UNITED STATES ENVIRONMENTAL PROTECTION AGENCY

REGION IX


Response to the November 2005 National Stone, Sand & Gravel Association
Report Prepared by the R.J. Lee Group, Inc
“Evaluation of EPA’s Analytical Data from the El Dorado Hills Asbestos
Evaluation Project”

April 20, 2006

 United States Environmental Protection Agency Region 9
Response to the November 2005 National Stone, Sand & Gravel Association report
prepared by the R.J. Lee Group, Inc:
“Evaluation of EPA’s Analytical Data from the El Dorado Hills
Asbestos Evaluation Project”

This document constitutes the United States Environmental Protection Agency Region 9
(EPA Region 9) response to the major findings and conclusions of the National Stone, Sand &
Gravel Association report “Evaluation of EPA’s Analytical Data from the El Dorado Hills
Asbestos Evaluation Project” prepared by the R. J. Lee Group (R. J. Lee Report). A more
detailed analysis will be completed after additional information is received from the R. J. Lee
Group and the National Stone, Sand & Gravel Association,1 and the United States Geological
Survey (USGS).
The R. J. Lee Report draws conclusions that are contradicted by the El Dorado Hills data
and by generally accepted scientific principles for measuring asbestos exposure.
Overview
The R. J. Lee Group review of the EPA data was contracted by the National Stone, Sand
& Gravel Association. The El Dorado County Office of Education funded the three reviewers
who wrote letters in support of the R. J. Lee Report and whose reviews are included in this
response.
The EPA Region 9 El Dorado Hills Naturally Occurring Asbestos Exposure Assessment
was designed to measure the exposures to asbestos fibers, if any, that resulted from sports and
play activities that disturbed dust and soil. EPA Region 9 adhered to accepted EPA standards for
sampling and analysis, including rigorous quality assurance/quality control, and to the standard
methodologies of EPA exposure and risk assessment.
The R. J. Lee Report Criticizes EPA Region 9 for Using Established Scientific and
Public Health Protocols - In assessing naturally occurring asbestos exposures in El Dorado
Hills, EPA evaluated asbestos exposures using the PCME (phase contrast microscopy
equivalent) asbestos fiber size classification. The PCME classification was used because human
epidemiological studies, which form the basis of knowledge of asbestos health effects, measured
asbestos fiber concentrations using phase contrast microscopy (PCM) analytical methods.
PCME is the standard term for fibers counted by more modern analytical methods that are of
equivalent size to those fibers that would be seen by PCM analysis, and includes fibers with a
length to width aspect ratio of 3 to 1 or greater. EPA considered PCME fibers in our analysis of
the El Dorado data to be consistent with the existing health databases and risk assessment
1

On March 9, 2006, EPA Region 9 sent a letter to the R.J. Lee Group and the National
Stone, Sand, & Gravel Association asking for additional information to support the findings and
conclusions of the R.J. Lee Report.
-1-

 procedures used by EPA, California EPA (Cal/EPA), the World Health Organization, and other
federal agencies and international organizations. This approach was rejected by the R.J. Lee
Group, which instead advocates use of asbestos fiber definitions which are not health based or
supported by the majority of experts in the health community, and which would not allow
comparison to the existing epidemiologic data on asbestos related cancers.
The R. J. Lee Report Claims that EPA Region 9 Misapplied Fiber Counting
Protocols - The R. J. Lee Report claims that EPA Region 9 inflated the fiber counts in the El
Dorado Hills air data by misapplying the International Standards Organization (ISO) method
10312 (the analytical method used by EPA to analyze the El Dorado air samples) and including
PCME structures with a 3 to 1 length to width aspect ratio in our analysis. The R. J. Lee Report
maintains that EPA should only have counted structures which met the general 5 to 1 aspect ratio
fiber size definition described in the body of the ISO 10312 method. However, Annex C and
Annex E of the ISO 10312 method specifically authorize the counting of PCME structures with a
3 to 1 aspect ratio. Another example of misleading information is the R.J. Lee Report’s
statistical evaluation and resulting conclusions regarding the concentrations of asbestos
structures detected in the EPA air samples. All of the established EPA, National Institute of
Occupational Safety and Health (NIOSH), and ISO analytical methods require the counting of
asbestos bundles, recognizing the significance of bundles to proper characterization of asbestos
fiber levels. The R.J. Lee Report did not include asbestos bundles in its analysis of the data,
thereby undercounting the number of structures.
The R. J. Lee Report Claims that EPA Region 9 Misidentified Amphibole Minerals The R. J. Lee Report concludes that EPA misidentified actinolite asbestos fibers in the El
Dorado soil samples by using inappropriate extinction angle criteria. The R. J. Lee Group
conclusion is contradicted by the National Institute of Standards and Technology (NIST) and the
major analytical methods used for analysis of asbestos in soil and bulk samples. The R. J. Lee
Report also cites an unpublished 1980 draft report to support its contention that structures found
in the EPA air samples are not asbestos, and ignores a subsequent 1981 published report by the
same author that actually supports the EPA approach.
The R. J. Lee Report Applies a Geologic Definition rather than a Public Health
Definition to Characterize Microscopic Structures - The R. J. Lee Report relies heavily on
the geologic distinction between asbestos fibers and cleavage fragments of the same dimensions,
with the implication that exposure to cleavage fragments is benign and of little or no health
significance. For the purposes of public health assessment and protection, EPA makes no
distinction between fibers and cleavage fragments of comparable chemical composition, size,
and shape. The EPA Region 9 approach, which is supported by most public health agencies and
scientists, as well as the American Thoracic Society, is based on the following: (1) The
epidemiologic and health studies underlying EPA and Cal/EPA cancer risk assessment methods
were based on exposures to both cleavage fragments and fibers, and were unable to distinguish
between the two, (2) The most recent panel of experts to review asbestos risk assessment
methods, the 2003 Peer Consultation Panel convened by EPA, concluded that “it is prudent at

