United States
Environmental Protection
Agency
Health Effects Research
Laboratory
Research Triangle Park NC 27711
EPA-600/1-7 8-033
May 1978
v»EPA
Maternal-Fetal
Tissue Levels of
Sixteen Trace
Elements in Eight
Communities
-------
RESEARCH REPORTING SERIES
Research reports of the Office of Research and Development, U.S. Environmental
Protection Agency, have been grouped into nine series. These nine broad cate-
gories were established to facilitate further development and application of en-
vironmental technology. Elimination of traditional grouping was consciously
planned to foster technology transfer and a maximum interface in related fields.
The nine series are:
1. Environmental Health Effects Research
2. Environmental Protection Technology
3. Ecological Research
4. Environmental Monitoring
5. Socioeconomic Environmental Studies
6. Scientific and Technical Assessment Reports (STAR)
7. Interagency Energy-Environment Research and Development
8. "Special" Reports
9. Miscellaneous Reports
This report has been assigned to the ENVIRONMENTAL HEALTH EFFECTS RE-
SEARCH series. This series describes projects and studies relating to the toler-
ances of man for unhealthful substances or conditions. This work is generally
assessed from a medical viewpoint, including physiological or psychological
studies. In addition to toxicology and other medical specialities, study areas in-
clude biomedical instrumentation and health research techniques utilizing ani-
mals — but always with intended application to human health measures.
-------
EPA-600/1-78-033
May 1978
MATERNAL-FETAL TISSUE LEVELS OF SIXTEEN
TRACE ELEMENTS IN EIGHT COMMUNITIES
by
John P. Creason
David J. Svendsgaard
Statistics and Data Management Office
and
Joseph E. Bumgarner
Cecil Pinkerton
Thomas A. Hinners
Environmental Monitoring and Support Laboratory
Research Triangle Park
North Carolina 27711
U.S. ENVIRONMENTAL PROTECTION AGENCY
OFFICE OF RESEARCH AND DEVELOPMENT
HEALTH EFFECTS RESEARCH LABORATORY
RESEARCH TRIANGLE PARK, N.C. 27711
-------
DISCLAIMER
This report has been reviewed by the Health Effects Research
Laboratory, U.S. Environmental Protection Agency, and approved for
publication. Mention of trade names or commercial products does
not constitute endorsement or recommendation for use.
-------
FOREWORD
The many benefits of our modern, developing, industrial society
are accompanied by certain hazards. Careful assessment of the relative
risk of existing and new man-made environmental hazards is necessary
for the establishment of sound regulatory policy. These regulations
serve to enhance the quality of our environment in order to promote the
public health and welfare and the productive capacity of our Nation's
population.
The Health Effects Research Laboratory, Research Triangle Park,
conducts a coordinated environmental health research program in toxicology,
epidemiology, and clinical studies using human volunteer subjects.
These studies address problems in air pollution, non-ionizing
radiation, environmental carcinogenesis and the toxicology of pesticides
as well as other chemical pollutants. The Laboratory participates in
the development and revision of air quality criteria documents on
pollutants for which national ambient air quality standards exist or
are proposed, provides the data for registration of new pesticides or
proposed suspension of those already in use, conducts research on
hazardous and toxic materials, and is primarily responsible for providing
the health basis for non-ionizing radiation standards. Direct support
to the regulatory function of the Agency is provided in the form of
expert testimony and preparation of affidavits as well as expert advice
to the Administrator to assure the adequacy of health care and surveillance
of persons having suffered imminent and substantial endangerment of
their health.
The developing fetus represents one of the most vulnerable subgroups
of the general population to the toxic effects of trace elements. This
investigation was designed to gain information on the levels of trace
elements present in the blood of a term fetus, and the relationship of
these levels to the levels found in the placenta and selected maternal
tissues. Once transplacental transfer of trace elements is established,
other researchers may attempt to ascertain the effects of the observed
levels of tissue trace elements on the developing fetus.
F. G. Hueter, Ph. D.
Acting Director,
Health Effects Research Laboratory
-------
CONTENTS
Disclaimer . ii
Foreword .iii
Abstract v
1. Introduction 1
2. Materials and Methods . 3
3. Results 8
4. Summary 19
5. Discussion 21
References 23
Tables 26
IV
-------
ABSTRACT
The developing fetus probably represents one of the most vulnerable
subgroups of the general population to the toxic effects of trace elements.
There have been numerous reports of abortion or fetal malformation due to
excessive exposure of the expectant mother to mercury and other trace
elements. This investigation was aimed at gaining information on the levels
of trace elements present in the blood of a term fetus, and on the relation-
ship of these levels to the levels found in the placenta and selected
maternal tissues. Once transplacental passage of the trace elements and
their levels in the placenta and fetus is established, other researchers
may attempt to ascertain whether these conditions actually result in overt
or subtle impairments of the developing fetus.
This study took advantage of the opportunity provided by normal
deliveries to obtain simultaneous tissue samples from a mother and her child.
Maternal-fetal sets consisting of maternal venous blood, cord blood, placenta,
maternal scalp hair, and pubic hair were collected and analyzed for the
following 16 elements; boron, barium, cadmium, chromium, copper, iron, lead,
lithium, manganese, mercury, nickel, selenium, silver, tin, vanadium and
zinc.
-------
SECTION 1
INTRODUCTION
Pollution of the environment by trace elements, many of which are toxic,
is a subject of increasing concern in our industrial society. Environmental
levels of many trace elements have increased along with industrial growth
(1-5). Exposure to trace elements in the environment is via multiple routes -
diet and water as well as air - and humans integrate their total environmental
exposure (1). Specific substances are known to concentrate selectively in
various tissues (6-7). Yet, little is known about how a great number of
trace elements accumulate in man, or what the long-term effects of such
accumulations might be. Ordinary tissue levels at low levels of exposure
have yet to be characterized for many.
The developing fetus probably represents one of the most vulnerable
subgroups of the general population to the toxic effects of trace elements.
Cases in point are the Minamata Bay incident with mercury (8) and numerous
reports of abortion or fetal malformation due to other trace element
exposures to the expectant mother (9-14). The present investigation was
aimed at gaining information on the levels of trace elements present in the
blood of a term fetus, and on the relationship of these levels to the levels
found in the placenta and selected maternal tissues. Once transplacental
passage of the trace elements and their levels in the placenta and fetus is
established, as it has been already for a few elements (9-16), other
researchers may attempt to ascertain whether these conditions actually result
in overt or subtle impairments of the developing fetus.
This study took advantage of the opportunity provided by normal
deliveries to obtain simultaneous tissue samples from a mother and her
1
-------
child. The tissues studied were either routinely collected or easily
available. Maternal-fetal sets consisting of maternal venous blood, cord
blood, placenta, maternal scalp hair and pubic hair were collected and
analyzed for the following 19 elements; arsenic (As), boron (B), barium (Ba),
beryllium (Be), cadmium (Cd), chromium (Cr), cobalt (Co), copper (Cu),
iron (Fe), lead (Pb), lithium (Li), manganese (Mn), mercury (Hg), nickel
(Ni), selenium (Se),. silver (Ag), tin (Sn), vanadium (V), and zinc (Zn).
The CHESS* network provided a framework for the nationwide collection
of tissue sets (17) for the specific purpose of characterizing the tissue
trace elements as to:
1. baseline concentrations
2. geographic trends
3. covariate effects (such as age, race and parity)
4. tissue interrelationships.
A knowledge of these parameters will permit the establishment of ranges of
elemental concentrations to be found across the United States. In addition,
geographic variation in tissue trace element concentrations may warn of
potential or existing environmental hazards to the extremely sensitive fetal
population.
* CHESS is an acronym for the Community Health and Environmental
Surveillance System.
2
-------
SECTION 2
MATERIALS AND METHODS
Collection of Samples
Maternal fetal sets were collected from each CHESS exposure sector in
eight CHESS areas. These areas were Charlotte, N. C.; Birmingham', Alabama;
Riverhead, N. Y.; Elizabeth, N. J.; Ogden, Utah; Salt Lake City, Utah;
West Los Angeles, California; and East Los Angeles, California.
Boundaries for these CHESS areas and descriptions of the areas have been
established in earlier studies (17-18). These areas have been grouped
geographically into regions designated as the southeast (Charlotte and
Birmingham), New York-New Jersey (Riverhead, N. Y. and Elizabeth, N. J.),
Utah (Ogden and Salt Lake City), and California (East and West Los Angeles),
Arrangements with local hospital departments of obstetrics provided the
collection mechanism. When possible, subjects were contacted during
prenatal care visits or classes. A local CHESS worker assisted the
obstetrics department in the selection of possible participants according
to the arrangements with the specific hospital. Subject participation was
on a voluntary basis. Consent forms were signed, and the mother's
questionnaire information completed during the prenatal care visits or
classes. Mother's and hospital information questionnaires, prenumbered
labels, and containers for tissue collection were supplied to each
hospital by the CHESS program. A local CHESS worker or a staff person
from the obstetrics department was responsible for completing the hospital
information form. Selection factors used in this study were as follows:
1. All samples should be from normal pregnancies and deliveries and
restricted to 16-39 year old mothers. "Normal" was defined by the
-------
knowledge and judgment of the clinicians and generally excluded
patients with major medical complications and/or coexisting illness,
and toxemia of pregnancy.
2. All study participants must have lived in the CHESS area for at
least the entire duration of their pregnancy.
3. Multiple births were excluded from the study.
Tissue sample collection procedures were as follows:
1. Maternal scalp hair was obtained from a recent haircut or trim and
mailed directly to the CHESS workers; instruction sheets and postpaid
envelopes, with individual prenumbered labels affixed, were given to
each participating mother at clinic visits or upon delivery; hair
samples were preferably obtained within two months of delivery.
