Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
4 Measurement of Exposure to Asbestiform Fibers N For more than 50 years, asbestos-containing dust in the workplace has been measured to characterize occupational exposure to these particles. These measurements were needed to correlate specific health effects in workers with their exposure to the dust, to ensure the proper functioning of dust control equipment, and to evaluate compliance with the fiber and/or dust standards or guidelines in effect at that time. In developing these measurement methods, attempts were made to balance and maximize specificity, sensitivity, and biological relevance for the different dust components. As measurement technology and knowledge of agents and disease mechanisms advanced, new sampling and analytical methods were developed with the goal of obtaining measurements that would be useful in protecting workers. Holt, 1957; Walton, 1982. ~ (See reviews by Ayer and Lynch, 1961; Attempts are now being made to determine the concentration of fibers in other environments, such as in buildings and in areas removed from known fiber sources. Techniques useful for the workplace are not always easily applied to other situations, where concentrations of materials are likely to be hundreds or thousands of times lower. In this chapter, the committee describes the measurement methods used to determine the concentration of asbestos in a given environment. Although the discussions are focussed on the specific methods used to measure asbestos, many of these methods may also be used to measure other asbestiform fibers. The development of the techniques is presented within a historical perspective. MEASUREMENT TECHNIQUES Table 4-l summarizes the principal methods used and identification of asbestiform fibers (Burdett the quantification _ , 1980~. The earliest methods measured mass. In the gross mass methods, airborne dust wan collected by filtration, precipitation, or impaction, and the total dust was determined by simple weighing on conventional balances. X-ray diffraction techniques were used to identify mineral phases present in the dust; magnesium analysis was used as an index of chrysotile asbestos 82
83 TABLE 4-1. Asbestiform Fiber Measurement Methodsa Measurement Collection Quant if icat ion Ident if icat ion ~. Mass, grose Filtcr Gravimetric Electrostat ic Gravime t ric prec ipitator ~ piezoe lec tric Mass, re- spirable Count Impingement Mineral ident if icae ion by x-ray; chrysot i le ident if icat ion by magnesium analysis Not appl icable Impac t ion Hi-vol/filter Beta-absorpt ion Mic roscop ic Horizontal e lu- Gravime t ric triator/f ilter Cycloneffilter Gravimetr~ Light micro~cope Not app 1 icab le Mineral ident i f icat ion by x-ray Mineral ident i f icat ion by x-ray; chrysot i le ident if icae ion by magne ~ ium ana ly ~ i s Mineral identification by x-ray; chry~ot i le itent if icat ion by magnesium analys is Identification by morphology Impac t ion Light mic rose ope Ident i f icat i on by morpho logy The rma 1 Light mic rose ope Ident i f icat ion prec ipitator by morphology Hembrane filter Light microscope Identification by phase contrast morphology; mineral 't ident i f ic at ion by · dispersion staining ~. Nuclepore filter TEM,b SEM,C .Hineral identification image recog- by SAED;d chemical nit ion composition by - EDXAe Nuc lepore f i 1 te r Light sca t t e ring Ident i f icat i on o f fibers by magnetic alignasnt aAdapted from Burdett et al., 1980. bTEM - Transmiss ion e lectron mic roscope . CSEM - Scanning electron microscope. tSAED - Selected area electron diffraction. eEI:~XA - Energy-dispersive x-ray analysis.