-2-


 this time to conclude equivalent potency [of cleavage fragments and fibers] for cancer,”2 (3) No
well-designed animal or epidemiological studies have adequately tested the hypothesis that
cleavage fragments with the same dimensions as a fiber are benign or that the human body
makes any distinction, (4) Studies that purport to show that cleavage fragments are benign are
questioned by many asbestos health experts, (5) There are no routine asbestos air analytical
methods, including those used by EPA, NIOSH, the Mine Safety and Health Administration
(MSHA), the American Society for Testing and Materials (ASTM), and ISO which differentiate
between cleavage fragments and crystalline fibers on an individual fiber basis.
The R. J. Lee Report’s “Virtual” Review of EPA Region 9’s Air Samples is
Inconsistent with Established Laboratory Practices - The R.J. Lee Group did not have access
to EPA’s actual air samples, nor did it collect any air samples of its own. Rather it reviewed
limited pictures and spectra data of a small number of EPA’s air samples and drew conclusions
based on those representations. Such a virtual review is not consistent with the National
Voluntary Laboratory Assurance Program (NVLAP) quality assurance procedures nor the
verification methods of the National Institutes of Standards and Technology.
Federal Courts Have Supported EPA - Many of the assertions of the R. J. Lee Report
are consistent with positions that the R.J. Lee Group took as an expert witness for W.R. Grace in
the Libby, Montana litigation. In this litigation, the written opinions of the District and Appeals
courts, while not specifically addressing the opinions of the R.J. Lee Group, rule in favor of EPA
and expressly hold that EPA’s experts and science are credible.3
Background
In October 2004, the EPA Region 9 Superfund site assessment program conducted an
assessment of exposures to naturally occurring asbestos (NOA) in El Dorado Hills, California.
Specifically, EPA Region 9 simulated the sports activities of children and adults at three schools
and a community park and, using personal air monitors, measured asbestos levels in the
breathing zones of participants. EPA Region 9 also collected samples of ambient air in the area
of the sampling at the same time the simulations were conducted to serve as reference samples.
The personal activity-based samples were then compared to the reference samples. The
Asbestos Hazard Emergency Response Act (AHERA)4 regulation Z-test for statistical

2

USEPA (U.S. Environmental Protection Agency) (2003). Report on the Peer
Consultation Workshop to Discuss a Proposed Protocol to Assess Asbestos-Related Risk, Final
Report. Office of Solid Waste and Emergency Response, Washington D.C. Page viii.
3

See U.S. v. W.R. Grace, 280 F Supp 2d 1149 (2003): U.S. v. W.R. Grace, 429 F. 3d
1224, 1245 (9th Cir. 2005) (Although debate regarding testing methodology and data analysis is
“exceedingly complex”, EPA did not ignore accepted scientific principles)
4

The Asbestos Hazard Emergency Response Act (AHERA) was passed by Congress in
1986 to provide for the inspection and mitigation of asbestos in school buildings. Regulations
implementing the Act were promulgated by EPA in 1987.
-3-

