U. S. RADIUM SUPERFUND SITE: USING LESSONS-LEARNED
FROM EXPERIENCE AT ANOTHER SITE TO EXPEDITE
ENVIRONMENTAL RESTORATION

John Prince
U.S. Environmental Protection Agency

Tom Urbaniak, P.E.
U. S. Army Corps of Engineers

Charyl Fines, P.E.
CDM Federal Programs Corporation

ABSTRACT

The U. S. Radium Corporation Superfund Site, and the Montclair/West Orange and Glen Ridge Radium Superfund Sites in Essex County, New Jersey, encompass a densely-populated area of commercial and residential properties. These sites are contaminated with radiological wastes, probably from the U. S. Radium facility that operated in the area from 1917 to 1926. The company refined radium from carnotite ore and manufactured many products containing radium, including a zinc-based luminous paint. Process wastes were used as backfill material for properties in the cities of Orange, West Orange, and South Orange, and as fill material for new housing developments in Glen Ridge, West Orange, and Montclair.

Several study areas for the site were defined: the former U.S. Radium facility; a nine-block area around the facility; and a number of "satellite" locations identified through various means, including review of company records and an aerial gamma overflight survey. Investigations of the approximately 400 properties targeted are ongoing; of the 350 properties investigated to date, 134 have been found to be contaminated.

In 1994, the U. S. Army Corps of Engineers (USACE) contracted with CDM Federal Programs Corporation (CDM) to perform remedial design work, beginning with the first 59 properties that were to be remediated. The cleanup of the site is being sequenced, such that remedial action has begun on the first properties, while the remedial design is underway on upcoming phases of remedial action, and investigations to identify other properties are ongoing.

Working with USACE, the U. S. Environmental Protection Agency (EPA) initiated remedial design and remedial construction work at the Montclair/West Orange and Glen Ridge (MWG) sites in 1990. After gaining four years of experience at these sites in contamination delineation, construction work sequencing, and community relations, EPA and USACE began remedial design work at the U. S. Radium site. Because the nature and extent of the contamination at this site are well defined and similar remedial design and remedial action work had been performed at the neighboring Essex County, New Jersey, radium sites, EPA was hoping to expedite the cleanup process while reducing the remedial design effort by applying lessons learned.

SITE HISTORY

The Radium Luminous Materials Corporation, renamed U. S. Radium Corporation in 1921, refined radium from carnotite ore and manufactured a number of products containing radium, including a zinc-based luminous paint. The refining facility processed approximately 2 tons of carnotite ore daily, with each ton yielding 5 to 7 milligrams of radium. The ore, containing 2 to 4 percent uranium oxide, originated from Paradox Valley, Colorado.

Process wastes were temporarily stored at an inactive part of the facility, and later disposed of off site. It is likely that these process wastes were used as fill material at several residential and commercial properties in the City of Orange, as well as fill material for housing development in the towns of Montclair, Glen Ridge, and West Orange, New Jersey.

Radium processing operations ceased at the plant in 1926. The U. S. Radium Corporation stopped handling ore at the facility due to the availability of higher grade ore from imported sources. At this same time, several court cases alleged that illnesses afflicting plant employees were caused by exposure to radioactive materials at the plant.1

In 1978, the New Jersey Department of Environmental Protection (NJDEP) was informed that EPA had identified the U. S. Radium Corporation facility as a potential hazardous waste site. The NJDEP Bureau of Radiation Protection (BRP) performed several radiological surveys at the U.S. Radium site, including a preliminary radiation survey in March 1979 that indicated further investigation was necessary. In June 1979, BRP began a detailed investigation of the site that included the initiation of a long-term gamma radiation monitoring program; collection of soil samples from adjacent properties and building material samples from two buildings at the High and Alden Street Properties; and a radon monitoring program. Gamma radiation monitoring indicated extensive areas of elevated gamma readings. All but two of the soil samples collected exhibited elevated Radium-226 activity. Average measurements of radon and radon decay products obtained during this investigation exceeded nonoccupational criteria of 0.02 Working Levels (WL), roughly equivalent to 4.0 picoCuries per liter (pCi/l) at 50% equilibrium.

In May 1981, at NJDEP's request, EPA conducted an aerial gamma radiation survey of the area surrounding the facility. The results of this survey, along with the findings of subsequent investigations, led to the inclusion of the site on the federal Superfund National Priorities List (NPL) in the fall of 1983.