2. All pubic hair shaved off in prepping was rinsed free of detergent
and antiseptic and stored in a Lab-Tek* container.
3. Two 15 ml specimens of venous blood were obtained from the mother,
when possible. These specimens were collected after delivery and were
stored in minimal Pb B-D* vacutainer tubes.
4. Two 15 ml specimens of umbilical cord blood were drawn in minimal
Pb B-D vacutainer tubes after cutting the cord, but before placenta
delivery.
5. Three peripheral wedges of the placenta were cut in full thickness
after placenta! delivery and placed in a separate clean, nonsterile
Lab-Tek cup; the samples were noted to be free of gross pathology.
* Mention of commercial products or company names does not constitute
endorsement by the Environmental Protection Agency.
-------
All blood and placenta specimens were immediately stored in a freezer
and kept frozen until shipped in dry ice to an EPA contract laboratory for
analysis.
Data on community dustfall were obtained monthly for the 12 months prior
to the end of the study at a central site within each community.in
conjunction with other CHESS studies. The locations of these sites were
such that the sample from each of the communities of interest in these
studies was within 2.5 kilometers of the monitoring site. The dustfall was
analyzed for Cd, Cr, Cu, Pb, Mn, Ni, and Zr content, expressed as milligrams
per square meter per month of trace element fallout.
Analytical Methods
Hair specimens were washed with a detergent solution, rinsed and dried
according to the procedure of Harrison et al. (19). Bloods and placenta
samples were weighed and lyophilyzed prior to digestion which was achieved
by oxygen combustion for some elements and by dry ashing for others, as
indicated in Table 1. The ashing procedure consisted of wetting the tissue
with quartz-distilled sulfuric acid and heating at 550°C in a muffle oven.
To prevent any losses when volatile elements were to be analyzed, weighed
portions of the samples were prepared by the Schoniger Flask Technique (20).
Dustfall samples were acid extracted and metal concentrations were
determined by atomic absorption spectrophotometry (18).
Analytical methods for all tissues for each element are shown in Table 1
Standard laboratory quality control procedures were employed. In addition,
recovery of all 19 elements added to a dustfall sample, to a hair sample,
-------
and to a blood sample were evaluated using additions that were either twice
the detection limit or twice the endogenous level, whichever was larger.
Recovery rates were greater than 80 percent in all cases, and greater than
90 percent in most of the cases.
Statistical Methods
Before statistical analysis of the data, the values were edited for
"outliers", values so far removed from the main body of readings as to
warrant their removal from the population for statistical analysis purposes.
A statistical procedure was developed for this process, so that the
editing was completely objective. In this procedure the inherent population
variability, as measured by the standard deviation of the logarithms of the
values, was estimated from the central section of the sample. Limits were
then set at three standard deviations from the mean of the logs. Histograms
of the data were carefully examined to ensure the effectiveness of this
procedure, and to verify that a large number of seemingly valid observations
were not being eliminated. Logarithms of the concentration values were used
to normalize the data and to make significance tests valid.
If the total sample mass or concentration of trace element in the
sample resulted in the amount of trace element being below the minimum
detectable limit of the instrument, then the minimum detectable limit
divided by the sample mass was taken as the censored value. These censored
values were flagged, but were used as measured values in the analysis.
Computations involving data where the censored values exceeded 10 percent
were not considered reliable by the authors. Arsenic, Be, and Co were
excluded from statistical analyses because of the high frequency of
censored values in all five tissues.
-------
Standard statistical methodologies of correlation and linear models
were applied to the trace element concentrations to detect effects and
interrelationships of the measured variables (21). Covariate groupings
used in the statistical analyses are shown in Table 2.
Monthly dustfall trace metal concentrations (Pb, Cd, Cu, Zn, Mn, Ni,
and Cr) observed in each sector were averaged over the 12-month period
ending in June 1972, the month the study was completed. These mean metal
levels were used as crude indices of environmental exposure in a portion of
the study.
-------
SECTION 3
RESULTS
Demographic Characteristics
The study population consisted of 198 subjects who provided sufficient
tissue samples for analysis. Demographic characteristics considered were
smoking patterns, parity, socioeconomic status, and age. The distributions
of these characteristics by CHESS sector are given in Table 3. Considering
each of these factors singularly does not show the high degree of
interdependency between them. There are significant correlations between
socioeconomic status and maternal age, parity and maternal age, and
socioeconomic status and smoking status (Table 4). This limits the
reliability of the assignment of significant effects to specific
covariates. However, this study was not specifically designed to
investigate such delineations, but rather to take the covariates into
account in looking for geographic differences in trace element levels.
Study Population Trace Element Concentrations
Tables 5 through 9 display the basic parameters estimated for each trace
element distribution by tissue. The last two columns are the endpoints of
the interval expected to contain 68 percent of the population observations
if the observations were from a log normal distribution whose parameters
coincide with the estimated parameters. The percent censored refers to the
percent of observations where the metal level in the sample was less than
the minimum detectable level. Statistics computed from samples where this
level exceeds 10 percent are not considered reliable by the authors.
The ranking of tissues by the number of metals for which the percent
-------
censored exceeded 10 percent was maternal blood with eight, placenta with
seven, cord blood with six, pubic hair with four and scalp hair with three.
Arsenic, Be, and Co were highly censored in all tissues, and were therefore
not included in further analyses.
There were fewer scalp hair samples than pubic hair samples, primarily
because subjects were asked to mail in scalp hair samples whereas pubic hair
samples were taken as a matter of hospital procedure. When the weight of
scalp hair samples was too small for analysis of all elements, only Pb, Cd,
Cu, Zn, Hg, Li, and Fe were analyzed.
Few trace element studies have been done on maternal-fetal sets (22-27).
A comparison of the trace element levels found in maternal and cord blood and
placenta in these studies with our results (Table 10) shows excellent
agreement for the five elements available, with the exception of a study
of placenta by Dawson et al. (27). There is a marked disparity of Dawson's
results from the results of Baglin and Brill et al. and Finklea et al., as
well as from ours.
A comparison of the five trace elements in scalp and pubic hair reported
in previous publications with the findings of our study reveals two facts:
First, there is little information available on the trace element content of
hair for this type of population (2,22,23,28,29); Secorfd, there is generally
good agreement with the trace element levels published in these similar
studies of adult females (Table 11).
To our knowledge, for most of the 11 remaining trace elements in our
study, there were no published values for this type of population with which
to make a comparison. Hence, these reported levels should serve as a frame
-------
of reference for future studies.
An examination of the geometric means in Tables 5 through 9 reveals
that both scalp and pubic hair trace element concentrations are higher than
blood or placenta values for all the trace elements except Fe. The
exception of Fe is understandable since about 70 percent of the body iron
is contained in blood (30).
Placenta trace element levels were higher than cord and maternal blood
levels for Pb, Cd, Zn, Se, and Mn. For Pb and Se, these differences were
relatively small even though they were statistically significant, whereas in
Cd, Zn, and Mn, the levels in placenta were about twice those found in
maternal and cord blood. Zinc levels would be expected to be elevated in
placenta, since the placenta contains large amounts of zinc-rich vascular
tissue (31). It appears then that, with the possible exceptions of Cd and
Mn, the placenta is not acting as a sink for trace element storage.
In order to closely examine the relative levels of maternal and cord
blood trace elements, and the relative levels of scalp and pubic hair trace
elements, ratios of these values were computed for all cases where both
values were present. A standard t-test was then carried out on the logs of
these ratios.
, V
Concentrations of four of the 16 trace elements —Pb, Cd, Cu and Zn —
were lower in the cord blood than maternal blood. Only Cu and Zn had ratios
significantly different from one at the 0.05 level. The other 12 elements
were higher in the cord blood. Eight of the 12 elements higher in cord
blood had ratios significantly different from one. (Table 12).
The average scalp hair levels were higher than pubic hair levels for
all 16 trace elements, since every geometric mean ratio is greater than or
10
-------
equal to one (Table 12). Ten of the 16 ratios were significantly different
from one at the 0.05 level.
Many of these ratios agree quite well with those computed from other
studies. Using the data on the five trace elements in Tables 10 and 11 to
compute ratios, 12 of the 15 possible available ratio comparisons from
other studies varied in the same direction from unity as did the ratios
found in these data. Two of the three cases in disagreement involved
comparisons with data from Finklea et al. (23).
Tissue Trace Element Interrelationships
Several authors have attempted to use some or all of these tissues as
indices of trace element exposure (2,3,22,23,32-34). Since many of the
trace elements are present at elevated levels in polluted environments, one
might expect tissues acting as depositories for trace elements to show
correlations between elements within that tissue. An example of such a
phenomenon is Cd and Pb in scalp hair as found by Hammer et al. (33).
Although scalp hair has been shown to be an effective index for some
elements, there are several potential sources of variation (such as hair
grooming differences, use of dyes and tints, and hair sampling techniques)
which may be avoided by the use of pubic hair as an index. Hence, one
would expect to find (1) a number of significant correlations between
trace elements in scalp hair, and (2) if pubic hair is another reliable
index for these elements, even more significant correlations between trace
elements should be found in this tissue. Since the sample numbers for
pubic hair are substantially larger than those for scalp hair, one would
expect the second condition to hold even if scalp and pubic hair were
equally effective exposure indices. Of the 45 possible correlation
11
-------
coefficients, there were 22 significant positive correlations and one
significant negative correlation in pubic hair, and 16 significant
positive correlations in scalp hair. (Table 13). In 11 instances
significant correlations were found for the same combination of elements
in both pubic and scalp hair. The differences between the correlation
matrices for scalp and pubic hair were generally found in Pb, Se, and Fe.
These three elements have far more significant correlations in pubic hair
than in scalp hair.