 significance was applied to determine whether there were any statistically significant differences
between the personal exposure samples and the ambient reference samples. EPA Region 9
collected over 400 air samples and generated over 7000 data points. All of EPA Region 9's’s
analyses were conducted by accredited laboratories using recognized methods and procedures
with strict quality assurance control, including blind performance samples to check analytical
accuracy.
Amphibole asbestos, which many health scientists consider to be even more toxic than
chrysotile asbestos, was found in almost all the reference and activity-based samples. Of the 29
different sets of activity-based scenario measurements, application of the Z-test determined that
personal exposures from 24 scenarios were significantly elevated over the reference samples.
Most importantly, the data showed that children and adults participating in sports activities in
areas where asbestos occurs naturally in the surface soils, as it does in El Dorado Hills, can be
exposed to asbestos fibers of health concern at up to 62 times the corresponding reference levels.
EPA Region 9 released the data from the assessment in May 2005 and held a public
meeting in El Dorado Hills that was attended by more than 1000 members of the public. From
the outset of the assessment, EPA Region 9 made clear to the community that EPA’s only intent
was to gather data on potential exposures. The community and the State and local regulatory
agencies could then use the information to make decisions about the significance of those
exposures and determine appropriate control measures. Both EPA Region 9 and the Agency for
Toxic Substances and Disease Registry (ATSDR) have informed the community that exposure
levels are a main determinant of the risk of developing asbestos-related cancers and non-cancer
diseases, and that reducing the exposures reduces the risk. Consistent with its intent, EPA
Region 9 has actively engaged the State and local regulatory agencies to improve naturally
occurring asbestos mapping, monitoring, dust control, and regulation. El Dorado County has
recently adopted more stringent dust control ordinances.
Detailed Comments on the R. J. Lee Report
R.J. Lee Finding #1: “Based on Mineralogy, Sixty-Three Percent (63%) of the Amphibole
Particles Identified as Asbestos Fibers can not be Asbestos.”
The R. J. Lee Report argues that there is too much aluminum in 63% of EPA Region 9's
identified fibers for the fibers to be asbestiform.5 In addition, the remaining 37%
(sometimes the Report uses 35%) are not asbestos fibers based on their particle
dimensions.
EPA Response
Aluminum - Analysis of the EPA Region 9 El Dorado air samples was performed using
the International Standards Organization (ISO) method 10312, a state-of-the-art

5

Asbestiform: Having the form or structure of asbestos.
-4-

 Transmission Electron Microscope (TEM)6 method with energy dispersive spectroscopy
(EDS)7 that has strict counting rules and characterizes the dimensions and chemistry of
every fiber identified by the microscopist. Identification of fiber type was performed
according to the general guidelines of the International Mineralogical Association (IMA)
(Leake, 1997)8, the international standard for amphibole nomenclature. This same
approach for asbestos classification is recommended in the “Research Method for
Sampling and Analysis of Fibrous Amphibole in Vermiculite Attic Insulation”, EPA
600/R-04/004, January 2004, and was one of the tools used by Meeker et al (2003)9 to
determine the composition and morphology of amphiboles from Libby, Montana.
The R. J. Lee Report claims that 63% of the amphibole fibers identified by the EPA
laboratory10 as actinolite asbestos have concentrations of total aluminum that are too high
to form asbestos fibers. According to page 2 of the R. J. Lee Report, “Particles with
more than 0.3 aluminum atoms pfu [per formula unit] or about 1.5 percent Al O cannot
form in the asbestos habit due to crystal lattice constraints.” To support its argument, the
R. J. Lee Report cites three references. However, on close examination, two of the three
references do not agree with the upper threshold limit that the R.J. Lee Group puts on
total aluminum content (Leake et al, 1997) (Deer, Howie and Zussman, 1997)11. The
third reference (Verkouteren & Wylie, 2000)12 draws its conclusions on examination of a
2

3

6

Transmission Electron Microscopy (TEM) produces images of a sample by illuminating
the sample with an electron beam in a vacuum, and detecting the electrons that are transmitted
through the sample.
7

Energy Dispersive Spectroscopy (EDS) uses measurement of the energy and intensity of
X-rays generated when a selected area of a sample is irradiated with an electron beam to identify
the mineralogical composition of a structure.
8

B.E. Leake et al (1997). Nomenclature of Amphibole: Report of the Subcommittee on
Amphiboles of the International Mineralogical Association, Commission on New Minerals and
Mineral Names. American Mineralogist, Volume 82, pages 1019-1037.
9

G.P. Meeker et al (2003). The Composition and Morphology of Amphiboles from the
Rainy Creek Complex, Near Libby, Montana. American Mineralogist, Volume 88, pages 19551969.
10

In this document, the terms “EPA laboratory” and “EPA Region 9 laboratory” refer to
the private laboratories that conducted the analysis of the EPA soil and air samples under
contract to EPA Region 9.
11

W.A. Deer, R.A. Howie, and J. Zussman (1997). Rock-Forming Minerals: Double
Chain Silicates, Vol 2, second edition, p 137 - 145.
12