SITE DESCRIPTION

The U.S. Radium site is in the City of Orange, Essex County, New Jersey. The area around the site consists of commercial, light industrial, and residential properties. The site itself consists of three groups of properties, the U. S. Radium Corporation property, Vicinity Properties, and Satellite Properties. The former U. S. Radium Corporation Property is located in Orange, occupying approximately 2 acres on the corner of the intersection at High and Alden Streets. Nine buildings were reportedly located at the site during U. S. Radium's occupancy. Since then, some structures have been removed or incorporated into the seven buildings currently present at the site.

From the aerial gamma overflight, vicinity properties were identified by an 8.5 micro Roentgen per hour (uR/hr) isopleth surrounding the plant, and encompasses an area of nine city blocks with approximately 330 residential properties.

Activities associated with U. S. Radium's operations are also believed to have occurred at specific properties outside the nine-block area encompassing the Vicinity Properties and the manufacturing facility. These properties, referred to as Satellite Properties, have been identified as additional areas where radium-containing materials may have been handled or disposed. In some cases these Satellite Properties were remote locations where employees would paint watch dials with the luminous paint. One of the Satellite Properties was the home of Dr. Barker, the chief chemist at the U. S. Radium Corporation who took his work home with him.

PROJECT INVESTIGATION HISTORY

Investigations at the U. S. Radium Corporation site include the various radiological surveys conducted by the NJDEP from 1979 to 1985, the aerial gamma overflight conducted by EPA in 1981, field radiological screenings performed by CDM in 1985 and 1986, and the two Remedial Investigation/Feasibility Studies (RI/FS) completed by Malcolm Pirnie, EPA's contractor, in 1993 and 1995.

NJDEP Bureau of Radiation Protection

NJDEP maintained direct interest in the U. S. Radium project, even after the site was listed on the NPL, because the facility was still fully occupied. It wasn't until 1985, when the businesses in two of the existing seven buildings vacated the property, that the state discontinued direct involvement in the field investigations.

In addition to the surveys performed in 1979, BRP also investigated nine Satellite Properties and 31 Vicinity Properties in 1983-4. Elevated exterior gamma radiation exposure rates were found on six properties, and elevated indoor rates were identified at two properties. All nine Satellite Properties had gamma radiation levels above background, approximately 10 uR/hr.

CDM Surveys

Under a REM (Performance of Remedial Response Activities at Uncontrolled Hazardous Waste Sites) contract with EPA, CDM performed additional radiological surveys at the U. S. Radium site in 1985 and 1986. Results from gamma exposure rate surveys and radon surveys at the High and Alden Street properties were consistent with results obtained during previous investigations performed by BRP. Surface alpha scans were also performed at three buildings at the High and Alden Street properties during this investigation. Elevated alpha readings were detected in two of the three buildings scanned, indicating possible surface or building material contamination rather than contamination at depth.

CDM also investigated six Satellite Properties during its 1986 investigation. Radiological contamination was identified at three of these properties. Further investigation was recommended for the three properties to identify the source, nature and extent of contamination.

No surveys were performed on the Vicinity Properties at this time, as BRP was taking the lead for those properties.

Malcolm Pirnie RI/FS

In September 1989, Malcolm Pirnie initiated a RI/FS for two Operable Units. Operable Unit I (OUI) consists of approximately 400 Satellite and Vicinity Properties. Operable Unit II (OU II) consists of the U. S. Radium Corporation facility and four commercial properties. The four commercial properties were included in OU II based on historical accounts of activities related to the radium-processing and dial painting work carried out by the U. S. Radium Corporation in these properties.

The RI/FS for OU I was completed in January 1993 and the Record of Decision (ROD) was issued in September 1993, which identified excavation of material with radium activity above the cleanup criteria with offsite disposal as the appropriate remedial alternative. The selected remedy was similar to the one issued for the MWG sites as the remedial alternatives were taken directly out of the original MWG ROD and then pared down based on applicability to the U. S. Radium site. The cleanup criteria identified in the ROD were 5 picoCuries per gram (pCi/g) above background for both surface soil and subsurface soil. The ROD also required reduction of radon gas levels in excess of 4 pCi/l and radon decay products in excess of 0.02 WL.