There were nine significant maternal blood correlations and 18
significant cord blood correlations, with seven metals being significant
in both tissues. Five of these seven trace element pairs had nearly
identical correlation coefficients in both types of blood. (Table 14).
The lower number of significant correlations in maternal blood as
compared to other tissues is understandable, since maternal blood is more
a transport tissue than a storage tissue. Cord blood, on the other hand,
is quite possibly acting as a repository for these elements. Also, the
placenta could be acting as a selective filter, allowing certain trace
elements to pass more freely than others. The placenta is an extremely
complex organ whose interactions with trace elements is poorly understood.
We cannot investigate such subtle interactions in this study. However,
it is of interest to note that the pattern of trace element correlations
in the placenta is very similar to that found in cord blood (Table 15).
There were 21 significant correlations between trace elements in placenta,
12 of which were also significant in cord blood, and only five of which
were significant in maternal blood.
12
-------
Significant correlations of trace element concentrations between
tissues are shown in Table 16 for each element. The maternal-to-cord-blood
correlations had the largest number of significant coefficients (12 of 16),
followed by maternal blood to placenta (7 of 16) and cord blood to placenta
(6 of 16). Scalp and pubic hair showed little relationship to maternal
and cord blood and to placenta, with only nine significant positive
correlations and eight significant negative correlations out of the 96
correlations possible. Scalp hair to pubic hair correlations showed only
Ba with a significant positive coefficient and B and Fe with significant
negative coefficients.
A comparison of these between-tissue correlations with those of other
authors for the few metals available is given in Table 17. The comparison
of our ratio of scalp to pubic hair results with those of Baumslag et al.
(28) is poor; however, most of his subjects were black while most of our
subjects were white. In the maternal-to-cord-blood results, our study
found many more significant correlations than Baglin et al. (22). Our
sample sizes for both maternal-to-cord-blood and maternal-blood-to-placenta
correlations were much larger for Pb and Cd than were Bag!in's, which
could explain the differing results for these elements in the tissue
comparisons. Mercury was highly censored in blood in this study (41%) and
no significant correlations were found in this study, while they were
significant in both Bag!in's study for maternal-to-cord-blood and
maternal-blood-to-placenta, and the work of Dennis's study (24) for
maternal blood to cord blood. The agreements between these studies is
encouraging, but the disagreements point out the need for further studies
to provide clarification of the true interrelationships of many of these
13
-------
elements between tissues.
Effects of Covariates on Tissue Trace Element Levels
Some previous studies have shown significant effects of selected
covariates on tissue trace element levels for some elements (2,22,28,33).
Creason et al. (2) found that some scalp hair trace element levels were
affected by area of residence, age, smoking, and socioeconomic status.
Baumslag (28) found race, age, and parity effects in hair for some trace
elements in a population of mothers and their newborn. Geographic
differences in tissue trace element levels have been reported by Hammer
et al. (33) among others. In view of all the above findings, it is
important to ascertain whether or not these or other covariates affected
tissue trace element levels in this study.
Almost all of the black subjects were in the southeastern region
(Charlotte, N. C. and Birmingham, Alabama) in this study. To avoid confounding
the race effect with geographic effects, only subjects in the Southeast were
included in making the race comparisons.
In both cities in the Southeast, there was one sector in which all the
women were black and one sector in which all the women were white. The
women in each of these sectors were well matched with respect to education
but smoking was less prevalent among blacks (Table 3). In addition, the
blacks generally reported less frequent shampooing of their hair than whites.
A joint analysis of race and city effects was carried out on the
southeastern region subjects in order to compensate for city-to-city
differences (21). A significant race effect was noted for Pb, Hg, Se, Fe,
and Mn in pubic hair and Li, Fe, Cr, and Mn in scalp hair (Table 18). No
significant race effects were noted for any elements in maternal blood,
14
-------
cord blood or for any elements in maternal blood, cord blood, or placenta.
The blacks averaged higher scalp and pubic hair trace element levels in
both cities in the cases where a significant racial effect was noted.
Although a significant racial effect was not found for both hair types in
each of these instances, generally, the nonsignificant hair type- reflected
the same relation between races as the significant hair type. Exceptions
to this occurred in scalp hair (selenium was higher among whites in
Charlotte) and in pubic hair (lithium concentrations were higher
among whites in Charlotte and Birmingham).
Scalp hair Se was the only tissue trace element with a significant
city by race interaction, indicating larger differences between the blacks
and whites in Birmingham than in Charlotte.
In view of the race effect on tissue trace element levels, blacks in
the Southeast were not included in subsequent analyses.
In each of the five tissues the two sector mean trace element levels
within each region were compared (Tables 19 through 23). Also, for each of
these tissues the four pooled regional mean trace element levels were
compared. Tables 24 through 28 list the geometric means of those trace
elements for which a significant difference was found. Significant
geographic differences were found for at least three of the five tissues
in 13 of the 16 trace elements tested. A majority of the sector
differences within regions were concentrated in the Southeast and Utah.
Regional differences were more frequent and often stronger than sector
differences within regions. Most of the regional effects were the result
of the southeast being significantly different from the other regions.
Maternal blood, cord blood, and placenta consistently exhibited the same
15
-------
regional trends.
The geographic effects noted above were not adjusted for the possible
effects of other covariates such as age, parity, socioeconomic class, and
smoking status, and, in the case of scalp hair, the usage of dyes and the
frequency of shampooing. In order to detect important factors and to
assess the contribution of these factors to the distortion of geographic
effects, a regression analysis on each tissue type was carried out. First,
the linear effects of all appropriate covariates were included in the
analysis. Next, geographic effects were included along with the covariates.
Riverside, N.Y. was omitted from this analysis since only one person had
covariate information in that community. The two sectors in Los Angeles
were pooled because of their small sample sizes. There were 27
tissue/element combinations for which a significant city effect was found
even after accounting for the linear effects of the covariates (Tables 29
and 30). Scalp hair had two elements with significant geographic effects.
However, the sample sizes for scalp hair were much smaller than for the
other tissues, sometimes being as small as 30 subjects. Thus it would be
less likely to find differences between regions in this tissue than in the
others. Cadmium and Lithium were the only elements with significant
geographic effects for all three tissues, maternal and cord blood and
placenta, after adjusting for the covariates, while Li was the only
element with significant geographic effects for all five tissues. In every
case where significant geographic effects were found, the covariates that
were significant excluding geographic effects were no longer significant
when geographic effects were included. In the six instances when a
covariate was still significant after including geographic effects, the
16
-------
geographic effects were found not to be significant. These cases were Se and
B in cord blood, Fe in maternal blood, Cr in placenta, and Cr in scalp hair.
Of these six tissue/element combinations, only B in cord blood had been
found to vary geographically.
Because of the unequal distribution of factors across sectors, it can
not be assured that covariates found to be unimportant after including
geographic effects do not have some effect. The proper conclusion is that
variation was better explained statistically by geographic effects.
Because of this fact it was felt important to report the covariate findings
when geographic effects were not considered (Tables 29 and 30). Before
examining these findings, it should be noted that a principal components
analysis (36) on smoking, parity, age, and socioeconomic status indicated
that over 73 percent of the variation in these factors could be explained
by the first two principal components. This finding reflects the high
degree of interrelationship among these factors and means that the
ability of standard statistical techniques to separate important from
unimportant cofactors is rather limited.
The most noticeable covariate in the analysis is the socioeconomic
factor, especially in scalp and pubic hair. Fourteen tissue/trace element
combinations were significantly related to this factor when geographic
effects were excluded, versus at most four for any of the other covariates.
The fact that only three tissue/trace element combinations are significant
when city effects are included indicates that there are differences in
socioeconomic conditions across communities that are confounded with the
geographic differences. These differences are apparently explained better
by the geographical covariate (city effects) than by the socioeconomic
17
-------
covariate in this study, but it may be that socioeconomic differences play
an important role in environmental exposure to trace elements and/or their
uptake by the expectant mother.
Relationships of Dustfall Trace Elements to Tissue Trace Elements
The minimum, maximum, and arithmetic mean of dustfall trace metals for
the 12 months immediately preceding the conclusion of the study are given in
Table 31. Only seven elements were analyzed in dustfall: Cd, Pb, Zn, Cr,
Cu, Mn, and Ni. Some element measurements were not available for several
of the months during this time period, as noted in the table. When a
correlation analysis of dustfall versus each of the tissues was carried
out for each of the seven elements, only two of the 35 possible correlations
were significant: Cd in scalp hair (r=0.18, p=0.03) and Cr in maternal
blood (r=0.15, p=0.02). Because dietary intake is quite probably more
important than the inhalation route in determining tissue trace element
levels, multimedia indices of exposure should be used in looking for
relationships between environmental exposure and trace element body burdens.
Unfortunately, only dustfall data were available, and in many cases these
data were scanty. In addition, the dustfall data were not collected
specifically for this study. Possibly better correlations could have been
found if more emphasis had been placed on this phase of the study.
18
-------
SECTION 4
SUMMARY
Trace element levels for 16 elements in five different tissues —
maternal blood, cord blood, placenta, scalp hair, and pubic hair —
collected from 198 subjects in eight communities in four distinct geographic
sections of the continental United States (the southeast, New York-New Jersey,
Utah, and California) have been reported. There was generally good
agreement with tissue levels published in similar studies for five elements.
This is the first report, to our knowledge, of levels for the remaining
11 elements. The tissue trace element levels generally ranked from
highest to lowest in scalp hair, pubic hair, placenta, and cord and
maternal blood. The cord blood and maternal blood rankings varied depending
on the element involved, with Pb, Cd, Cu, and Zn concentrations higher in
maternal blood and the other 12 elements higher in cord blood.