J.R. Verkouteren and A.G. Wylie (2000). The Tremolite-Actinolite-Ferro-Actinolite
Aeries: Systematic Relationships Among Cell Parameters, Composition, Optical Properties, and
-5-


 small set of fibrous actinolite asbestos samples which the authors partition into asbestos
and fibrous “non-asbestos” byssolite using criteria which the IMA specifically
recommends against, and which is inconsistent with all standard asbestos analytical
methods. Perhaps most important is the fact that all three references agree that it is the
IMA criteria which primarily govern the general classification of amphibole type, not the
total aluminum content. These references therefore actually support the classification
approach taken by the EPA laboratory.
The R.J. Lee Group did not have access to the EPA air samples to conduct their own
analyses. Instead, the R.J. Lee Group looked at a limited number of photographs of the
recorded EDS spectra. Interferences by other elements in the sample can affect the
aluminum total in the spectra. This is especially important because the EPA samples
were of air releases from soil, not processed asbestos material. Soils contain nonasbestos mineral and biological particles that can influence element totals in an EDS
spectrum, most notably clay particles, which are high in aluminum. The laboratory used
by EPA Region 9 identified aluminum-rich actinolite asbestos, by applying the IMA
classification guidelines to its direct analysis of the actual sample.13
Particle Dimension - As previously stated, the R. J. Lee Report claims that 37% of the
fibers counted by EPA in the El Dorado Hills air samples are not asbestos fibers based on
their particle dimensions. The report claims that EPA Region 9 inflated the fiber counts
by including asbestos structures which do not meet the definition of a fiber as described
in ISO 10312. The general ISO 10312 method requires the counting of every asbestos
structure with a length to width aspect ratio of 5:1 or greater. As directed by Region 9,
the EPA laboratory counted structures with a 3:1 or greater aspect ratio. The R. J. Lee
Report states that EPA erred in counting structures with aspect ratios less than 5:1.
Annex C and Annex E of the ISO method clearly authorize the counting of PCME
structures with a 3:1 aspect ratio if the data are to be used for exposure or risk
assessment purposes, the stated goal of the El Dorado Hills assessment. In fact, the
ISO method contains numerous references to PCME fibers. PCME fibers are
defined as fibers greater than 5 microns in length, and 0.25 to 3 microns in width
with a 3:1 aspect ratio.14 PCME fibers form the basis for EPA’s IRIS toxicity
database and the asbestos risk models of California EPA and other federal and
international organizations.15

Habit, and Evidence of Discontinuities. American Mineralogist, 85, p. 1239 - 1254.
13

Personal communication with John Harris, Lab/Cor, January 2006.

14

World Health Organization (1986). Environmental Health Criteria 53, International
Programme on Chemical Safety, Asbestos and Other Natural Mineral Fibres, section 2.3.2.2.
15

The IRIS asbestos cancer inhalation unit risk, a measure of asbestos cancer potency, is
based on the EPA 1986 Airborne Asbestos Health Assessment Update (EPA/600/8-84/003F;
1986). Cal/EPA used a similar approach and data sets to derive its cancer unit risk. Both the
IRIS and the Cal/EPA cancer potency values rely on human epidemiological studies that were
conducted using phase contrast microscopy (PCM) analytical methods (some were midget
-6-


 The R.J. Lee Group also manipulates its statistical analysis of the El Dorado Hills air data
by ignoring counts of asbestos fiber bundles in its evaluations. Bundles are two or more
attached parallel asbestos fibers which can have a significant health impact when they are
inhaled and separate into individual fibers. Bundles were counted in the historical
epidemiological studies which form the basis of our knowledge of asbestos-related health
effects and EPA’s IRIS database. All of the established EPA, NIOSH, and ISO
analytical methods require the counting of asbestos bundles, recognizing the
significance of bundles to proper characterization of asbestos fiber levels.
The R. J. Lee Report further states that EPA’s data inflated the asbestos fiber count by
ignoring the Agency’s own “definition” of asbestos. To support this claim, the R.J. Lee
Report cites the glossary of “Method for Determination of Asbestos in Bulk Building
Materials”, EPA 600/R-93/116, 1993, which states, in part, “With the light microscope,
the asbestiform habit is generally recognized by the following characteristics: Mean
aspect ratios ranging from 20:1 to 100:1 or higher for fibers longer than 5 microns.” The
building material analytical method is designed to detect commercially processed
asbestos in items like floor tiles, roofing felts, paper insulation, paints, and mastics, not
naturally occurring asbestos on air filters or in soil samples. To present the 20:1 aspect
ratio for commercial grade asbestos as a universal EPA policy, and to advocate its use as
an appropriate standard for analyzing air samples of naturally occurring asbestos is
inappropriate and contradictory to use of the PCME dimensional criteria as a tool for
assessing exposure risk.
The R. J. Lee Report also states that the diffraction pattern analyses produced by the EPA
laboratory for the El Dorado Hills air samples demonstrates that the particles identified
by the laboratory are not asbestos.16 The report cites a 1980 unpublished draft study by
S.J. Ring to support its conclusion. The R. J. Lee Report does not mention a 1981
published article by the same author which revises the findings such that they no longer
support the conclusion of the R. J. Lee Report and, in fact, support the data produced by