84 content. When there was a need to collect and measure samples over short times, such as in the evaluation of controls or brief exposure episodes, the mass of the small amount of material could be measured by very sensitive piezoelectric or beta-absorption instruments. Major drawbacks to these analytical methods were the insensitivity of the x-ray me thod in the detection of small part ic le a , the nonspecificity in the resolution of chrysotile from the other serpentine minerals, and the similar nonspecificity of the magnesium assay. Because much of the mass measured by gross methods consisted of particles too large to penetrate into the lung, techniques were often used to remove the large r particles before assay. The horizontal, parallel plate elutriator was preferred in the United Kingdom, whereas industrial hygienists in the United States tended to use small cyclone devices. All the mass methods yield results stated in terms of mass of dust per unit volume of air. In occupational environments, the units commonly used are milligrams of dust per cubic meter of air, whereas the much lower dust masses found in nonoccupational ambient environments are more conveniently expressed as nanograms of dust per cubic meter of air. Counting methods are far more sensitive than mass determinations, since samples with too little mass to be weighed are usually adequate for counting.] Furthermore, since small particles far outnumber large particles, counting emphasizes the respirable dust. Lantly, fibers can be counted separately from other particles.2 Particles deposited directly on microscope elides by impaction or thermal precipitation can be counted by light microscopy. However, a more even dispersion can be obtained by impinging a jet of dust-ladened air on a surface submerged in a liquid. The liquid is then transferred from the impinger to a counting cell where the particles are allowed to settle so they can be seen and counted in the same focal plane. These methods have low and differing efficiency ant resolving power. The membrane filter, however, is a very efficient dust collector. After being rendered transparent, thereby making the f ibers visible, the fi lter can be examined by phase contrast microscopy. The best resolution is For example, 1 ng of chrysotile dust would yield 400 fibers 5 Am in length and 0.5 Am in diameter. A nanogram is about a thousand times lighter than most analytical balances can weigh with precision and accuracy. 2As noted in Chapter 2, shape alone does not determine whether a particle is asbestifonm. In a workplace where asbestos fibers were the major dust present, the distinction was presumably not of major practical importances For occupational environments, asbestos fibers are counted if they are more than 5 Am long and at least three times longer than they are wide (National Institute for Occupational Safety and Health, 1977~.
85 obtained by the transmission electron microscope, which can resolve particles made up of only a few hundred atoms. Somewhat larger particles may be identified by techniques that reveal their chemistry (a probe technique) or crystallographic characteristics (by electron diffraction). Results of impinger counts are usually expressed in millions of particles per cubic foot; dust concentrations measured by other methods are typically expressed as particles or fibers per cubic centimeter. In some electron microscope techniques, fibers or dispersed fibrils are counted, and the results are then converted to units of mass per volume. MEASURING ASBESTOS DUST IN THE WORKPLACE The Impinger Technique Early investigators of workplace exposures to asbestos fibers in the United States used the impinger technique, then commonly used in mines. Dust was collected in an alcohol medium, usually over a short period (e.g., 20 to 30 minutes), and the suspension was examined by light microscopy at lOOX total magnification. All particles in the dust were counted. Very few fibers were seen, partly because of the low resolving power of that optical system. The counting of large numbers of samples was tedious, and interob~erver measurement differences led to systematic bias. The first asbestos dust "standard" in the United States was based on measurements made with impingers by Dreessen et al. ( 1938~ . mese investigators correlated observed health effects with measured dust exposures in the asbestos text ile industry and tentatively concluded, with reservations, that limiting exposure to 5 million particles per cubic foot (5 mppcf) of air may be effective in preventing asbestosis. No corre let ion with cancer of any type was attempted. They recognized, as did later investigators, that counts of all particles provided a very indirect index of disease potential. The Membrane Filter Technique In the membrane filter technique, efficient, convenient collection media are used for assaying the work environment (Edwards and Lynch, 1968; Holmes, 1965; Leidel et al., 1979~. A portion of the filter may be rendered transparent and then examined with a phase contrast light microscope. Fibers with an aspect ratio greater than 3 to 1 are counted on a prescribed, representative area of the filters (National Inst itute for Occupational Safety and Health, 1977~. This technique is sufficiently sensitive to allow fibers in workplaces to be counted with measurable precision and accuracy. dSee Chapter 2 for a discussion of mineralogical definitions.