Correlations of trace elements within tissues were reported. The
pattern of correlations in the placenta was very similar to that found in
cord blood. Maternal blood had the fewest significant interelement
correlations.
Correlations of trace element concentrations between tissues for each
element showed that the maternal to cord blood correlations had the
largest number of significant coefficients. Scalp and pubic hair showed
little relationship to maternal and cord blood and to placenta.
Race was significantly associated with several scalp hair and pubic
hair trace element levels with the blacks averaging higher levels than the
whites. No significant race effects were noted for maternal blood, cord
blood, or placenta.
19
-------
Significant geographic variation in trace element levels were found for
many elements in all five tissues. Regional differences were stronger than
differences within regions. Maternal blood, cord blood, and placenta
consistently exhibited the same regional trends. In examining covariates
and their possible influence upon tissue trace element concentrations,
it was found that the socioeconomic factor was a possible influence,
especially on scalp and pubic hair. Since there were distinct differences
between regions in socioeconomic levels, the exact relationship could not
be established in this study. Age, parity of the mother, and smoking
habits were not found to be strong influences, although a few trace elements
showed some possible relationships to these factors. In scalp hair, shampoo
frequency was related only to Se, while use of hair treatments was related
only to Cr.
20
-------
SECTION 5
DISCUSSION
This study was not designed to afford a thorough investigation of the
covariates that could be affecting trace element levels in the five tissues.
The purpose of considering covariate information here was to detect
important sources of variation in the data. Such detection affords other
researchers in the field background information for improved design of
studies and a reference point for comparison of results.
In epidemiology, causation is rarely shown. However, even valid
associations require that strict designs be carried out. When a multitude
of possible factors are being considered, such designs are impractical.
In view of these considerations, and of the strong interrelationships
between some covariates, the reader is warned that indicated significant
associations with covariates may be spurious. The reader is also warned
that this population of women was selected by convenience and controlled
for several factors; therefore, all statistical inferences should be
specific to this population. Careful judgment as to the applicability of
these inferences to the general population must be used. We have included
results from the few other available studies throughout our report in
order to afford readers an opportunity to make this judgment for
themselves.
Even in view of the above restrictions, this report has provided
insight into many gaps in our knowledge of the transplacental transfer of
trace elements. It has also served to establish baseline levels for a
large number of trace elements in maternal and fetal tissues in a realistic
setting, and has shown that these levels vary widely across the country for
21
-------
many elements.
More definitive studies are obviously needed in this critical area.
The transplacental transfer of some elements found in contaminated
environments has been demonstrated. The excessive exposure of the fetus
to Pb in urban environments may already be occurring, based upon the cord
blood levels reported in this and other studies (26, 32). There could be as
yet unknown metabolic effects of pregnancy on trace element distribution
and toxicity in the placenta and developing fetus, as well as in the
expectant mother. The levels of exposure of the placenta and the fetus to
16 trace elements have been reported here. However, the mere presence of
a pollutant in tissue is not always sufficient evidence for disease
potential. Now, the toxicity of the various trace elements to the
expectant mothers, to the placenta, and to the developing fetus must be
established in order to protect them from existing or future hazards of
environmental trace element exposure.
22
-------
REFERENCES
1. Pinkerton, C., Creason, J., Hammer, D., and Colucci, A., Multimedia
indices of environmental trace metal exposure in humans. In Trace
Element Metabolism in Animals - 2, W. Holkstra et al. eds., University
Park Press, Baltimore, Md., 469 (1974).
2. Creason, J., Hinners, T., Bumgarner, J., and Pinkerton, C., Trace
elements in hair, as related to exposure in metropolitan New York.
Clinical Chemistry 2J_, 603 (1975).
3. Yamaguchi, S., Matsumoto, S., Kaku, S., and Hoshide, M., Relationship
between mercury content of hair and amount of fish consumed. HMSHA
Health Reports 86_, 904 (1971).
4. Hunt, W., Pinkerton, C., McNulty, 0., and Creason, J., A study in trace
element pollution of air in 77 midwestern cities. In Trace Substances
in Environmental Health-IV, D. D. Hemphill, ed., University of Missouri
Press, Columbia, Mo., 56 (1970).
5. Creason, J., McNulty, 0., Heiderscheit, T., Swanson, D., and Buechley,
R., Roadside gradients in atmospheric concentrations of cadmium, lead
and zinc. In Trace Substances in Environmental Health - V, D. D.
Hemphill, ed., University of Missouri Press, Columbia, Mo., 129 (1971).
6. Voors, A., Shuman, M., and Gallagher, P., Zinc and cadmium autopsy
levels for cardiovascular disease in geographic context. In Trace
Substances in Environmental Health - VI, D. D. Hemphill, ed.,
University of Missouri Press, Columbia, Mo., 215 (1972).
7. Strain, H., Berliner, W., Lankau, C., McEvoy, R., Pories, W., and
Greenlaw, R., Retention of radioisotopes by hair, bone and vascular
tissue. J. Nucl. Med. 5_, 664 (1964).
8. Harada, Y., Clinical investigations on Minimata Disease. Congenital
(or fetal) Minimata Disease. In Minimata Disease, M. Kutsuna, ed.,
Kumamoto University, Japan, 73 (1968).
9 Angle, C. and Mclntire, M., Lead poisoning during preonancy. Amer. J.
Dis. Child 108, 436 (1964).
10. Karlog, 0. and Moller, K., Three cases of acute lead poisonina.
Acta Pharm. Tox. J5., 8 (1958).
11 Palmesano, P., Sneed, R., and Cassady, G., Untaxed whiskey and fetal
lead exposure. J. Pediat. 75., 869 (1969).
12. Lugo, G., Cassady, G. and Palmesano, P., Acute maternal arsenic
intoxication with neonatal death. Am. J. Dis. Child 117, 328 (1969).
23
-------
13. Snyder, R., Congenital mercury poisoning. N. Engl. J. Med. 284,
1014 (1971).
14. Engleson, G. and Herner, T., Alkyl mercury poisoning. Acta Pediat.
Scand. 41_, 289 (1952).
15. Carpenter, S., Placental permeability of lead. Envir. Health
Perspectives, 129 (May, 1974).
16. Barltrop, D., Transfer of lead to the human fetus. In Mineral
Metabolism in Pediatrics, D. Barltrop and W. Barland, eds., Davis,
Philadelphia, 135 (1969).
17. Riggan, W., Hammer, D. et al. , CHESS, a community health and
environmental surveillance system. In Proceedings of the Sixth
Berkeley Symposium on Mathematical Statistics and Probability
(Vol 6), University of California Press, Berkeley, Calif., Ill (1972).
18. U. S. Environmental Protection Agency, Health consequences of sulphur
oxides: a report from CHESS 1970-71. U. S. EPA, RTP, N.C. (1974).
19. Harrison, VI., Yurachek, J. and Benson, C., The determination of trace
elements in human hair by atomic absorption spectroscopy. Clinical
Chemistry Acta. 23_, 83 (1969).
20. Southworth, B., Hodecker, J. and Fleischer, K., Determination of
mercury in organic compounds. Anal. Chem. 30, 1152 (1968).
21. Graybill, F., An Introduction to Linear Statistical Models, Vol. 1,
McGraw-Hill, Inc., New York (1961).
22. Baglan, R., Brill, A., et al., Utility of placental tissue as an indica-
tor of trace element exposure to adult and fetus. Envir. Res. 8, 64
(1974).
23. Finklea, J., Creason, J. et al., Transplacental transfer of toxic
metals. Presented before the Subcommittee on Toxicology of Metals,
Permanent Commission and International Association on Occupational
Health, Buenas Aires, Argentina, Sept., 1972.
24. Dennis, C. and Felu, F., The relationship between mercury levels in
maternal and cord blood. The Science of the Total Environment 3,
275 (1975).
25. Harris, P. and Holley, M., Lead levels in cord blood. Pediatrics 49,
606 (1972).
26. Scanlon, J., Umbilical cord blood lead concentrations. Amer. J. Dis.
Child 121, 325 (1971).
24
-------
27. Dawson, E., Croft, H., Clark, R. and McGanity, W., Study of seasonal
variations in nine cations of normal term placentas. Amer. J. Obstet.
and Gynoc. 102 (3), 354 (1968).
28. Baumslag, N., Yeager, D., Levin, L., and Peterinq, H., Trace metal
content of maternal and neonate hair. Arch. Envir. Health 29, 186
(1974). ~
29. Sorenson, J., Levin, L. and Petering, H., Cadmium, copper, lead,
mercury and zinc concentrations in the hair of individuals living in
the United States. Interface 2_ (2), 17 (1973).
30. Underwood, E. J., Trace Elements in Human and Animal Nutrition, 3rd
ed., Academic Press, New York, N. Y. (1971).
31. Tipton, I., The distribution of trace metals in the human body. In
Metal-Binding in Medicine, M. Seven and L. Johnson, eds., J. B.
Lippencott Co., 31 (1960).
32. Rajegowa, B., et al., Lead concentrations in the newborn infant. J.
Pediatrics 80, 116 (1972).
33. Hammer, D., Finklea, J., et al., Hair trace metals and environmental
exposure. Amer. J. Epid. 93_ (2), 84 (1971).
34. Hammer, D., Finklea, J., et al., Trace metals in human hair as a simple
epidemiologic monitor of environmental exposure. In Trace Substances in
Environmental Health - V, D. D. Hemphill, ed., University of Missouri
Press, Columbia, Mo., 25 (1971).
35. Morrison, D. F., Multiyariate Statistical Methods, McGraw-Hill, Inc.,
New York, N. Y. (1967).