impinger data converted to PCM counts) that could not distinguish fibers that were 5 microns in
length or less. PCM cannot distinguish between fibers and cleavage fragments. PCM is not as
powerful as current Transmission Electron Microscope (TEM) methods (400X vs 20,000X) as
TEM can see the thinner/shorter fibers. However, since EPA's (and Cal/EPA 's) toxicity
database relies on human health studies that used PCM, current EPA risk procedures use the
more powerful TEM method but report the PCM equivalent (PCME) fibers and only use the
PCME counted fibers in a risk assessment. This is because the IRIS asbestos file specifies that
only PCME fiber counts be used with inhalation unit risk for risk calculation. See also the
reference cited in footnote 11.
16

Diffraction pattern analyses irradiates a sample with x-rays and then takes an x-ray
photograph.
-7-

 EPA.17
R.J. Lee Finding #2: “The Laboratory Procedures did not Comply With the NVLAP
Quality Assurance Standard.”
The R. J. Lee Report says that the false positive rate in our air samples was 35% when
the acceptable limit in the National Voluntary Laboratory Accreditation Program
(NVLAP) is 10%.
EPA Response
The laboratories used by EPA Region 9 for analysis of the El Dorado Hills air and soil
samples are accredited through the National Voluntary Laboratory Accreditation
Program (NVLAP). NVLAP is administered by the National Institute of Standards and
Technology, a non-regulatory agency within the U.S. Commerce Department. A large
part of the accreditation process involves on-site audits performed by NVLAP-certified
inspectors who review laboratory operational and quality assurance compliance
parameters, including documentation proving compliance with NVLAP requirements for
verification analyses. A laboratory must demonstrate that all analysts reporting data meet
the false negative and false positive requirements set forth by NVLAP before an
accreditation certificate is issued. To make a determination that a laboratory did not
comply with NVLAP verification standards would require a very detailed examination of
all laboratory generated raw data, project specific information, such as a site-specific
EPA issued Quality Assurance Project Plan, laboratory instrument log books, and other
data and information not supplied in an analytical report. Interviews with the laboratory
manager, quality assurance manager, and involved analysts are also mandatory to make
judgement on a laboratory’s possible non-compliance. The R.J. Lee Report’s conclusion
that the EPA laboratory was not in compliance with NVLAP, based on a cursory review
of count sheet and other limited data without the in-depth examination detailed above, is
therefore invalid and cannot be used to question EPA’s analytical results.
EPA chose NVLAP-accredited laboratories for the El Dorado Hills assessment as a
minimum quality requirement. For supplemental quality assurance, the laboratories were
subjected to on-site audits performed by EPA’s Quality Assurance Technical Support
group, and both laboratories were sent performance evaluation samples prior to analysis
of the El Dorado samples. In addition, the laboratory conducting the air sample analysis
was sent double blind performance evaluation samples during the sampling event. In all
cases, the laboratories successfully identified the amounts and types of asbestos present
on the blind samples within acceptable limits. Further, the El Dorado Hills air and soil
data were validated by a third party in accordance with standard EPA quality assurance
17

S.J. Ring (1981). Identification of Amphibole Fibers, Including Asbestos, Using
Common Electron Diffraction Patterns. In Russell P.A. and Hutchings A.E. (Eds), Electron
Microscopy and X-ray Applications to Environmental and Occupational Health Analysis, Vol.
2:175-198, Ann Arbor Science Publ., Inc.
-8-