86 Prior to the 1960s, fibers of several lengths were counted separately and reported (Lynch, 1965) . During the 1960s, the U. S. Public Health Service followed the counting strategy developed in the British textile industry, and counted only fibers >5 Am in length--a length that was later incorporated into the U. S. occupat tonal asbestos standard (U. S. Occupational Safety and Health Administration, 1971~. The longer asbestos fibers were believed to be the agents responsible for asbestosis (Beattie and Knox, 1961~. In addition, when only the "longer" fibers were counted, greater precision was attained from repetitive fiber counts on the same spec imen (Addingly, 1966) . MEASURING ASBESTOS DUST IN THE AMBIENT ENVIRONMENT The number of fibers >5 Am in length counted on membrane filters by phase contrast light microscopy is used as an index for exposure in the industrial workplace. However, these fibers may constitute only a small portion of the total number of fibers present. When the fibers collected on membrane filters, which collect particles as small as 0.01 Am in diameter, are counted by transmission electron microscopy, up to 100 times more fibers may be detected than are visible by light microscopy. (See Lynch et al., 1970, for accounts of studies in the textile industry, and Rohl et al., 1976, for measurements in the brake repair industry.) The ratio of transmission electron microscope fibers to fibers visible in the light microscope may be a function of fiber type, industry, degree of manipulation, distance from emission source, and other factors. Fibers in the ambient environment far from point sources of asbestos emissions are gene rally much shorter than 5 ~m, thinne r than O. 5 ~m, and, thus, predominantly smaller'ehan the resolution capacity of the light microscope (Spurny and Strober, 1981~. In the ambient environment, electron beam instruments can be used to measure fiber concentration and to characterize single, isolated fibers (Lange r and Pooley, 1973~. Other mineral particulates may pose serious background problems. For example, in areas where rocks ant minerals are crushed for processing, particles resembling asbestos may be emitted into the ambient environment (Larger et al., 1979~. Because chrysotile accounts for more than 90: of the asbestos used in the United States, the Environmental Protection Agency (EPA) has used it as an index of asbestos exposure. However, it was considered impractical to recover the fibers routinely without introducing artifacts or altering fiber size. Therefore, only the chrysotile mass, as determined with the electron microscope, has been monitored routinely (Thompson, 1978). In the standard technique, large volumes of air are pulled through membrane filters. Portions of the filters are ashed in nascent oxygen at low temperatures to remove interfering organic matter. This ash is then dispersed in water by ultrasound, and the residue in filtered onto another membrane filter. Then, either this filter is "rubbed out" in a
87 nitrocellulose medium and portions of the nitrocellulose film examined by electron beam instrumentation or it is directly transferred to an electron microscope grid for analysis. Both methods reduce the fibers to unit fibrils, thus enhancing homogenization of the specimen and reducing scan time, but information about the original nature of the fibers is lost. RELATIONSHIPS AMONG VARIOUS EXPOSURE MEASUREMENT METHODS . Data on numbers of fibers in the workplace have been used in corre rating exposure and heal th e f fee t s in various occupat iona 1 studie s (British Occupational Hygiene Society Committee on Asbestos, 1983; Dement et al., 1982; Liddell et al., 1982~. In order to be able to compare diverse studies and to assess health risks from ambient exposures, it would be useful to establish a relationship among the various methods of determining exposure. Specifically, what are the relationships among the several methods used in estimating asbestos dust in the workplace and in the ambient environment? Consistent relationships among these methods do not exist. They are subject to analytical error and subjective bias (Thompson, 1978~. As examples, the electron microscope technique and its associated sampling and analytical techniques have an experimental error of approximately 15: to 30: of