25
-------
TABLE 1. SAMPLE PREPARATION AND ANALYSIS METHODS
Metals Preparation Analysis
Cd, Pb Oxygen combustion AA aspiration
Cu, Mn, Zn, Fe Oxygen combustion AA aspiration
As, Hg, Se Oxygen combustion AA on vapor
Li Oxygen combustion Flame photometry
Ag Oxygen combustion ES
Ba, Be, B, Cr Dry ashing ES
Co, Ni, V, Sn Dry ashing ES
Remarks:
1. Managanese in hair was evaluated from ES data when detection from AA
was found to be inadequate.
2. AA = atomic absorption
ES = emission spectroscopy
26
-------
TABLE 2. GROUPINGS USED TO ANALYZE COVARIATE INFORMATION
Maternal Age
Parity
Smoking Status
Socioeconomic
Status
(Education)
Shampoo
Frequency
Scalp Hair
Color
Pubic Hair
Color
Scalp Hair
Treatment
16-19 20-24 25-29 30-34 35-39
2-3 4+
Nonsmoker Exsmoker Current Smoker
High School
At least once a week Less than once a week
Black Brown Blonde Red
Black
No
Nonblack
Yes (a) Dyes
(b) Tints
(c) Shampoo Rinses
27
-------
TABLE 3. DEMOGRAPHIC CHARACTERISTICS OF MOTHERS BY CHESS SECTOR
Charlotte
Whites
SMOKING
PATTERNS
(lever
Ex
Current
Unknown
PARITY
One
Two or three
_ Four or more
r-o
oo
Unknown
EDUCATION
< High school
Iliyh school
> High school
Unknown
AGE
16 to 19
20 to 24
25 to 29
30 to 39
Unknown
7
1
18
1
9
11
6
1
19
5
2
1
11
5
8
2
1
Charlotte
Blacks
17
2
8
0
8
14
5
0
18
6
3
0
11
10
4
2
0
Biniiinqhani
Whites
8
2
16
0
13
9
4
0
17
7
2
0
12
12
2
0
0
Birmingham
Blacks
13
0
10
0
11
7
5
0
12
8
3
0
6
9
6
2
0
Riverhead Elizabeth
New York New Jersey
1
0
1
0
19
1
0
0
19
0
1
0
19
0
1
0
0
19
8
3
6
3
10
6
4
0
8
5
4
3
4
8
5
1
2
Ogden Salt Lake CHy
Utah Utah
12 7
1 4
0 9
0 0
9 11
3 5
1 4
0 0
0 4
1 8
12 8
0 0
2 3
5 13
6 3
0 1
0 0
West East
Los Angeles Los Anyelcs Overall
5 7 85
2 5 20
0 1 08
0 2 25
3 6 81
4 7 66
0 0 29
0 2 22
0 2 80
1 1 43
5 10 49
] 2 26
1 1 51
4 4 71
2 4 40
0 4 12
0 2 24
TOTAL
27
27
26
23
20
20
13
20
15
198
-------
TABLE 4. CORRELATION BETWEEN COVARIATES
(N=122)
Maternal
Age
Smoking
Status
Smoking
Education Status
0.23* -0.07
-0.39*
Parity
0.46*
-0.12
0.12
*Significance level less than 0.05
29
-------
TABLE 5. TRACE METAL LEVELS IN MATERNAL BLOOD*
ELEMENT
PB
CD
CU
ZN
HG
LI
SE
FE
BA
B
CR
NI
AG
V
SN
MN
rfOSS
186
186
183
181
177
185
182
137
185
187
178
182
187
181
186
186
PERCENT
CENSORED
0.00
3.23'
0.00
0.00
25.42
1.62
0.00
MEAN
33.27
3.17
79.34
662.00
0.96
0.56
11.68
0.00 48565.
0.00
4.81
0.56
14.29
41.71
48.07
11.83
5.38
"These values give a range
concentrations, assuming
8.65
10.04
11.1
8.5
0.40
1.3
4.6
3.49
MINIMUM
4.70
0.10
25.20
270.00
0.10
0.03
3.30
15200.
2.00
1.00
1.0
0.7
0.06
0.3
0.9
.60
MAXIMUM
178.00
31.30
149.00
1075.00
6.60
2.38
34.60
67100.
40.00
49 -.00
90.0
85.0
3.80
3.3
24.0
23.00
GEOM MEAN
27.81
1.72
75.72
646.00
0.41
0.39
10.84
48000.
7.43
7.70
6.9
3.3
0.26
1.2
3.7
2.71
that should include approximately 68% of the
the underlying distribution of concentrations
GM/GSD'
15.80
0.52
54.91
514.00
0.11
0.16
7.36
40644.
4.37
3.69
2.8
1.2
0.10
0.8
2.0
1.35
GM x GSof
48.95
5.67
104.43
810.00
1.49
0.96
15.96
56687.
12.61
16.09
17.4
12.3
0.65
1.7
6.9
5.44
population
is log normal .
*Levels in ug/100 ml.
30
-------
TABLE 6. TRACE METAL LEVELS IN CORD BLOOD*
ELEMENT #OBS
PB
CD
CU
ZN
HG
LI
SE
FE
BA
B
CR
NI
AG
V
SN
MM
180
185
136
135
185
185
180
185
183
187
184
182
186
187
187
186
PERCENT
CENSORED
.56
4.32
0.00
0.00
22.70
1.62
0.00
0.00
0.00
0.00
0.00
5.49
29.03
41.71
7.49
0.54
MEAN
31.77
2.76
45.46
499.00
1.33
0.78
12.71
50825
10.01
13.62
12.4
8.1
0.53
1.6
5.6
4.18
MINIMUM
2.70
.10
10.30
170.00
0.10
0.05
3.70
27200.
2.00
0.70
1.0
0.7
0.08
0.5
0.9
0.85
MAXIMUM
136.00
18.20
127.00
1100.00
11.50
5.21
35.60
73900.
45.00
94.00
74.0
56.0
9.00
6.0
42.0
21.00
GEOM MEAN
26.86
1.65
41.77
478.00
0.53
0.51
11.83
50369.
8.32
10.74
7.9
4.5
0.33
1.4
4.31
3.56
GM/GSD^
14.56
.52
27.33
354.00
0.15
0.19
8.16
43857
4.71
5.44
3.2
1.6
0.13
2.2
2.2
2.00
GM x GSD^
49.55
5.26
64.08
645.00
2.22
1.36
17.15
57848.
14.72
21.20
19.6
12.9
0.84
2.3
8.5
6.34
TThese values give a range that should include approximately 68" of the population
concentrations, assuming the underlying distribution of concentrations is log normal.
*Levels in ug/100 ml.
31
-------
TABLE 7. TRACE METAL LEVELS IN SCALP HAIR*
ELEMENT
PB
CD
CD
ZN
HG
LI
SE
FE
BA
B
CR
NI
AG
V
SN
HN
i?OBS
114
112
110
109
112
112
112
113
61
63
62
63
62
61
62
62
PERCENT
CENSORED
0.38
3.57
0.00
0.00
0.00
0.00
0.89
0.00
0.00
0.00
0.00
0.00
0.00
8.20
0.00
0.00
MEAN
20.158
1.0074
17.958
143.3
2.2246
.1723
0.9445
36.09
2.491
1.053
1.176
1.674
0.1400
0.140
1.405
1.203
MINIMUM
2.500
0.0960
4.140
50.0
0.1040
0.0180
0.0300
1.88
0.170
0.020
0.020
0.130
0.0080
0.004
0.070
0.060
MAXIMUM
78.900
4.5000
80.300
280.0
16.5000
1.1300
6.3000
236.00
14.000
8.000
7.900
10.000
1.1000
1.400
5.200
5.500
6EOM MEAN
14.908
0.7736
14.977
134.9
1.4125
0.1318
0.6596
20.68
1.789
0.523
0.591
1.011
0.0744
0.061
0.874
0.735
TThese values give a range that should include approximately 63°', of the
concentrations, assuming the underlying distribution of concentrations
GM/GSO1
6.800
0.3658
8.380
94.1
0.5227
0.0642
0.2359
7.57
0.753
0.153
0.169
0.376
0.0241
0.016
0.288
0.264
GM x GSD*
32.685
1.6362
26.769
193.4
3.3168
0.2705
1.5215
56.54
4.251
1.786
2.066
2.723
0.2293
0.228
2.653
2.045
population
is log normal .
*Levels in yg/g.
32
-------
TABLE 8. TRACE METAL LEVELS IN PUBIC HAIR*
ELEMENT
PB
CD
CU
ZN
HG
LI
SE
FE
BA
B
CR
NI
AG
V
SN
MM
PERCENT
rUab CENSORED rVA"
153
158
148
158
158
155
151
155
no
107
108
no
113
103
113
154
0.00
5.06
o.oo
0.00
1.90
5.31
0.00
4.52
0.00
o.oo
o.oo
o.oo
o.oo
41.75
0.88
0.00
8.905
0.4309
11 .764
110.69
3.7600
0.1977
0.7031
29.86
2.168
0.729
1.614
0.693
0.1334
0-051
0.785
1.981
MINIMUM
0.828
0.0100
2.710
6.14
0.0060
0.0030
0.0410
0.60
0.070
0.010
0.030
0.030
0.0040
0.006
0.020
0.040
MAXIMUM
35.100
3.2300
51.900
885.00
43.1000
0.8420
4.6200
233.00
11.000
6.800
17.000
5.400
0.8800
0.540
7.000
17.210
GEOM MEAN
5.363
0.2539
9.813
73.09
0.7121
0.1070
0.4867
15.48
1.372
0.416
0.794
0.414
0.0736
0.029
0.430
0.950
GM/GSD'
3.306
0.0813
5.476-
29.17
0.1142
0.02SO
0.2046
4.44
0.482
0.133
0.250
0.151
0.0223
0.010
0.131
0.275
GM x GSJ^
14.271
0.7924
17.588
183.14
4.4407
0.3957
1.1578
53.98
3.902
1.254
2.522
1.133
0.2428
0.080
1.409
3.277
TThese values give a range that should include approximately 68% of the population
concentrations, assuming the underlying distribution of concentrations is log normal.