 procedures and were found to be acceptable for all uses.
R. J. Lee Finding #3:“The Soil Samples do not Demonstrate the Presence of Amphibole
Asbestiform Minerals.”
The R. J. Lee Report states that the actinolite asbestos fibers identified in the El Dorado
Hills soil samples contain too much aluminum to be asbestiform and that the extinction
angles of the fibers indicate that they are non-fibrous cleavage fragments. The R.J. Lee
Group’s analysis of 23 split soil samples from EPA’s October 2004 sampling event found
no asbestos in the samples.
EPA Response
Aluminum - The R. J. Lee Report states that the aluminum content of the fibers in the
soil samples was too high to be asbestiform actinolite and that it was indicative of nonasbestiform actinolite and another amphibole, hornblende, which contains approximately
10-20% by weight Al2O3 (5.3-10.6% by weight aluminum). Both the laboratory
performing EPA’s El Dorado soil sample analysis and the laboratory which analyzed the
EPA air samples noted significant quantities of hornblende in the samples, but did not
count or report those particles as asbestos. Please see the EPA response to Finding #1 for
a further discussion of the aluminum issue.
Extinction Angles - The extinction angle of a fiber evaluated by polarized light
microscopy is one of many criteria used to identify mineralogical composition. The
extinction angle for amphibole asbestos fibers is the difference in degrees between the
long axis of the fiber and the angle at which the fiber optically disappears (the
polarization direction where the light passing through it becomes “extinct”) when the
fiber is rotated under a polarized light microscope. The R.J. Lee Report states that
amphibole asbestos fibers have a zero-degree extinction angle and that non-asbestos
cleavage fragments have non-zero extinction angles. Therefore, because the EPA soil
sample analysis reported extinction angles which, according to the R.J. Lee Group,
averaged 12o, the report alleges EPA incorrectly identified cleavage fragments as
asbestos fibers.
The R.J. Lee Report’s conclusion regarding extinction angles is contradicted by the
National Institute of Standards and Technology (NIST) and the major analytical
methods used for analysis of asbestos in soil and bulk samples. NIST certifies and
provides Standard Reference Materials (SRM) for laboratory instrument calibration and
laboratory accuracy measurement. The NIST Tremolite/Actinolite SRM 1867A is a
special set of three samples certified by NIST to be of ultra-high purity tremolite,
actinolite, and anthophyllite asbestos and is considered the “gold standard” for asbestos
analytical laboratories. The material is rigorously characterized and is accompanied by a
six-page document that describes the properties of each sample. It is required that all
analytical laboratories accredited by NIST/NVLAP have the material in their possession
and that they use it to calibrate their operations and to test their analysts. The NIST SRM
-9-


 1867A certificate which accompanies the samples of tremolite and actinolite states that
the reference tremolite can have an extinction angle of up to 16.6 + 0.3o and that the
actinolite can have an extinction angle of up to 15.9+0.2o. When the EPA laboratory
processed the NIST actinolite standard in the manner of the El Dorado Hills soil samples,
the extinction angles of the fibers in the processed standard sample were consistent with
allowed maximum extinction angles for tremolite/actinolite asbestos (~ 10o to 20o) and
the extinction angles of the fibers seen in the EPA soil samples.18
Further, the laboratory methods of EPA, NIOSH, and other agencies for analysis of
asbestos in bulk material all state that tremolite-actinolite asbestos fibers may have zero
(parallel) or non-zero (inclined or oblique) extinction angles. EPA Method 600/R93/11619, the standard method used by all NIST/NVLAP accredited laboratories to test
building materials for the presence of asbestos, states in Table 2-2, Optical Properties of
Asbestos Fibers, that tremolite-actinolite asbestos has extinction “parallel and oblique (up
to 21o).” NIOSH Method 900220, the method used for analysis of the El Dorado Hills soil
samples, states directly that actinolite and tremolite fibers exhibiting inclined extinction
are to be considered asbestos. The method further states that “If anisotropic fibers are
found (during PLM analysis), rotate the stage to determine the angle of extinction.
Except for tremolite-actinolite asbestos which has oblique extinction at 10-20o, the other
forms of asbestos exhibit parallel extinction... Tremolite may show both parallel and
oblique extinction.”21
R.J. Lee Finding #4: “The ISO 10312 Analytical Method can not Distinguish Between
Asbestos Fibers and Non-Asbestos Cleavage Fragments.”
The R.J. Lee Report states that the ISO 10312 method contains the disclaimer that “The
method cannot discriminate between individual fibers of asbestos and non-asbestos
analogues of the same amphibole material,” and, therefore, EPA inflated the asbestos air
concentrations by counting “cleavage fragments.”
EPA Response
The ISO 10312 method cannot differentiate between fibers and cleavage fragments with

18

M. Bailey (2006). Identification of Asbestiform Tremolite/Actinolite. Naturally
Occurring Asbestos Workgroup Meeting Presentation.
19

USEPA (U.S. Environmental Protection Agency) (1993). Method for the
Determination of Asbestos if Bulk Building Materials. EPA Method 600/R-93/116.
20

NIOSH (National Institute for Occupational Safety and Health) (1992). Asbestos
(Bulk) by PLM.. Method 9002 (Issue 2).
21