the measurement value. Relative standard deviations of 45X are not unusual in light microscope counts. In addition, measurements made in a particular environment at different times will vary because the actual concentrations vary. The different techniques measure a variety of indices, which often do not remain in constant proportion to each other from sample to sample . For example, with the phase contrast light microscope, fibers longer than 5 Am are counted as a single species, whereas shorter fibers are not counted at al 1. Therefore, a given fiber count obtained by this technique would undoubtedly represent very different numbers of fibers and mass concentrations than the same fiber count obtained by electron microscopy. In some cases, reproducible conversion factors may be de termined when large numbers of paired samples are analyzed by the various methods. However, these conversion factors usually cannot then be applied to samples obtained under a different set of conditions. Table 4-2 summarizes some reported attempts to determine conversion factors among the various methods. The ratios in that table are based on direct, independent estimates, except for those in parentheses, which were calculated from other ratios. Although the accuracy of these estimates is not known, an order of magnitude either way would probably embrace most situations. These ratios are equivalent only in the sense that they would be expected if side-by-side measurements were made in an environment similar to that in which the data were originally obtained. 1
88 TABLE 4-2. Relationships Among Methods of Measuring Exposure to Asbestos in the Workplace Equivalent Values Expected from Various Measurement Me thods Phase Contras t Elect ran Light Microscope Microscope Impinger (PRIM) ~ >5-~- (E>l) (EM Mass Base Valueb (mppef) long fibers/cm3) fibers/cm3) (mg/m3) 1 n~ppcf 1 6 (360)d (0.2) ~ impinger) c 1 >5-m long fibers/ O. 17 1 60 O. 03 cm3 (pCLMye 1 Et! fiber/cm3 (O. 0028) 0.017 1 O. 0005 ~ EM count ~ 1 mg/m3 (5 ~30f 2, OOOg 1 (mass) aRatios developed by Cook and Ilarklund, 1982; Davis et al., 1978; Dement et al., 1982; Lynch et al., 1970; Rohl et al., 1976, and the British Occupational Hygiene Soc iety (Walton, 1982) were used to construct the table. Some adjustment was necessary to achieve consistency. bGiven the base value indicated in column 1, the other columns show the equivalent value to be expected from the indicated method. Thus, 1 mpp`:f by impinger would be equivalent to 6 >5-m-long fibers/cm3 measured by PCLM or 360 fibere/cm3 measured by the EM. Numbers have been rounded. CCollected in an impinger and counted at lOOX light field. mppcf = millions of particles per cubic foot. dRatios in parentheses are calculated from other ratios. eCollected on membrane filters and counted by ACID at 430X. Ethic rat lo converts to 30 ID fibers/ng versus the nominal 20 fibers/ng sometimes used. "This ratio converts to 2,000 total EM fibers/ng. The data for this table were obtained from workplace dust clouds or other environmental samples containing high concentrations of asbestos. . Fiber size/weight relationships are also presented (Table 4-3) to indicate the ratios that might be expected under various conditions. The electron microscope (EM) count/mass ratio (2,000 fibers/cm3 per mg/m3) in Table 4-2 is equivalent to 2,000 EM fibers per nanogram,
89 C) · - e~ · - _ C~ :% ~ C~ C _ · - 0 dV · - ~ ~ O O C) °O (U U. N .," .,' U. o V ~ C~ · - a ~S v ~o c S" ~ 0 _ E · - o 2 ·^ U~ - 0 ~ :~ D ·r. · - o I~sD O~D O -O 0 _·e ·e O ·e_ _ 0 ,e_' U ~ . · - 000 0~0 m0 O`D^0_4 ~ C~·e·- ·e O ·-__ _ _ U~ O O~ 0=000 0~0 =0 ~ ° 0 0 ~ C%e C~ ·. ~·- _. ·e e- ·e O _ _ _ _ _ _ C~ J U~ O O O O O O O O O O 0 `0 0 0 0 ~ O ~U~ _ 0 _ ~ ·- 0 ·e ~ e- ·- _ · e. · · ~_ _ _ O ~e ~ ~ _1 _ 000 ° ~° ° ° OO O O O O e ~e~ ~ ~ _~D ClD O ~O ~O - -~ ·- ~ ~ ~ e- e. ·- ~ ·- ~ ee ~ee _ C~ C~ _ ~ ,e-1 ~ _ ~ _ ,e_. · O ~`0 ~ _ _ 00 0 0 0 0 0 0 0 0 0 0 ·· O ~0 ^1 0 - O ~O e~ e. ~ ·- 0 ·. ~J e- O ee O ,-1 e~ _ ~ e_' 0 ~e · e e ~1 ~e _ _. 00 00 00 0= 0 - 0' 0~ 0- O ·. O ~ ~ e- ~ ·. (D ·. ~ ~1 l~t e. _ e. _ e. _ ~e. · e ^1 C ~`0 _ _ · O O ~e _ ~_ j0 e~) ~-_ ~O OO OO ~O e~ ~eCO ~O-O ·. O Z ~ ~ ·-0 ·-~- CX} (U ~ ~1e~ _~ _ ~e. · tV ~ `00` ~~ ~e - =.~1 tV ._O ~- E" ~ O0 1 . '+ ~el J ~L~ 42_ ~ OO~ ~O C ~rl··· ·· · ~ OOO OO - C~l t- _ O P. he .'