*Levels in yg/g.
33
-------
TABLE 9. TRACE METAL LEVELS IN PLACENTAL TISSUE*
ELEMENT
PB
CD
CD
ZN
HG
LI
SE
FE
•BA
B
CR
NI
AG
V
SN
MN
?OBS
165
169
166
166
164
167
160
169
167
169
167
169
168
169
168
167
PERCENT
CENSORED
0.00
0.00
0.00
0.00
9.15
0.00
0.00
0.00
0.00
20.12
4.19
43.20
60.12
88.76
30.36
0.60
MEAN
37.25
4.41
42.21
1344.
2.39
0.65
14.39
30420.
10.1
8.3
6.3
3.4
0.47
1.5
5.0
9.4
MINIMUM
8.10
0.90
4.40
600.
0.10
0.05
6.10
3800.
3.0
2.0
1.0
0.8
0.09
0.8
2.0
1.0
MAXIMUM
150.00
15.80
118.00
3200.
81 .90
3.35
24.80
91400.
31.0
47.0
27.3
32.0
4.40
8.0
29.0
50.0
GEOM MtAN
31.56
3.74
33.75
1267.
0.65
0.49
13.85
28930.
8.9
6.1
4.8
2.2
0.26
1.3
3.9
6.9
i
GM/GSD7
17.91
2.08
15.82
906.
0.14
0.21
10.42
14280.
5.4
2.9
2.3
1.0
0.09
0.9
2.1
3.3
GM x GSD^
55.61
6.73
72.01
1771.
2.90
1.12
18.41
47100.
14.5
12.7
10.1
5.2
0.71
2.1
7.4
14.3
values give a range that should include approximately 68% of the population
concentrations, assuming the underlying distribution of concentrations is log normal.
*Levels in ug/100g.
34
-------
TABLE 10. A COMPARISON OF MEAN TRACE ELEMENT LEVELS FOUND IN
SEVERAL STUDIES IN MATERNAL AND CORD BLCOD AND PLACENTA*
Cd
Fe
Hq
Pb
Zn
MB
CB
PL
MB
CB
PL
MB
CB
PL
MB
CB
PL
MB
CB
PL
This
Study
3.0
2.7
4.4
48565
50325
30424
0.96
1.33
2.39
33.0
31.0
37.0
661
498
1344
Baglin 5 Brill
et al. (22)
1.7
1.6
1.7
42072
58575
12183
0.87
1.15
2.08
16.5
12.3
30.5
663
317
1175
Finklea
et al.(23)
3.7
4.6
6.8
-
0.50
0.60
1.40
33.0
46.0
96.0
-
Dennis
et al -(24)
0.68
0.75
Harris &
Hoi ley (25)
14.0
12.7
.
-
Scanlon
(26)
_
22.1
-
Dawson
et a]. (27
896
-
390
_
1713
*Maternal blood (MB) and cord blood (CB) in pg/100g.
35
-------
TABLE 11
TABLE 11. A COMPARISON OF GEOMETRIC MEAN TRACE ELEMENT LEVELS*
FOUND IN SEVERAL STUDIES IN MATERNAL SCALP AND PUBIC HAIR
Cd
Fe
Hg
Pb
Zn
SH
PH
SH
PH
SH
PH
SH
PH
SH
PH
This
Study
0.8
0.3
20.7
15.5
2.2
0.7
14.9
6.9
135
73
Baumslag
et al.(28)
22.1
15.0
-
31.5
16.6
136
151
Finklea
et al. (23)
3.'l
_
12.Vb)
Creason
et al. (2)
0.6
24.0
1.0
11.0
112
Baglin &
Brill
et al.(22)
-
1»
-
-
Sorenson
et al. (29
0.8
-
17.1
14.1
174
(a) Females ages 15-50
(b) Arithmetic Means
(c) Median
*A11 values are in yg/g of scalp hair (SH) and pubic hair (PH).
36
-------
TABLE 12. RATIOS BETWEEN TRACE ELEMENT LEVELS OF THE TWO
TYPES OF HAIR AND THE TWO TYPES OF BLOOO
Element
Pb
Cd
Cu
Zn
Hg
Li
Se
Fe
Ba
B
Cr
Mi
Ag
V
Sn
Mn
=f
of
Obs.
97
99
90
96
100
99
94
98
38
37
39
39
33
37
38
49
Geometric
Mean of
Ratio Between
Scalp Hair
and Pubic Hair
Trace Element Levels
2.09*
2.95*
1.54*
1.79*
1.80*
1.15
1.32*
1.00
1.52*
1.59
1.04
2.69*
1.60
2.61*
2.22*
1.05
1
of
Obs.
173
175
177
173
146
175
170
180
170
175
163
166
164
142
174
172
Geometric
Mean of
Ratio Between
Cord Blood
and Maternal Blood
Trace Element Levels
0.94
0.93
0.55*
0.73*
1.49*
1.35*
1.08
1.04*
1.12
1.41*
1.15
1.15
1.28*
1.25*
1.18*
1.33*
* Ratio significantly different from 1 at .05 level.
37
-------
TABLE 13. SIGNIFICANT CORRELATIONS BETWEEN
ELEMENTS WITHIN THE TWO HAIR TYPES
Pb
Pb ^ ^
Cd .62^
Cu
Zn
Li .21
Se
Fe .35
Ba
Cr
Mn
Cd Cu Zn Li Se
.28 .40 .25 .24 .23
.18 - .24
.26^ fi^6f .40 .46 .38
^ ^/
.24X '/• / .43 .24
xx
.38^
v42 ^/^
.28 - - .49 - ^
_____
.30
_____
Fe Ba Cr
.35
.23
.51
.62
.39 -.24
.28
^/> X
•s
.47 x .42
.53 .69 v
.30 .64 .69
Mn
.20
-
_
-
_
.63
.46
x v
Remarks:
1. Only includes elements whose censoring level was small in all tissues in
order to simplify the task of making valid comparisons of the correlation
structure between tissues. Significance level less than 0.05 two sided.
2. Sample sizes between Pb, Cd, Zn, Li, Se and Fe varied from 157 to 141 in
pubic and from 113 to 106 in scalp hair. For the remaining metals sample
sizes varied from 105 to 101 in pubic hair and from 61 to 58 in scalp hair.
-------
TABLE 14. SIGNIFICANT CORRELATIONS BETWEEN
ELEMENTS WITHIN THE TWO BLOOD TYPES
Pb Cd Cu
Pb v .45 .18
Cd .20 ^
Cu .15 -" x c
Zn - - ^-1*.
Li -
Se
Fe
Ba
Cr - -.24
Mn
Zn Li Se Fe Ba 'Cr Mn
.27 .17 .16 - -
.22 - .16 .21 - -.24
.38 - .17
?o0 - ~ .35 .17
^- x - --"16 -20
~~- /;
.46 -.17 - %v
\
.19 - .20 v .30
.17
Remarks:
1. Remark #1 on Table 13 applies.
2. Sample sizes varied from 184 to 172.
39
-------
TABLE 15. SIGNIFICANT CORRELATIONS BETWEEN
ELEMENTS WITHIN PLACENTA
Pb Cd Cu Zn Li Se Fe Ba
Pb .17 .32 .21 .24 - - .17
Cd - .22 -.16 .30 .32 .20
Cu .56
Zn - .25 .26
Li -
Se .18
Fe .28
Ba
Cr
Mn
Cr Mn
- -
.42
-
.35
-.21
.20
.42
-
.24
Remarks:
1. Remark 1 on Table 13 applies.
2. Sample sizes varied from 155 to 167.
40
-------
TABLE 16. A SUMMARY OF SIGNIFICANT1 CORRELATIONS BETWEEN TISSUES BY ELEMENT
Ba
B
Cd
Cr
Cu
Fe
Pb
Li
Mn
Hg
Ni
Se
Ag
Sn
V
Zn
Tissue Pairs
MB
to
CB
.17**
(177)
.29***
(179)
.26***
(170)
.17**
(177)
.33***
(181)
.20***
(174)
.41***
(178)
$jj&
.14*
(171)
//$&/
yyW'
.15**
(174)
MB
to
PL
-.13*
(164)
.23'**
(162)
.13*
(157)
.15**
(164)
-.14*
(159)
.26**
(159)
'
(159)
CB
to
PL
.24***
(161)
-.22***
(159)
.14*
(158)
.31***
(159)
-.13*
(159)
.14"
(153)
MB CB MB CB PL PL SH
to to to to to to to
SH SH PH PH SH PH PH
-.18* .29*
(106) (39)
.25* -.21** -.28**- .17* -.36**
(59) (103) (51) (104) (37)
-.18* -.26
(104) (103)
.22* .36***
(54) (57)
-.19**
(138)
.15* -.24** -.20**
(148) (99) (98)
23*
(51)
.17*
(102)
V
'//
7 .17*
(109)
1 Two-sided test, i.e. Ho:?=0 vs H^p/O
* Significant at }G% level
** Significant at 3S level
*** Significant at 1" level
\y///A ;.|ore than IDS of pairs contain at least 1 censored value.
