Document LJGzw775dZ481X8LNDLzb1a9g

R&S 041103 r Generic Damage in Man Caused by Environmental Agents "--------- ---- Copyrighted Materia! Do .Net Reproduce Without Publisher's Permission MONITORING FOR CHROMOSOMAL DAMAGE IN EXPOSED INDUSTRIAL POPULATIONS D, Jack Kitian Dante J. Picc'Cano Occupational Health and Medical Research Dow Chemical U.S.A. Freeport, Texas : . - W r ' A ` , . 4 The Texas Division of Dow Chemical U.S.A. is located about 50 miles south of the city of Houston, Texas. While the area immediately around our 3500 acre site is predominantly rural and residential, we are part of a large industrial and petrochemical complex that extends along the northern coast of the Gulf of Mexico and the southeastern coast of the North American continent. The Texas Division has some 7000 full time employees who work around or use a wide variety of chemi cals (Table I). These chemicals have become the subjects of intense investigation in recent years because of their toxic potential. For many .of these compounds, the potential for toxicity includes the possibility of genetic alteration in an individual's somatic or gametic cells. The possibility of genetic damage, especially the associ ation between an increased rate of chromosomal aberrations and carcinogenesis, is a matter of serious concern. One of the participants in this conference, Dr. James German (1972), has observed that many human cancers arise from a single cell and that this cell often has one or more chromosomes that differ in morphology from any in the complement of the noncancerous cells of the affected person. In addition, chromosomal insta bility, leading to an increased number of chromosomal abnor malities, mutation, and cancer is found in the several rare genetic disorders that feature increased chromosomal instabi lity together with a greatly increased expectancy for cancer to develops. Certain agents - radiation, some chemicals, and certain viruses - have in common the ability to produce in- Copyrighi 1979 by Academic Press, inc. 101 All rights of reproduction in an> form reserved. ISBN 0-12-089550-1 -,-rttti 102 Genetic Damage in Man TABLE I.- .Dow Texas_J)ivisionJ Freeportj Texas. 7000 employees. Manufacturer or user of a variety of chemicalsa including Benzene Toluene diisocyanate Ethylenimine Toluene diamine Vinyl chloride Phosgene Vinylidene chloride Carbon tetrachloride Trichloroethylene Chloroform Epichloro hydrin Hexachlorobenzene Hexachlorobutadiene creased rates of cells with chromosomal abnormalities and to increase the risk of malignancy. The comprehensive surveillance program established for Dow Texas Division workers is a good example of how cytogenet ic monitoring can be integrated with a system of industrial monitoring. The surveillance program (Table II) incorporates four major areas of investigation: Periodic Evaluation, which includes a history and physical examination at preset inter vals and appropriate laboratory tests; Cytogenetic Evaluation, done on a routine basis and in the event of accidental overex posure; Conventional Epidemiologic Studies, to determine the mortality and morbidity experience of exposed workers as com- TABLE XI. Vinyl Chloride (VC) Workers Comprehensive surveillance program Periodic evaluation History Physical Lab work Cytogenetic evaluation Routine exposures Over-exposures Conventional epidemiology Mortality Morbidity Special epidemiology Reproductive study (VC .workers' wives) age in Man lemioals. ind to for -ogenet;rial cates , which Lnt^rlu .on, overexie the is com- * 'f. v Chromosome Damage 103 pared-to a suitable control ..group; and Special Epidemiology, which, in a study underway now, will compare the reproductive patterns of the wives of vinyl chloride workers to the child bearing experience of the wives of a matched control group. The medical management team of Dow's Texas Division first became aware of the need for a program of cytogenetic monitor ing of certain workers in the early 1960s when reports of the mutagenicity of ethylenimine (El) in non-mammalian systems began to appear in the literature. Dow had started research on El in 1958 and started to develop a commercial process in 1961. Actual production began in 1965. Because of the early reports of possible mutagenicity, cytogenetic monitoring of El workers was made part of the medical surveillance program already in effect. When the cytogenetic program was started, the list of potential mutagens was short, and many compounds - such as vinyl chloride - were thought to be essentially harmless. Looking back, we can see that our first efforts at cyto genetic analysis were relatively crude and our findings in conclusive. Continued work on the many technical problems, however, allowed us to become more confident in our results and to expand the cytogenetic