JO which corresponds to a distribution of fibers with a mean length of 3 Am and diameter of 0.3 Am. Such a distribution suggests the presence of substantial numbers of fibers longer than 5 ~m, which would be visible under a light microscope. The ambient air in environments far from asbestos sources have few fibers longer than 5 ~m. In general, those remote ambient environments will contain many more, but smaller, fibers in a given mass than would the workplace clouds on which Table 4-2 was based (Spurny and Strober, 19813. For example, if the fibers in the remote environment had average lengths of 1 Am and diameters of 0.1 ~m, there would be 70,000 EM fibers/ng (instead of the 2,000 determined for the workplace) and the ratios would be altered accordingly. EXPOSURE TO CHRYSOTILE IN THE AMBIENT ENVIRONMENT . . . Chrysotile has been detected in urban air (Selikoff et al., 1972) and in lunge of urban dwellers (Larger et ale ~ 1971; Pooley, 1972; Pooley et al. ~ 1970). Ambient levels of chrysoeile asbestos are usually expressed as mass concentrations (ng/m3~. To estimate health risks from these ambient exposures, the mass measurements need to be converted to the equivalent fiber concentrations that are used as dose measurements in workplaces, for which dose-response curves have been developed. There is no single way to do this conversion since, as explained earlier, the dust clouds are quite different, especially in regard to the sizes of fibers they contain. One approach is to convert the ambient mass data into numbers of fibers of the shortest length (5 ~m) generally counted in the workplace, with an assumed diameter of 0.5 ~ and aspect ratio 10:1. There are 400 f ibers of this ~ ize in 1 ng . For a mass concentrat ion of 20 ng/m3, which is typical of outdoor environments not near known sources , this conversion yie Ids a concentration of 0.0080 fibers/cm3. Even in workplaces, however, most fibers are shorter than 5 ~m. Assuming that workplace fibers average 3 Am x 0.3 Am (the assumption made in Table 4-2) and applying the equivalency factors in Table 4-2, a typical equivalent concentration for 20 ng/m3 would be 0.040 fibers/cm3. However, if we were to assume an average remote ambient fiber size of 1 Am x 0.1 ~m, then a concentration of 1.4 fiber/cm3 would weigh 20 ng/m3. Measurements of ambient concentrat ions observed at single sampling locations may vary over several orders of magnitude. Seasonal changes in wind direction, especially near emission sources, account for much of this variability. For example, in studies reported by Thompson (1978), 20 specimens obtained downwind from an emission source had average asbestos fiber mass concentrations ranging from 0.03 to 8,200 ng/m3. However, for industrial cities in the continental United States from 1969 to 1970, average airborne asbestos mass concentrations ranged from 0.6 to 95.0 ng/m3, or two orders of magnitude. During 1971 and 1972,
l 91 44 samples similarly obtained contained concentrations ranging from 0.4 to 27.7 ng/m3. COMPLICATING FACTORS IN ENVIRONMENTAL ASSAYS In an asbestos workplace, all the fibers may be assumed to originate from the fiber being used. However, because remote ambient environments, by definition, are distant from known asbestos sources, the identity of fibers there can neither be assumed a priori nor easily determined with any certainty, especially by light microscopy (Langer, 1979~. The light microscope specified for analysis of membrane filter specimens (phase contrast microscopy) yields only size and shape information, which may allow the analyst to "identify" fibers by morphology alone. With the increased resolution of the electron microscope, the internal structure of the elementary chrysotile fibril may be visualized (Larger and Pooley, 1973~. For chrysotile asbestos, morphological information and the behavior of ache fiber under the electron beam are usually sufficient information for identification (see discussion in Langer et al., 1974~. However, other fibers require additional diagnostic procedures. Selected area electron diffraction (SAED) yields crystal data reflecting characteristic structural elements that may enable the microscopist to distinguish among types of fibers. Chemical information may also be obtained by means of either energy dispersive x-ray analysis or crystal spectrometry probe techniques. For fibers in remote ambient samples to be accepted as asbestiform, accurate fiber identification is needed. For example, Spurny and Strober (1981) have shown that more than 90: of "mineral fibers" in nonurban areas sampled in Europe were not asbestos, but, rather, were such materials as fibrous gypsum and even ammonium sulfate. In a study of the fibrous content of the lungs of the general population, Churg (1983) found approximate ly as many nonasbestos mineral fibers as asbestos fibers. Therefore, proper diagnostic tools are needed to characterize fibers in remote ambient samples as asbestiform. Furthermore, when extrapolating health risks from the workplace deco such remote environ- ments, it music be recognized not only that fiber concentrations and size distributions are different in the two environments but also that fiber types may include nonasbestiform varieties. The asbestiform properties enumerated in Chapter 2 cannot as Yet be measured on microscopic sampler. FUTURE MEASUREMENT OF EXPOSURE TO ASBESTIFORM FIBERS Walton ~1982) has noted that "there is no practicable alternative to the membrane filter/phase contrast optical microscope for routine use in the occupational environment . " Nonetheless, the method is too insensitive and nonspecific to yield the information needed to assess fiber exposure in the nonoccupational environment.