MB - Maternal Blood
C3 Cord Blood
PL Placenta
SH = Scalp Hair
PH Pubic Hair
41
-------
TABLE 17. A COMPARISON OF SIGNIFICANCE TESTS
FOR CORRELATIONS BETWEEN TISSUES*
Scalp to Pubic Hair
This Baumslag et al.(28)
Study
(N=95) (N=50)
Cd NS 0.01
Cu NS 0.01
Fe 0.05 (neg. r) 0.01 (pos. r)
Pb NS 0.01
Maternal to Cord Blood
This Bag!an & Brill Dennis ocl
et al-(22) et aU24) Hoi ley (25)
(N=600) (N=43) (N=22)
Cd 0.01 NS
Fe 0.01 0.05
Pb 0.01 NS(N=10) NS
Hg NS 0.01 0.01
Se 0.10 0.01
Zn 0.05 NS(N=12)
Maternal Blood to Placenta
Cd
Fe
Pb
Hg
Se
Zn
This
Study
(N>155)
0.05
0.05
0.10(neg. )
NS
NS
0.01
Baglan & Brill
et al. (22)
(M=600)
NS(N=34)
0.05
NS(N=73)
0.05
0.01
0.05
The probability of getting a value as large as observed if the true
correlation was zero is given. NS means not significant at the .10
level (p>0.10).
42
-------
TABLE 18. SIGNIFICANCE LEVELS OF TESTS FOR RACE EFFECTS
AND RACE BY CITY INTERACTION I,'! BIRMINGHAM AND
CHARLOTTE*
Effect
Cr
Fe
Hg
Li
fin
Pb
Se
Race Effect
Scalp
Hair
<0.05
<0.001
<0.05
<0.01
Pubic
Hair
<0.05
<0.02
<0.01
<0.01
<0.01
Race by City Interaction Effect
Seal p
Hair
<0.05
Pubio
Hair
*No significant effects were found for maternal blood, cord blood, or placenta.
43
-------
TABLE 19. TESTS OF SIGNIFICANCE OF REGIONAL DIFFERENCES
IN SCALP HAIR TRACE ELEMENT LEVELS*
Trace
Elements
Pb
Cd
Zn
Li
Se
Fe
Ba
V
Sn
Mn
Between
Southeast
-
0.05
0.01
0.05
0.01
0.01
0.05
Sectors Within the Same Region
NY-HJ Utah California
0.05
0.05
0.05
0.01 0.01
-
NO TEST
NO TEST
NO TEST
NO TEST
Between
Regions
0.01
0.01
-
0.001
0.001
0.05
* Values given are the probability of the observed difference in sample geometric
mean levels between categories assuming no difference in the original population.
Only values of 0.05 or less are listed. Elements omitted from the table had no
significant differences in any of the above comparisons.
44
-------
TABLE 20. TESTS OF SIGNIFICANCE OF REGIONAL DIFFERENCES
IN PUBIC HAIR TRACE ELEMENT LEVELS*
Trace
Elements
Pb
Cu
Zn
Hg
Li
Se
Fe
Ba
B
Cr
Sn
Mn
Between Sectors Within the Same Region
Southeast NY-NJ Utah California
-
0.01 0.01
0.05
0.01
-
0.05 0.05
-
0.01 0.01
0.05
-
0.05
Between
Regions
0-001
0.001
-
0.001
0.001
0.001
0.001
0.01
-
0.001
* Values given are the probability of the observed difference in sample geometric
mean levels between categories assuming no difference in the original population.
Only values of 0.05 or less are listed. Elements omitted from the table had no
significant differences in any of the above comparisons.
45
-------
TABLE 21. TESTS OF SIGNIFICANCE OF REGIONAL DIFFERENCES
IN MATERNAL BLOOD TRACE ELEMENT LEVELS*
Trace
Elements
Pb
Cd
Cu
In
Li
Se
Ba
B
Cr
Ni
Sn
Between Sectors Within the Same Region
Southeast NY-NJ Utah California
0.05 0.05
0.05
-
0.05
0.05
-
0.05
0.001
-
-
0.05
Between
Regions
-
0.001
0.01
0.01
0.001
0.05
0.05
0.001
0.001
0.001
* Values given are the probability of the observed difference in sample geometric
mean levels between categories assuming no difference in the original population.
Only values of 0.05 or less are listed. Elements omitted from the table had no
significant differences in any of the above comparisons.
46
-------
TABLE 22. TESTS OF SIGNIFICANCE OF REGIONAL DIFFERENCES
IN CORD BLOOD TRACE ELEMENT LEVELS*
Trace
Elements
Cd
Cu
Zn
Li
Fe
B
Cr
Ni
Mn
Between Sectors Within the Same Region
Southeast NY-NJ Utah California
-
0.05
0.05 0.05
0.05
-
-
-
0.05
Between
Regions
, 0.001
0.01
0.005
0.001
0.05
0.01
0.001
0.001
-
*Va1ues given are the probability of the observed difference in sample geometric
mean levels between categories assuming no difference in the original population.
Only values of 0.05 or less are listed. Elements omitted from the table had no
significant differences in any of the above comparisons.
47
-------
TABLE 23. TESTS OF SIGNIFICANCE OF REGIONAL DIFFERENCES
IN PLACENTA TRACE ELEMENT LEVELS*
Trace
Elements
Pb
Cd
Cu
Zn
Li
Se
Fe
Mn
Between Sectors Within the Same Region
Southeast NY-NJ Utah California
0.05
_
-
-
0.05
-
0.001
-
Between
Regions
-
0.001
0.001
0.05
0.001
0.001
0.001
0.001
Values given are the probability of the observed difference in sample geometric
mean levels between categories assuming no difference in the original population.
Only values of 0.05 or less are listed. Elements omitted from the table had no
significant differences in any of the above comparisons.
48
-------
TABLE 24. SECTOR GEOMETRIC MEANS OF TRACE ELEMENTS IN MATERNAL BLOOD*
IJD
Charlotte Whites
Charlotte Blacks
Birmingham Whites
Birmingham Blacks
Riverhead
Elizabeth
Ogden
Salt Lake City
West Los Angeles
East Los Angeles
// of Obs.
Max Min
26-21
26-23
26-25
23
20-18
20-18
8-7
19-17
6-5
14-13
B
3.75
5.45
7.03
7.17
9.55
10.79
9.04
11.02
12.31
15.41
Ba
5.53
6.40
7.67
8.10
7.03
6.90
3.76
9.05
8.62
9.53
Cd
1.004
0.988
1.102
0.892
4.586
2.184
3.263
3.855
3.712
2.947
Cr
9.33
9.70
9.92
9.41
5.82
5.22
4.87
4.07
3.58
5.08
Cu
78.6
81.6
77.2
86.8
60.2
69.4
61.0
71.0
88.8
81.2
Li
0.440
0.422
0.591
0.656
0.248
0.193
0.263
0.345
0.728
0.280
NJ.
5.61
5.74
5.80
6.57
3.90
2.92
1.65
1.64
2.19
1.65
Pb
35.3
28.2
26.1
24.3
27.2
25.9
18.5
31.8
31.7
27.4
Se.
9.45
10.52
10.93
11.91
11.59
12.53
10.22
11.27
8.03
9.93
Sn.
2.95
3.84
4.25
4.34
3.54
3.62
4.97
3.60
3.04
3.28
Zn.
580.1
612.2
657.4
573.2
709.0
711.4
630.5
678.8
667.0
709.0
'Levels in yg/lOO nil.
-------
TABLE 25. SECTOR GEOMETRIC MEANS OF TRACE ELEMENTS IN CORD BLOOD*
Charlotte Whites
Charlotte Blacks
Birmingham Whites
Birmingham Blacks
Riverhead
Elizabeth
Ogden
Salt Lake City
en
° West Los Angeles
East Los Angeles
t of Obs.
Max Min
25-22
26-24
25-24
23-21
20-18
19-18
11-8
19-16
6
15-13
JL
9.
8.
10.
10.
14.
15.
9.
9.
8.
13.
01
63
83
74
58
26
17
01
82
03
Cd
1.216
1.140
1.006
0!573
3.448
2.861
3.297
3.033
2.071
2.976
10.64
13.62
11.66
10.63
6.16
4.43
6.48
4.06
5.24
5.42
Cu_
42.9
45.5
48.0
50.3
36.5
29.6
44.8
37.0
56.7
36.5
49819.
49994.
48526.
45600.
52794.
53006.
53517.
51946.
50019.
52384.
0.756
0.763
0.621
0.781
0.308
0.226
0.416
0.444
0.838
0.313
Mn_
3.95
4.60
3.44
3.71
2.76
3.76
4.28
2.77
2.73
3.20
Hi
5.63
6.13
7.98
5.31
3.92
3.77
3.08
2.53
3.20
2.32
Iii
405.1
415.2
492.1
431.3
517.2
565.0
610.4
489.1
430.1
541.6
*Levels in pg/100 ml.
-------
TABLE 26. SECTOR GEOMETRIC MEANS OF TRACE ELEMENTS IN PLACENTA*
Charlotte
Charlotte
Whites
Blacks
Birmingham Whites
Birmingham Blacks
Riverhead
Elizabeth
Ogden
Salt Lake
West Los
Cast Los
City
Angeles
Angeles
# of Obs.
Max Min
21-18
21-18
21-18
18-17
19-17
20-17
12-11
19-18
6-5
15-14
2
2
2
2
5
4
5
4
4
5
Cd
.822
.555
.341
.625
.400
.823
.242
.855
.316
.983
Cu.
51.8
47.9
48.4
52.4
25.0
21.4
25.9
19.5
19.7
29.6
Fe_
31259.
24727.
12624.
19570.
33036.
31234.
35841.
27591.
35741.
31215.
y_
0.685
0.711
0.820
0.874
0.194
0.235
0.477
0.287
0.479
0.608
Mn_
4.48
3.72
4.03
4.42
9.51
12.11
12.52
10.04
8.77
12.15
P_b
26.9
31.8
39.6
36.9
32.2
27.8
28.7
27.0
29.2
36.6
Se_
12.36
13.96
11.43
12.03
13.70
15.74
15.61
14.97
17.91
14.68
Zn.
