the four packing plant samples had 137C contents ranging

from 293 to 12,500 pci/kg (90.4 to 3,710 poi/g K). 'Two of these animal

had 137s levels that were of a general magnitude about equal to rangingg

frDS oDB to two thmee) those of the other "eastern" samples and the other

two had levels 10 to 100 times as high. It was suspected at this point

*Cormenwt by the packing plant manager on the feed of these and
subsequent animal from this source werce probably based on inspection of
the condition of the animal and quality of the meat.


that the higher levels of 1370 in the meat were due to animal diets

composed primarily of grase. Several weeks later, four additional samples

were obtained from the same packing plant along with as much identifying

information as obtainable in order to further investigate this source of

meat. As shown in Table 9, all of these animals were cows of varying

breeder, the meat were of low grade, and again two samples were~ identified

as from grass fed animal and two from animals that had received some

grain. A range of results very similar to the first four samples was

obtained, with the totally grase-fed animals having the highest 13708


Sample Animal 137C 1370/
No. Description Feed Heat Grade pCifkg pfke pCify

Samples from University of Florida Beef Research Unit (trimmed from short loin)
Collected 1{7/67
7 Steer Dry lot: corn, Good 204 j; 29 3.75 ]; .30 54.3 j 8.9
8 Steer citrus puilp Good 142 + 24 3.31 + .24 42.6 + 7.8

Sampoles from local packing plant buying only local animala (trimmed from unidentified portions)*
Collected 4(26/67
36 293 + 24 3.24 + .24 90.4 + 9.9
39 4 animal identified only as "two had 507 + 29 3.51 + .27 144.0 + 13.7
37 some grain, two fed primarily on grass" 1,840 + 36' 3.69 +C .27 498.0 + ;37.8
38 12,500 + 73 3.36 + .28 3,710.0 + 309.0
Collected 5/10(67
50 Rereford cow Some grain Utility 567 + 27 3.90 + .26 145.0 + 12.0
53 Bereford cow Some grain Cutter 808 +e 41 4.09 +e .38 198.0 + 21.2
52 Jersey cross cow Grass-fed Not specified 5,530 + 68 4.97 + .40 1,130.0 jo 97.2
51 Brahama cow Grass-fed Canner 9,840 + 64 3.62 + .27 2,720.0 j; 200.0

"Confidence intervals represent the two-standard deviation counting error only and do not include other
sources of variance.

Commpents concerning feed of these animals were supplied by packing plant. They are probably based on
conditiDR Of animals and appearBaCe Of meat.





A total of 165 vegetable samples were collected, analyzed, and

evaluated for gammna-emitting radionuclide content. Of these, 29 were

collected during the preliminary sampling in May and Jurne, 1966, from eix

different stations in four regions. The remaining 136 were collected

during January through July, 1967, from 19 stations in the five -designated

regions. As surmmarized in Table 10, the 1966 aamples represented 10

different crope and the 1967 eamples represent the 12 designated crop


The principal radionuclides detected in vegetable samples in addition-

to 40K were 137Cs and naturally occurring 226Ra and ite daughters.

Following the arrival of fresh fallout, presumned to be from Chinese

nuclear weapons teete, short-lived nuclides (half-lives on the order of

one weekc) appeared briefly and moderately long-lived activities (half-

lives ranging from one month to one year) appeared and gradually


Ceatum-137l results are considered first in this chapter, and a

consideration of other nuclide results follows.

"Crope, categories, stations, and regions are defined in Chapter III.

Year. Category .;-SE SW C NE NW-a State Total

Leaf, Stem 3 0 11 0 0 14
Fruit, Seed, Pod 1 7 3 0 2 13
Root. Tuber 0 0 2 0j
All Categories 4 7 16 g 2 29


Leaf, Stem 14 13 13 9 9 58
Fruit, Seed, Pod 14 12 116 7 50
Root. Tuber 1 6 0 12 j[ju
All Categories 29 31 14 37 25 136

1966 6 1967

Leaf, Stem 17 13 24 9 9 72
Fruit, Seed, Pod 15 19 4 16 9 63
Root. Tuber -1 6 12 9 3
All Categories 33 38 30 37 27 165

Cesium-137 Content6 of:Florida Vegetables--
Average Levels and Reaion 'Variation

The resulted are sumrmarized by region for each of the two

period and for the total study in Table 11 and in Figure 7.

feature of these results is the wide range of values reported

all regions.




A striking

for nearly

S- May-June, 1966 January-July, 1967 1966 & 1967
Region N R R 'N R Nl 2 R

SE 4 50 + 10 0-111 29 24 + 6 0-176 33 27 + 5 0-176
SW 7 71 j 9 24-110 31 104 + 7 0-357 38 98 + 6 0-357
C 16 64 + 7 0-561 14 118 +11 26-356 30 89 + 6 0-561
13E 0 NS NS 37 48 + 6 0-214 37 48 +6 0-214
N1W 2 24 + 20 23- 25 25 23+ d 0- 67 27 23 + 6 0- 67

Regions 4 56 + 7 24- 71 5 63 + 3 23-118 5 60 + 4 23-98
Samples 29 61 1 5 0-561 136 58 + 3 0-357 165 59 + 3 0-561

Year 1966 1967 1966-1967



N = auaber of samples or regilons
P I average, pCilkg wet weight
R I range of sample or region average, same
NS = not sampled

units as mean

Confidence intervals of means are based on the two-standard deviation
counting error only and do not include other sources of variation.

NS-Not sampled
**** Average of all


The 1370s content of the 29.eamples from 1966 averaged 61 pCi/kg

(vet weight) while the 136 siamples from 1967 averaged 58 pCi/kg with an

overall sample average for all 165 samples of 59 pCifkg. The data

suggest a geographic trend with maximum values in the regions designated

Southwest and Central, the lowest values in the Southeast, and Northwest,

and intermediate values in the Northeast.

The apparent regional effect was tasted for the 1967 sampling period

with an analysis of variance. Examination of the data showed that the

variance increased in a general fashion as the mean increased and that

frequency distributions of sample means were skewed to the right. Both

observations suggest that in spite of other sources of variation, the

analytical results were strongly influenced by the behavior of radio-

active decay, which can be described by a Poisson distribution (variance

equal to the mean). Accordingly, in order to stabilize the variance,

the data was transformed for the purpose of the analysis of variance by

taking the square root of the reported sample means. From this analysis,

shown in Table 12, it can be seen that there was sufficient evidence at

the ar = 0.01 level to indicate a regional effect.

The Scheffe procedure was need to test the transformed data to

determine which of the regional averages were significantly different.194

As is shown by the ranking in Table 13, the central, southwest, and

northeast regions constitute a group for which there is insufficient

evidence to indicate a significant difference between regions, but all of

these regions are significantly different from both the northaest and

southeast regions. The northeast, northwest, and southeast regions also

constitute a similar group, differing significantly from the central and

southwest regions.

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