During the RI stage of the RI/FS in 1991, EPA identified five properties that required interim measures to address health risks related to unusually high radon levels and indoor gamma radiation exposure. EPA had radon mitigation systems installed in two of the residences and one commercial building, along with the installation of lead shielding along the wall of the basement and removal of contaminated materials under the basement slab, to reduce the gamma exposure in two houses.

The USACE entered into an interagency agreement with EPA in 1992 to perform the remedial design for the U. S. Radium site. USACE contracted with CDM in April 1994 to perform the predesign investigation and prepare the design documents for the first 59 properties. Concurrently, work on the RI/FS for OU II continued and was completed in May 1995 by Malcolm Pirnie. The ROD was issued in August 1995 identifying excavation and offsite disposal as the selected remedy. This alternative provided for the excavation of 18,400 cubic yards of radium-contaminated soil, along with 150 cubic yards of construction material resulting from the demolition of buildings, or parts of buildings contaminated with radionuclides. During this remedial action, relocation of active businesses at some of the commercial properties would be required.

PREDESIGN INVESTIGATION STRATEGY

Along with U. S. Radium, the Superfund sites at Montclair/ West Orange and Glen Ridge in Essex County, New Jersey, had also been included on the NPL in 1983. The two sites include three noncontiguous study areas in residential communities three miles from the U. S. Radium site. Although defined as separate Superfund sites, the source of their radiological contamination was linked to the process tailings from the U.S. Radium Corporation facility during the program initiated by NJDEP to investigate former radium processing facilities within the state. The rural areas of the Montclair/West Orange and Glen Ridge communities apparently were used as disposal sites for the process tailings from U. S. Radium Corporation. Some radium-contaminated soil was also moved from the original disposal locations and used as fill material in low-lying areas. Houses were subsequently constructed on or near the radium waste disposal areas. In a few instances, part of the fill was mixed with concrete for construction of foundations or sidewalks.

Field investigation work for the RI/FS began at the MWG sites in 1983. CDM completed the supplemental RI/FS in April 1989 and the ROD for the MWG sites was issued by EPA in September 1989. In 1989, under an Interagency Agreement (IAG) with EPA, USACE implemented a pilot program on four residential properties, which included the demolition of all structures, excavation of contaminated soil and disposal off site. Nine properties were added to the pilot program, but rather than demolish the structures, the foundations were supported during soil excavation and the structures were fully restored. The cost of this initial pilot program was $10 million. In 1990, the full-scale remedial construction program began and continues to date, through six phases of construction and remediation at 264 properties. Predesign/design work for the first 245 properties was completed at a cost of $16.3 million. The budget for the entire MWG project, including construction costs and which is scheduled to be completed by the year 2000, is $250 million.

In developing the work plan for performance of the predesign and design work, EPA and USACE suggested that CDM build on the experiences of the ongoing remediation at the MWG Superfund sites. EPA, USACE, and CDM identified several ways to reduce the design costs and schedule:

The joint experiences of EPA, USACE, and CDM from the MWG sites were the basis for a separate design IAG between EPA and USACE. USACE - Kansas City District contracted with CDM to provide predesign and design services for the first 59 U. S. Radium properties identified for remediation. Phase 1, including only 18 of the 59 properties, was planned to be a small phase so that EPA could start construction work and still have the opportunity to work out kinks in the design by receiving feedback during the remedial action. However, the completion of the Phase 1 remedial design documents coincided with the federal government shutdown in 1995-1996, and no federal funding was available for remedial action. Therefore, by the time Remedial Action (RA) funding became available, the Phase 2 design was also completed, so the two phases were combined. The first step was to minimize the scope of the predesign field investigation to save time and money. This could be accomplished by utilizing the data collected during the RI/FS, using gamma meters rather than soil sampling to delineate the soil contamination, and development of a sampling plan to meet specific design objectives.

Use of RI/FS Data

EPA provided CDM with extensive gamma survey data collected during the RI/FS performed by Malcolm Pirnie. However, the gamma data had been collected using instruments with different probes than those to be used by CDM during the predesign field investigation. The varying efficiency and sensitivity of the probes do not allow for direct comparison of the measurements. CDM developed a correlation between the counts per minute (cpm) recorded on the gamma scintillation detectors with a 1"x 1" Sodium Iodide (NaI) crystal used by Malcolm Pirnie and the gamma scintillation detectors with a 2" x 2" NaI crystal used by CDM based on the cpm recorded at 20 boreholes for each instrument. The data included in the RI/FS was then translated, based on this correlation, then plotted in cross section to delineate the vertical extent of the contamination and to identify data gaps for additional boreholes. An average of five additional boreholes were installed for each residential property. The data from the additional boreholes was then plotted along with the original correlated borehole data and checked for consistency.