1123.6
1253.2
1206.6
1066.6
1267.9
1422.2
1528.2
1404.5
1308.0
1298.3
"Levels in pg/100 g.
-------
TABLE 27. SECTOR GEOMETRIC MEANS OF TRACE ELEMENTS IN SCALP HAIR*
Charlotte Whites
Charlotte Blacks
Birmingham Whites
Birmingham Blacks
Riverhead
Elizabeth
Ogden
Salt Lake City
c_n
ro West Los Angeles
East Los Angeles
# of Obs.
Max Min
10-7
12-10
9-7
11-7
10-5
18-5
10-6
18-0
4-3
12-7
Ba_
2.058
4.020
1.567
2.224
0.842
1.823
1.224
Missing
0.701
0.994
Cd_
1.297
1.129
0.952
0.946
0.539
0.671
0.513
0.703
1.041
0.597
Is.
29.65
113.69
19.01
37.64
22.68
26.35
7.17
10.73
5.24
10.31
0.
0.
0,
0
0
0
0
0
0
0
LJ_
,081
.160
.163
.179
.188
.236
.054
.127
.106
.078
• Hi
0.648
2.127
0.465
0.555
0.719
0.568
0.360
Missing
0.921
0.653
Pb_
23.79
24.64
15.23
23.30
10.55
14.06
6.45
11.93
23.81
13.93
Se_
0.679
0.425
0-539
0.959
0.424
1.033
0.389
1.062
0.591
0.442
Sn_
0.878
2.006
0.421
0.612
1.774
1.708
0.742
Missing
0.385
0.420
Zn_
108.0
139.2
126.8
163.7
152.5
134.3
112.6
155.5
138.4
115. 6
*Levels in
-------
TABLE 28. SECTOR GEOMETRIC MEANS OF TRACE ELEMENTS IN PUBIC HAIR*
°f Qbs". CCuFHLMPSeSn Zn
Max Min
Charlotte Whites 15-12 0.406 2.194 0.535 7.01 7.08 0.128 0.102 0.593 5.89 0.346 0.658 54.4
Charlotte Blacks 14-7 0.515 2.082 0.629 8.64 20.02 0.473 0.083 1.459 8.79 0.494 0.857 62.5
Birmingham Whites 21-16 0.455 2.120 0.949 6.71 9.06 0.294 0.057 0.910 5.14 0.325 1.006 65.9
Birmingham Blacks 18-7 0.409 2.153 1.332 9.80 11.47 0-852 0.036 2.130 8.66 0.749 0.626 79.3
Riverhead 19-12 0.252 1.205 1.015 13.91 21.80 0-955 0-128 0.959 5.32 0.526 0.252 100.5
Elizabeth 20-13 0.373 1.077 0.558 8.45 10.54 1.030 0.076 0.649 4.57 0.281 0.157 43.2
Ogden 11-4 0.166 0.669 1.333 9.16 39.69 5.003 .217 1.388 7.64 0.500 0.402 115.1
Salt Lake City 20-7 1.062 0.955 2.016 14.23 36.14 0-798 0.302 1.217 12.03 1.026 0.304 100.4
West Los Angeles 6-4 0.901 0.485 0.714 18.60 44.04 1.298 0.351 0.867 8.86 0.525 0.417 134.2
East Los Angeles 15-11 0.259 1.033 0.529 8.86 19.14 1.145 0.123 0.422 7.43 0.406 0.248 59.9
^Levels in pg/g.
-------
TABLE 29. TESTS OF SIGNIFICANCE OF THE EFFECT OF SELECTED FACTORS
ON MATERNAL BLOOD, CORD BLOOD AND PLACENTA TRACE ELEMENT LEVELS*
Maternal
Blood
Cord
Blood
Placenta
Without Controlling for City Effects
Smoking SES Age Parity
Cd - - -
Li
Fe - - +0.01
Ba - - -
B -0.5 - - -0.05
Cr - - -
Cd +0.025
Cu - - -
Zn - - -
Li
Se - +0.005
Ba - +0.05
B +0.01
Cr
Ni - - -0.05 +0.05
Ld - +0.05
Cu - -0.005
Zn
Li - -0.05
Se
Fe
Cr -0.025
Mn -0.05
Controlling for City Effects
Smoking SES Age Parity City Effects
<0.05
<0.001
+0.025
<0.01
<0.001
<0.005
<0.01
<0.01
<0.001
<0.005
+0.005
+0.01
<0.001
-
<0.001
<0.005
<0.05
<0.001
<0.01
<0.001
-0.05 -0.05
<0.001
Remarks: A negative sign appearing before the p-value indicates that the element level decreased with increased smoking,
higher socioeconomic status, older age groups or higher parity groups.
* Values given are the probability of the observed difference in sample geometric mean levels between categories assuming
no difference in the original population. Only values of 0.05 or less are listed. Elements omitted from the table had
no significant differences in any of the above comparisons.
-------
TABLE 30. TESTS OF SIGNIFICANCE OF THE EFFECT OF SELECTED FACTORS
ON SCALP HAIR AND PUBIC HAIR TRACE ELEMENT LEVELS*
en
en
Scalp
Hair
Pubic
Hair
Pb
Hg
Li
Se
Fe
Cr
V
Pb
Cu
Zn
Hg
Li
Se
Fe
B
Sn
Mn
Without Controlling for City Effects
Shampoo Usage
Smoking SES Age Parity Frequency of Dyes
-0.05 - ...
-0.025 -
-0.05 - ...
+0.005
-0.005 -
+0.025
-0.05
NO TEST NO TEST
+0.025
+0.05
-0.025
+0.005
_
-
_
_
- - -
Controlling for City Effects
Shampoo Usage City
Smoking SES Age Parity Frequency of Dyes Effects
_ _ - - - - _
0.025
0.01
+0.025
-
NO TEST NO TEST 0.001
0.025
-.
0.001
0.005
0.00]
0.005
0.05
0.005
-0.01
Remarks: The positive sign before p-values under the heading of Shampoo Frequency or Usage of Dyes indicates that
element levels tended to be higher among those women who shampooed more frequently or who used dyes.
* Values given are the probability of the observed difference 1n sample geometric mean levels between categories assuming
no difference in the original population. Only values of 0.05 or less are listed. Elements omitted from the table had
no significant differences in any of the above comparisons.
-------
TABLE 31. MEAN DUSTFALL TRACE ELEMENT CONTENT BY SECTOR*
Birmingham, Ala.
Charlotte, N.C.
Whites
Blacks
Salt Lake City,
Ogden, Utah
LA Basin, Calif.
West
East
Elizabeth, N.J.
Riverhead, N.Y.
Birmingham, Ala.
Charlotte, N.C.
Whites
Blacks
Salt Lake City,
Ogden, Utah
LA Basin, Calif.
West
East
Elizabeth, N.J.
Riverhead, N.Y.,
Birmingham, Ala.
Charlotte, N.C.
Whites
Blacks
Salt Lake City,
Ogden, Utah
LA Basin, Calif.
West
East
Elizabeth, N.J.
Riverhead, N.Y.
# of Months
Observed
12
12
11
Utah 11
12
1
4
3
12
12
12
12
Utah 11
12
1
4
3
12
12
12
11
Utah 11
12
1
4
3
12
Cadmium
Minimum
0.01
0
0.01
0.01
0.01
0.02
0.02
0.09
0
Lead
0
0.21
0.10
0.57
2.41
3.84
5.91
4.15
0.30
Zinc
5.45
1.07
0.75
1.04
1.19
4.38
3.00
9.43
0.19
Chromium Copper
# of Months # of Months
Maximum Mean Observed Minimum Maximum Mean Observed Minimum Maximum Mean
1.58 0.13 5 0.11 0.56 0.34 7 0.55 11.93 2.88
0.19 0.07 4 0 0.23 0.13 5 1.18 4.52 1.97
2.03 0.18 3 0.07 0.16 0.10 5 0.64 4.92 1.50
0.28 0.09 4 0.06 0.29 0.16 6 1.38 2.44 1.76
0.18 0.05 5 0.05 0.18 0.11 7 0.55 2.15 1.23
0.02 0.02 1 0.06 0.06 0.06 1 0.62 0.62 0.62
0.26 0.06 2 0.10 0.34 0.22 4 0.89 2.07 1.21
0.12 0.14 3 0 0.46 0.21 3 2.82 7.88 5.22
0.30 0.04 4 0.02 0.11 0.05 8 0.39 13.41 3.38
Manganese Nickel
19.82 7.73 6 2.23 12.89 6.59 7 0.01 0.89 0.30
8.92 4.80 5 1.31 2.87 1.86 6 0 0.50 0.35
9.86 4.62 6 0 2.55 0.73 6 0 0.30 0.09
17.94 6.94 6 0.63 2.34 1.45 6 0.02 0.55 0.30
6.98 5.02 7 0.32 3.13 1.38 7 0.01 0.46 0.19
3.84 3.84 1 0.61 0.61 0.61 1 000
7.39 7.06 4 1.22 1.68 1.51 4 0.10 0.31 0.17
14.00 9.18 3 0.87 1.06 0.98 3 0.41 0.52 0.47
4.39 1.74 8 0.05 0.70 0.35 8 0 0.76 0.21
84 42 32 50 *Dustfall expressed as mg/m2/month,
16.90 7.92
19.29 8.14
6.32 4.61
6.30 3.89
4.38 4.38
5.39 4.54
10.88 10.12
26.38 5.90
en
en
-------
TECHNICAL REPORT DATA
i'i U'G±c read Jnitnictioni, on ihe /vi > r/-;'j;v~
INCLASSIFIED
COSATi 1 idd/Group
06, T
21 NO. OF PAGES
62
22 PR'CE
EPA Form 2220-1 (9 73)
57
------- |