92 Several basic objectives should guide the development and eventual selection of a method for measuring fibers in nonoccupational or remote environments. First, the method should yield data that are useful in conducting epidemiological studies relating exposure and disease and in making decisions designed to reduce health risks. The ideal method should measure a characteristic, parameter, or index with biological relevance, i.e. , the measurement should be related to the risk of the disease end point being studied. Possible types of measurement include fiber number, mass, length, diameter, and surface charge. Because of the great extent of environmental variability, developing accurate informa- tion about the concentrations of fibers in the air will be more dependent on the number of samples collected than on limitations of analytical techniques. Current methods for determining ambient concentrations of f ibrous particles could benefit from substantial improvement. However, sufficient standardization is needed to allow comparisons of data from various laboratories 80 that a data bank of ambient concentrat ions can be established for use by epidemiologists and other researchers. Sensitivity and specificity improved as the light microscope was superseded by the electron microscope (EM) with its greater resolving power. One issue to be considered now concerns the relative merits of using the transmission electron microscope (TEM) and the scanning electron microscope (SEM). Other issues concern methods of preparing the fibers for the EM without disturbing them and development of improved identification techniques (Hiddleton and Jackson, 1982) . The SEM has been used extensive ly to examine environmental f ibers and has produced some dramatic photographs of fibers In situ. Although the SEM direct preparation method provides little opportunity for contamination, the image resolution, contrast, and x-ray resolut ion of the SEM have not been sufficient for precise mineralogical identification. With an energy-di~persive x-ray attachment, the SEM can now provide analytical information for identifying minerals, but it still does not provide structural data. Because of its high resolving power, the TEM has been more generally applied to studies of environmental fibers, especially when confidence in fiber identification is required (Chatfield, 1979, 1982; Chatfield and Dillon, 1978~. Researchers do not agree on the best method of preparing representative and quantitative EM samples. The fibers in air or water must be deposited evenly and unaltered on the flat surface of an EM grid and spaced far enough apart to be readily counted and examined, yet not so far apart that there are too few to count. In most modern methods, the sample is collected either on mixed cellulose eater Hillipore filters or on polycarbonate Nuclepore filters. To transfer the deposit directly onto an EM grid, the fitter must be dissolved, usually by washing gently with solvent. With uncoated Hillipore filters, as much as 80X of the fibers is lost.
93 The most satisfactory direct transfer preparation technique involves the carbon coating of particles on the surface of a l!tuclepore filter. This technique is part of the current EPA inte rim procedure ~ Samud ra et al., 1977) . Nuclepore filters are preferred because, unlike the Millipore filters, they have a smooth, featureless surface. Because of this property, vacu~m-coating with carbon produces a replicate that surrounds and traps the particles, holding them in their original position as the filter dissolves. The large amount of surface detail on Millipore filters makes them unsuitable for carbon coating. Rigorous fiber identification is not always necessary, especially in occupational or other defined environments. Morphology alone is often adequate, especially for chrysotile. For environmental samples, which may contain many fiber-shaped particles of different minerals, selected area electron diffraction (SAEI)) and energy dispersive x-ray analysis (EDXA3 may be used to obtain crystallographic and chemical information for more precise identifications. RECOMMENDATIONS Concentrations of asbestiform fibers in urban and rural loca~cions, and at various distances from known sources, should be routinely monitored so that fiber levels and population exposures can be determined with respect to time and location. Fiber characterization is also needed. If feasible, these data should be used in conjunction with health studies to determine any effects on the exposed populations. Characterization of f ihrous dusts should inc lude to the extent possible the length, diameter, quality, and type of all fibers present and their concentrations, both as mass and number. Direct transfer techniques and TEM examinat ion of the preparer ions, or other techniques that allow examination of particles as they existed in the aerosol, should be used. Fiber monitoring techniques for use in nonoccupational environments should be standardized so that results from various studies are comparable. Automated instrument techniques are needed to permit analysis of the large number of samples required to monitor exposure of different segments of the U. S. population over time. REFERENCES Addingly, C. G. 1966. Asbestos dust and its measurement. Ann. Occup. Hyg. 9: 73-82. Ayer, H. E., and J. R. Lynch. 1961. Motes and fibers in the air of asbestos processing plants and hygienic criteria for airborne asbestos. Pp. 511-522 in C. N. Davies, ed. Inhaled Particles and ilapours, Pergamon Press, Oxford.