A similar correlation and data verification process was performed for surface soil readings.

Soil Activity to Gamma Emitter Correlation

In addition to gamma survey data, soil and building material sampling and analysis was planned to delineate soil and building contamination. To reduce analytical costs and field investigation time (turnaround for radiological analytical results is normally 45 days), CDM collected 22 downhole soil samples and had them analyzed for Radium-226. At each of the sample locations, a gamma measurement was recorded. The Radium-226 results were then correlated against the gamma measurements to determine the gamma value in cpm that would represent a 5 pCi/g radium activity in soil.

A separate correlation was generated for surface measurements to account for differences in geometry. CDM verified the surface gamma readings to soil Radium-226 activity obtained with 2" x 2" NaI detectors at the Department of Energy calibration pads in Grand Junction Colorado by analyzing five surface soil samples and comparing these activity with the field gamma measurements.

The gamma cutoff values developed from these correlations were then used in the field, enabling the field crew to use real-time gamma measurements for identification of soil contamination rather than wait for laboratory results. As a quality control measure, two soil samples were collected on each property at "borderline" measurement locations, to substantiate the conclusions reached using the gamma meters and to check against the original correlation.

Phase 1 and 2 predesign work was completed in spring 1996 at a cost of $ 1.5 million and 9,191 labor hours. This represented a reduction in average level of effort from 240 labor hours per property at MWG sites to 160 labor hours per property at U. S. Radium, and a reduction of average analytical costs from 25 samples per property to two.

USACE issued the remedial action contract for these two phases in December 1996. Concurrently with Phase 1 and 2 remedial action, predesign field investigations continued on subsequent Phases 3 and 4. Unlike the Phase 1 and 2 properties, there was minimal data available from earlier investigations for properties that were included in Phases 3 and 4. As a result, it became necessary to increase the predesign effort in terms of numbers of soil borings and soil samples. CDM was able to reduce costs in other ways by implementing additional data collection techniques based on experience gained at both the MWG and U. S. Radium sites. This included the use of pen-based computers for data collection and an onsite gamma spectroscopy for use as a quick-count system for field screening of soil samples.

Pen-Based Computers

CDM had piloted the use of pen-based computers for recording field gamma measurements in lieu of field logbooks during field work at MWG. The gamma measurements were electronically recorded in a database, by entering the data on the computer screen in the field. The data is then downloaded for use in a spreadsheet format, ASCII file, or for translation to a .dxf. format for use on Integraph drawings included on the design documents. This process eliminates a data entry step from logbook to computer, and also the quality control check of this data entry step. This allows more time to check for data anomalies and to confirm that the data fully supports the design.

The data collection requirements for this type of radiological measurement vary from the typical geological logging of a borehole, so special programming was required to adapt the computers for this use. In addition, after several weeks of using the computers, the field technicians requested some additional programming changes to simplify the data entry process. Once these modifications had been made, it was observed that use of these computers did not change the time, faster or slower, necessary to complete the field work at MWG. However, use of these computers did shorten the time needed to complete the work between data collection and remedial design, thus offsetting the cost of the technology. From this experience, it was determined that the pen-based computers would be implemented for use at U. S. Radium. This has been accomplished successfully, with the predicted resulting reduction in data processing time and improved data management capabilities.

Onsite Quick-Count System

When the field work at U. S. Radium was being planned, it was deemed unnecessary to field screen soil samples as very few samples were being collected. This was a departure from the procedure at the MWG site, where soil samples were being collected from every 6-inch interval of a soil boring as part of predesign field investigations. As full gamma spectrography laboratory analysis of every sample was determined to be infeasible and unnecessary, an onsite gamma spectroscopy system was used to screen soil samples that were sent to the lab for laboratory analysis by quick-counting the soil samples for Radium-226 or any other gamma emitters.