NIOSH (National Institute for Occupational Safety and Health) (1992). Asbestos
(Bulk) by PLM.. Method 9002 (Issue 2). Qualitative Assessment, Item c, page 4.
-10-

 the same dimensions and chemical composition. No routine analytical method has a
protocol for distinguishing fibers from cleavage fragments on an individual particle basis.
Additionally, from a health standpoint, there is no evidence that supports making the
distinction.
Cleavage fragment is a geologic term which refers to structures that form when nonfibrous forms of asbestos minerals split along crystallographic planes, as opposed to
asbestos fibers which form from crystalline growth. The R.J. Lee Report maintains that
there is a toxicological difference between asbestos structures which formed as fiber
crystals and fibers which formed by cleavage plane separation. Page 3 of the R.J. Lee
Report states that cleavage fragments are “not known to produce asbestos-like disease.”
It is the position of EPA, the U.S. Centers for Disease Control and Prevention,
Agency for Toxic Substances and Disease Registry (ATSDR) and National Institute
for Occupational Safety and Health (NIOSH), and the American Thoracic Society,
among others, that microscopic structures of amphibole and serpentine minerals
that are asbestiform and meet the size definition of PCM fibers, should be counted
as asbestos, regardless of the manner by which they were formed. There are four
reasons why the health agencies have taken this position: (1) The epidemiologic and
health studies underlying EPA, and California EPA, cancer risk assessment methods were
based on exposures to both cleavage fragments and fibers, but were unable to distinguish
between the two, (2) The most recent panel of experts to review asbestos risk
assessment methods, the 2003 Peer Consultation Panel convened by EPA, concluded that
“it is prudent at this time to conclude equivalent potency [of cleavage fragments and
fibers] for cancer,”22 (3) No well-designed animal or human epidemiological studies
have been conducted to date to test the hypothesis that cleavage fragments with the same
dimensions of a fiber are benign, or that the human body makes any distinction, and
studies that purport to show that cleavage fragments are benign are questioned by many
asbestos health experts,23 (4) There are no routine air analytical methods, including those
used by EPA, NIOSH, the Mine Safety and Health Administration (MSHA), the
American Society for Testing and Materials (ASTM), and the ISO which differentiate
between cleavage fragments and crystalline fibers.

22

USEPA (U.S. Environmental Protection Agency) (2003). Report on the Peer
Consultation Workshop to Discuss a Proposed Protocol to Assess Asbestos-Related Risk, Final
Report. Office of Solid Waste and Emergency Response, Washington D.C. Page viii.
23

Both Addison (Addison J, Davies LST. 1990. Analysis of amphibole asbestos in
chrysotile and other minerals. Ann Occ Hyg, Apr;34(2):159-75) and members of the U.S. EPA
2003 Peer Consultation panel raised concerns about interpretation of the Davis study (Davis JM,
McIntosh C, Miller BG, Niven K. 1991. Variations in the carcinogenicity of tremolite dust
samples of differing morphology. Ann NY Acad Sci, Dec;643:473-90 ), which attempted to
compare the toxicity of asbestos fibers and cleavage fragments. These concerns reflected the
lack of peer review, use of intra peritoneal injection instead of inhalation exposure, significance
of mesotheliomas caused by structures reported as cleavage fragments, purity of the cleavage
fragment samples and issues related to fiber dimensions.
-11-

 In terms of epidemiological data and health outcomes, the cleavage fragment argument is
without merit. For the purposes of public health assessment and protection, EPA makes
no distinction between fibers and cleavage fragments of comparable chemical
composition, size, and shape.
There are no recognized analytical protocols, including those used by EPA, NIOSH,
MSHA, ASTM, and ISO, which include criteria to differentiate between cleavage
fragments and crystalline fibers. All these methods require that structures which meet
their definition of the specific counting rules for an asbestos fiber be counted. The
requirements are based on the fact that, in the words of an expert from the United States
Geological Survey, “At a microscopic level, distinguishing between these forms on
single [asbestos] particles, can be extremely difficult to impossible.”24 As noted above,
R.J. Lee made a very similar claim with regard to cleavage fragments as the expert
witness for W.R. Grace in the Libby, Montana, Superfund cost recovery litigation. The
EPA analytical experts who reviewed the R.J. Lee Group’s testing methodology related
to the Libby site found that the R.J. Lee laboratory could not demonstrate any reliable
criteria with which to distinguish, at the microscopic level, asbestos cleavage fragments
from asbestos fibers of the same size, shape, and composition. The Ninth Circuit Court
of Appeals recognized the competing scientific arguments but found that EPA’s position
was consistent with the record of evidence and accepted scientific principles.25
R.J. Lee Finding #5: “Applying the Latest Science and Definitional Techniques, the El
Dorado Hills Study Shows no Significant Exposure to the Type of
Amphibole Asbestos Fiber Connected To Health Risk.”
The R. J. Lee Report claims that the latest science for measuring the risk posed by
asbestos is the Berman-Crump Asbestos Risk Assessment Protocol (“Berman-Crump”)
which proposes that amphibole asbestos fibers which are more than 10 microns long and
less than 0.5 microns wide (protocol fibers) are the most toxic. Of the 2,386 fibers which
the R. J. Lee Report states the EPA laboratory identified, the R.J. Lee Report concludes
that only 7 fibers meet the “Berman-Crump” definition. Therefore, the R.J. Lee Group
maintains that EPA has overstated the risk from exposure to asbestos fibers in El Dorado
Hills.
EPA Response
The “Berman-Crump” protocol that the R.J. Lee Report references is in fact a draft EPA
method. EPA had the method reviewed by a peer consultation panel in 2003. The panel
made a number of important recommendations that must be addressed before the method
can be used for EPA risk assessments. A number of important revisions have been made
24

G.P. Meeker, USGS, (2002). Review of Expert Report of R.J. Lee.