94 Beattie, J., and J. F. ~ox. 1961. Studies of mineral content and particle size distribution in the lungs of asbestos textile workers. Pp. 419~433 in C. N. Dairies, ed. Inhaled Particles and Vapours, Pergamon Press, Oxford. British Occupational Hygiene Society Committee on Asbestos. 1983. A study of the health experience in two U.K. asbestos factories. Ann. Occup. Hyg. 27 :~-55. Burdett , G., J . M. LeGuen, A. P. Rood , and S . J . Rooker. 1980. Com- prehensive methods for rapid quantitative analysis of airborne particulates by optical microscopy, SEM and TEM with special reference to asbestos. Stud. En~riron. Sci. 8:323-328. Chatfield, E. J. 1979. Preparation and analysis of particulate samples by electron microscopy with special reference to asbestos. Scanning Electron Microac. I: 563-578. Chatfield, E. J. 1982. Analytical procedures and standardization for asbestos fiber counting in air, water and solid samples. Pp. 91-107 in J. Small and E. Steel, eds. Asbestos Standards: Materials and Analytical Methods. NBS Spec. Pub. No. 619. National Bureau of Standards, Gaitheraburg, Md. Chatfield, E. J., and M. J. Dillon. 1978. Some aspects of specimen preparation and limitations of precision in particulate analyses by SEM and TEM. Scanning Electron Microac. I: 487-496. Churg, A. 1983. Nonasbestos pulmonary mineral fibers in the general population. Environ. Res. 31:189-200. Cook, P. M., and D. R. Marklund. 1982. Sample preparation for quantitative electron microscope analysis of asbestos fiber concentrations in air. Pp. 53-67 in J. Small and E. Steel, edn. Asbestos Standards: Materials and Analytical Methods. NBS Spec. Pub. No. 619. National Bureau of Standards, Gaitheraburg, Md. Davis, J. M. G., S. T. Beckett, R. E. Bolton, P. Callings, and A. P. Mlddleton. 1978. Mass and number of fibres in the pathogenesis of asbesto~-related lung disease in rate. Br. J. Cancer 37:673-688. Dement, J. M., R. L. Harris, Jr., M. J. Symons, and C. Shy. 1982. Estimates at dose response for respiratory cancer among chrysotile asbestos textile workers. Ann. Occup. Hyg. 26:869-~82. Dreessen, W. C., J. M. Dalla Valle, T. I. Edwards, J. W. Miller, R. R. Sayers, H. F. Eason, and M. F. Trice. 1938. A study of asbestosis in the asbestos testile industry. Public Health Bull. 241:217. Edwards, G. H., and J. R. Lynch. 1968. The method used by the U.S. Public Health Service for enumeration of asbestos dust on membrane filters. Ann. Occup. Hyg. Il:1-6. Holmes, S. 1965. Developments in dust sampling and counting techniques in the asbestos industry. Ann. N.Y. Acad. Sci. 132:288-297. Holt, P. F. 1957. Pneumoconlosis: Industrial Diseases of the Lung. E. Arnold & Co., London. 268 pp. Larger, A. M. 1979. Significance of aspect ratio in regulation of asbestos fiber exposure. Ann. N.Y. Acad. Sci. 330:601-604. Larger, A. M., and F. D. Pooley. 1973. Identification of single fibers in human tissues. Pp. 119-225 in P. Bogovski, J. C. Gilson, V. Timbrell, and J. C. Wagner, eds. Biological Effects of Asbestos, IARC Scientific Publ. No. 8. TuterDational Agency for Research on Cancer, Lyon. sampling and counting N.Y. Acad. Sci. 132:288-297. Industrial Diseases of the LUDQ.