This screening method did prove to be useful at the U.S. Radium site on several occasions when measurements taken with the gamma scintillation meters were not sufficient to verify the results of one soil excavation effort. The analyzer is a gamma spectrograph that utilizes a 3" x 3" NaI crystal. Quick-count readings can have a large degree of error (1-2 pCi/g) as they do not allow time for a full ingrowth period and samples are not dried or homogenized, but they do provide a greater measure of confidence that the laboratory results will be below cleanup criteria. For example, a quick-count result of 2 pCi/g or lower, adjusted for possible error of 2 pCi/g, would still be less than 5 pCi/g and, therefore, be sufficient to make decisions in the field regarding excavation limits.

At the U. S. Radium facility, the Remedial Action Contractor (RAC) was using a portion of the open lot to set up the temporary facilities compound during remedial action. To address worker and local health and safety concerns, an area of contaminated soil was removed. Elevated gamma meter readings were observed on the walls of the excavation, as expected, but the floor of the excavation was believed to be clean. Shine from the excavation walls was interfering with the gamma meter readings on the floor. CDM used the onsite quick-count system to screen the floor soil samples. This enabled the RAC to backfill immediately and continue with mobilization efforts.

Phase 3 predesign work on 50 Vicinity properties was completed in fall 1996 at a cost of $1.6 million and 10,516 labor hours. The remediation contract for this phase was issued by USACE in December 1997. Phase 4 preliminary investigation and predesign work on 25 Vicinity properties and all the OU II properties was completed in January 1998 at a cost of $1million and 6,050 labor hours. The difference between MWG and U.S. Radium was less significant in these two phases, with the average level of effort increasing to 210 labor hours per property as compared to 240 for MWG.

INTERIM MEASURES

While completing the predesign investigations, certain properties were identified as requiring immediate attention due to above criteria indoor radon levels or gamma radiation levels. EPA determined that interim measures should be taken at these locations to mitigate health risk concerns. These occurrences lined up very well with the experience at the MWG sites which allowed EPA to implement interim remedies very quickly. Radon mitigation systems were installed in each of these houses, resulting in a reduction of the radon levels in each home to less than 2 pCi/l.

There were occasions, however, when the field investigation encountered readings that were outside the bounds of typical contaminant patterns. One home with elevated radon also had unusually high gamma radiation levels in one bedroom on the second floor. After closely investigating the apparent source of the gamma radiation, it was determined that there was a narrow strip of concrete material below the wood floor in the bedroom that had several "hot" areas, probably resulting from tailings used as an aggregate source for the concrete. The concrete was not acting as a structural member, so it was immediately removed from the bedroom, lowering the gamma radiation from 1.9 million cpm to 30,000 cpm. This property is now being fully remediated under Phase 1 and 2 activities.

REMEDIAL DESIGN

The remedial design for the first phase was placed on a fast-track schedule to meet EPA's fiscal year end 1995 deadline to obtain remedial action funding. This only allowed 8 weeks to complete the 95 percent design submission for the Phase 1 properties. This was accomplished by electronically transferring the gamma data from spreadsheet files to .dxf. files using CDM's Environmental Data Manager (EDM). This enabled the design team to overlay the gamma measurements onto the civil survey map for each property. Excavation limits could then be established based on the overlay. Structural and landscape restoration plans were then developed from the excavation plans.

Primary excavation limits were established based on gamma meter readings. To address the potential for overlooking contamination due to the limited number of soil borings, the primary areas were bounded by five-foot wide secondary areas which would be excavated in increments, as post-excavation testing determined.

Phase 1 and 2 design work was completed at a cost of $1 million and 10,940 labor hours. Phase 3 design work was completed at a cost of $850,000 and 9,572 labor hours. The remediation contract for this phase was issued by USACE in December 1997. Phase 4 design work was completed in January 1998 at a cost of $670,000 and 7,136 labor hours. The average level of effort of 200 labor hours per property for these four phases represents a reduction from 300 per property at MWG. Some of this reduction may be attributed to the use of electronic data transfer from the field to design drawings, but it should also be recognized that the depth of contamination at the MWG properties increased the design difficulties, especially related to maintenance of existing structures.

In addition to the planned design packages, several properties were identified as requiring remedial design work based on information collected during the remedial action on the first two phases. In some cases, contaminated soil extended beyond property lines and in other cases, properties adjacent to those being remediated were added as property owners allowed access. In both situations, it was necessary to generate property design packages in very short periods of time to avoid repetition of excavation in the same area and subsequent impacts to the implementation of the design schedule. The design package was separated into exterior yard work and interior structural work. The exterior work was completed first, as it required less design time and had to be handled with the adjacent open property excavation. The interior excavation and building restoration work could be completed later to accommodate the RAC's schedule.