25

U.S. v. W.R. Grace, 429 F.3d at 1245.
-12-

 to the draft method since 2003, but at this time the method has not been independently
peer reviewed. It will not be adopted by EPA as a risk assessment tool unless and until it
passes rigorous internal and external peer review.
The expert peer panel has recommended that the fiber size for the draft EPA risk
assessment method be adjusted to include fibers greater than 5 microns in length and up
to 1.5 microns in width.26 The change is designed to account for lung deposition of fibers
that results when fibers are inhaled through the mouth, and not filtered by the nasal
passages. The broadening of the fiber definition to include inhalation by “mouth
breathers” is especially relevant to the El Dorado Hills data. Our investigation measured
personal asbestos exposures of individuals participating in sports activities, where
physical exertion would likely increase breathing through the mouth. The PCME fibers
counted in the EPA air samples are actually consistent with the latest science of
EPA, as reflected in the recommendations of the peer consultation panel. In
addition, the EPA peer consultation expert panel recommended that cleavage fragments
be treated as any other asbestos fiber of the same morphology and chemical
composition.27
EPA Region 9 focused on obtaining an accurate count of PCME structures, consistent
with our risk assessment protocols and those of Cal/EPA and other health agencies. The
counting rules which EPA set for the laboratory were designed to stop counting when a
statistically-significant number of PCME fibers were detected. By concentrating on
PCME structures, other fiber size classifications may not have been counted to statistical
significance. This may have resulted in under counts of other fiber sizes (e.g. the
“Berman Crump” protocol fibers referred to in the R. J. Lee Report). EPA Region 9's
study counted PCME structures so that the data could be directly compared to
human health epidemiological studies. These epidemiological studies form the basis
for risk assessment models currently used by EPA, Cal/EPA and other federal agencies
and international organizations.
R. J. Lee Report Peer Reviews
The R. J. Lee Report was reviewed by three individuals, although research of one of the
individuals was extensively quoted in the report and therefore the independence of the
reviewer is debatable. The three reviewers generally agree with the conclusions of the R.
J. Lee Report regarding aluminum content, fiber chemistry, cleavage fragments, and
extinction angles.
Both the R. J. Lee Report and one of the reviewers support use of the original “Berman26

USEPA (U.S. Environmental Protection Agency) (2003). Report on the Peer
Consultation Workshop to Discuss a Proposed Protocol to Assess Asbestos-Related Risk, Final
Report. Office of Solid Waste and Emergency Response, Washington D.C. Page 5-5.
27

Ibid, page 5-1.
-13-

 Crump” protocol and calculate a “Berman-Crump” fiber air concentration of 0.0002
fibers/cubic centimeter, using the EPA fibers which they assert meet the “BermanCrump” definition. The peer reviewer then compares that concentration with an ambient
concentration of 0.0008 fibers/milliliter measured in New York City, and states that the
“Berman-Crump” value in El Dorado Hills is extremely low. This comparison is flawed
for at least two reasons. Significantly, the New York City numbers are based on fibers
counted against a totally different size classification (essentially comparing apples to
oranges), but the reviewer also fails to recognize that a concentration of 0.0002 f/cc
translates in the protocol to an increased cancer risk of 1 in 1,000 exposed
individuals. This number is disturbingly high and is outside the acceptable cancer risk
ranges of EPA, Cal/EPA, and most other state and federal health agencies.
Conclusions
EPA Region 9 has carefully reviewed the R. J. Lee Report and believes that it makes
largely unsupported and incorrect conclusions about the EPA Region 9 El Dorado Hills
Naturally Occurring Asbestos Exposure Assessment. EPA Region 9 has asked the
United States Geological Survey (USGS) to conduct an independent study of the El
Dorado County area to address several mineralogical questions raised by the R. J. Lee
Report. The USGS study will use sophisticated analytical techniques (such as electron
probe micro analysis) to more completely characterize the naturally occurring asbestos in
terms of mineral identification and particle morphology.
All of the EPA Region 9 work in El Dorado Hills was, and continues to be, consistent
with the EPA’s standard operating and quality control procedures for asbestos work
throughout the country.

-14-