95 Langer, A. M., A. Sastre, and I. J. Selikoff. 1971. Chrysotile asbestos in the lUDgs of persons in New York City. Arch. Environ. Health 22:348-361. Langer, A. M., A. D. Mackler, and F. D. Pooley. 1974. Electron micro- scopical investigation of asbestos fibers. Environ. Health Perapect. 9: 63-80. Langer, A. M., C. M. Maggiore, W. J. Nicholson, A. N. Rohl, I. B. Rubin, and I. J. Selikoff. 1979. me contamination of take Superior with amphibole gangue minerals. Ann. N. Y. Acad. Sci. 330:549-572. Leidel, N. A. ~ S. G. Bayer, K. A. Busch, and R. D. Zumwalde. 1979. USPHS/NIOSH membrane filter method for evaluating airborne asbestos fibers. NIOSH Pub. No. 79-127. National Institute for Occupational Safety and Health, Cincinnati. Liddel, F. D. K., G. W. Gibbs, and J. C. McDonald. 1982. Radiological changes and fibre exposure in chrysotile workers aged 60-69 years at Thetford mines. Ann. Occup. Hyg. 26: 889-898. Lynch, J. R. 1965. Asbestos study proceedings and findings. Trans- actions of the American Conference on Governmental Industrial Hy- gienists, May 1965. American Conference of Governmental Industrial Hygienists, Cincinnati. Lynch, J. R., and H. E. Ayer. 1966. Measurement of dust exposures in the asbestos textile industry. Amer. Ind. Hyg. Assoc. J. 27:431-443. Lynch, J. R., H. E. Ayer, and D. L. Johnson. 1970. The interrelation- ships of selected asbestos exposure indices. Amer. Ind. Hyg. Assoc. J. 31:598-604. Middleton, A. P., and E. A. Jackson. 1982. A procedure for the estima- tion of asbestos collected on membrane filters using transmission electron microscopy (TEM). Ann. Occup. Hyg. 25:381-391. Pooley, F. D. 1972. Electron microscope characteristics of inhaled chrysotile asbestos fiber. Br. J. Ind. Med. 29:146-153. Pooley, F. D., P. D. Oldham, H. Um~Chang, and J. C. Wagner. 1970. The detection of asbestos in tissues: Pneumoconiosis. Pp. 108-116 in H. Shapiro, ed. Proceedings of the Third International Conference. Oxford University Press, Capetown. Pott, F. 1978. Some aspects on the dosimetry of the carcinogenic potency of asbestos and other fibrous dusts. Staub-Reinhalt Luft 38:486-490. Rohl, A. N., A. M. Lange r, M. S. Wolff, and I. Weisman. 1976. Asbestos exposure during brake lining maintenance and repair. Environ. Res. 12:110-128. Samudra, A. U., C . F. Harwood , and J . D. Stockhalm. 1977. Electron microscope measurement of airborne asbestos concentration. EPA- 600/1-77-178. Environmental Protection Agency, Research Triangle Park, N.C. Selikoff, I. J., W. J. Nicholson, and A. M. Langer. 1972. Asbestos air pollution. Arch. Environ. Med. 25:1-13. Spurny, K. R., and W. Strober. 1981. Some aspects of analysis of single fibers in environmental and biological samples. Int. J. Environ. Anal. Chem. 9:265-281.
96 Thompson, R. J. 1978. Ambient air monitoring for chrysotile in the U.S. Pp. 355-363 in C. C. Gravatt, P. D. LaPleur, and K. F. J. Heinrich, eds. Proceedings of Workshop on Asbestos: Definitions and Measurement Methods. NBS Spec. Pub. No. 506. National Bureau of Standards, Gaitheraburg, Md. U.S. Occupational Safety and Health Administration. 1971. Occupational Safety and Health Standards, Part 1910.93 Air Contaminants (Gases, Vapors, Fumes, Dusts, and Mists). Fed. Regist. 36:10503-10506. Walton, W. H. 1982. me nature, hazards and asseasment of occupational exposure to airborne asbestos dust: A review. Ann. Occup. Hyg. 25:115-247.