REMEDIAL ACTION

Phase 1 and 2 construction began in the fall of 1996, and is scheduled to be completed in early 1999. Phase 3 construction is scheduled to begin in the spring of 1998 and will continue until the following summer. USACE issued both remedial action contracts to the same preplaced contractor who had performed the Phase 1-5 work at MWG, Sevenson Environmental Services, Inc. This decision was made primarily because both EPA and USACE recognized the value of the prior experience at MWG to the U. S. Radium remedial action, both in terms of specific experience with the work and with the community relations requirements of working on private residences.

During the Remedial Action, CDM supports EPA and USACE by providing post remediation verification services which include gamma surveys, soil sampling and analysis, gamma exposure rate measurements and radon measurements to verify that the cleanup criteria have been achieved. After the excavation limits are reached and prior to backfill, gamma surveys are the first step in verifying that cleanup has been achieved. Then gamma exposure rate measurements are taken using a Pressurized Ionization Chamber (PIC). Finally soil samples are taken and analyzed by the onsite quick-count system. The results from the quick count are used again to verify that the cleanup has been achieved and that the original radium soil activity to gamma meter correlation is valid. After the quick count results are reviewed, the excavation is backfilled. Upon completion of the backfill operation, gamma exposure rate measurements are taken again.

As discussed in previously, in some situations additional soil contamination is discovered beyond the limits of the original excavation plan. Removal of this material is simple to address using the post excavation confirmation criteria. However, when the soil contamination crosses a property line onto an adjacent property that has not been previously investigated, it is necessary to prepare designs for adjacent property remediation very quickly. If access can be obtained from the adjacent property owner and the soil contamination is limited to exterior soil away from any structures, it is possible to complete the additional excavation without formal design submittal procedures. This requires close coordination among CDM, EPA, USACE and the RAC.

Once all work on the property is completed, including all restoration work, an alpha-track detector is placed in the house for a one-year duration. This is the final verification measurement to confirm that the radon level is less than 4 pCi/l.

As of January 28, 1998, excavation at 30 of the Phase 1 and 2 properties has been completed. All of the delineated primary excavation areas (6,948 cubic yards) were removed. Of the 6,355 cubic yards of potential secondary excavation area, approximately 51% ( 3,258 cubic yards) was removed. This compares to Phases 1 to 6 at the MWG sites, where the actual soil volume excavated (126,646 cubic yards) exceeded the total primary and secondary design volume of 117,490 cubic yards by only 8%. This indicates that the assumptions defining secondary excavation for U.S. Radium may be too conservative.

The aspect of this remedial action that receives the greatest community response is the restoration of remediated properties. Technically and economically, restoration is a minimal consideration when developing design documents. However, from a community relations perspective, property restoration is a key aspect in the entire remedial action program. During the construction work, EPA, USACE and the RAC must deal with the concerns of the individual homeowner and tenant on a daily basis. In addition, the overall impact on the community from the appearance of newly restored structural and landscaping features, is significant. From the experience at MWG, EPA and USACE have been able to anticipate community concerns and address them before they become larger issues.

SUMMARY

The U. S. Radium project is currently in its fourth year of design work and its second year of construction. Since 1994, EPA has been able to complete predesign field investigation and prepare design packages for 134 OU I properties, and all of the OU II properties, including the U.S. Radium facility. Remedial action has begun for Phase 1 and 2, and is anticipated to be completed in early 1999. Phase 3 has begun and will be completed in the summer of 1999. Phase 4 remedial action will begin in late 1998.

The experience from the MWG project has resulted in the following benefits to the U. S. Radium project:

In addition, community relations with the residents of the City of Orange are much better than could be expected considering the extremely disruptive nature of this remediation. The residents are seeing the real value of this entire program by having their own houses remediated and restored. Because of the Phase 1 and 2 remediation efforts, residents who had previously denied access to their properties are contacting EPA directly and requesting to be included in the program.

REFERENCE

1. C. Clark, "Radium Girls; Women & Industrial Health Reform 1910-1935", University of North Carolina Press, 1997.

The opinions and conclusions expressed in this paper do not necessarily reflect those of the U.S. Environmental Protection Agency, and no endorsement should be inferred.

BACK