surveillance program to a number of other worker groups, including those exposed to benzene and vinyl chloride. Participation in a collaborative study of rat bone marrow cytogenetics was of particular help to us (Kilian et at. , 1977a, Kilian et at, , 1977b). Six cytogenetics laboratories, including our own, joined together, first, to devise methods to minimize interlaboratory differences and, second, to mea sure interlaboratory variation in results following the stand ardized administration of a known mutagen - triethylenemelamine - to rats from a common source. A preliminary workshop was held to resolve-scoring differences, to develop a joint protocol and common glossary, and to reach agreement on uni form reporting methods. This study showed (Fig. 1) a good pattern of agreement among the laboratories, particularly for certain categories of aberration. It also (Table III) allowed us to evaluate the variability of individual scorers in our own laboratory. As the list of potential chemical mutagens grew longer, the need for well-controlled cytogenetic studies of human effects, using simplified, repeatable methods became more pressing. We have previously published details of our methods (Kilian and Picciano, 1976); I would like to review these and the conditions we find are necessary for meaningful results. The groups being monitored should be of significant size and relatively stable and healthy. Employee turnover rates should be low, there must be adequate technology for accurate monit oring of both chronic and acute exposures, and there has to R&S 041105 104 CO UJ 50 - O _J < 5 oo: 40 - z CQ < Ll. 30 - O o_ f-- & 20 - S'8 (X. c Q_ 3 c in o < 10 0 Genetic Damage in Man TEM (pg/kg) FIGURE 1. Graph showing degree of agreement among six laboratories for averagearc sin /proportion of abnormal cells by laboratory and dose level of triethylenemelamine in cooper ative study of rat bone marrow cytogenetics. be medical and genetic expertise for the evaluation of what ever changes might be found in an individual. It is very help ful if exact levels of exposure to various compounds, alone or in combination, are known. The number of individuals available for study should be large enough to permit evaluation of match ed control groups. In addition, the availability of findings from periodic medical examination and records from comprehen sive health insurance plans - for the workers and their fami lies - allows the development of cooperative study plans in volving the industrial physician, epidemiologist, and cyto geneticist to evaluate not only the worker's medical status, but any reproductive consequences as well. ge in Man >0 0 .verage 3,20 ;o '0 six l cells ?ooper- vhat- { help- Lone or lilable E matchiings rehen- f ami5 in^toatus, Chromosome Damage TABLE III. Proportion abnormal Cells by Scorer and Dose 'for Laboratory 00 - - __________ __ ____________ Scorer number Dose quantity 000 100 200 300 400 01 .00 .00 .10 .31 .41 02 .00 .03 .19 .185 .66 03 .01 .02 .29 .205 .28 05 .03 .10 .20 .40 .72 07 .01 .02 .08 .21 .53 All scorers .010 .034 .172 .260 .538 All doses .16 .21 .16 .29 .17 .20 A standardized medical history - including exploration of family and occupational background - is taken prior to pre employment examination, and the worker's status in regard to those factors - such as exposure to chemicals, drugs, radiat ion, and virus infections - that are known to affect the chro mosomes is recorded. The medical status of workers exposed to toxic chemicals is reassessed at periodic intervals there after. Standard, culturing and harvesting procedures for peri pheral blood are used. Abnormalities are recorded on raw data sheets by cyto genetics laboratory technicians. The data are transferred to 80-column computer punch cards. Since one of our objectives is to set up longitudinal life-time studies to evaluate the consequences of chromosomal breakage, having the necessary information in computer.language is very helpful. Scored cells are identified by vernier setting and slide number. Early in our cytogenetic experience, we scored 20 cells per individual and, later, increased this to 50 cells, whenever possible. At the present time, however, we are analyzing at least 200 cells per culture to ensure the validity of the test system. As a general rule, results of cytogenetic evaluation of specific worker groups are compared to those found upon eva luation of preemployment examinees, that is, persons who have had chromosome analysis done as part of routine preemployment examination and who were judged, on the basis of the history taken at that time, to have no condition that would predispose them to increased aberration rates. In some cases, it is poss ible for the worker to serve as his or her own control. During the more than 12 years that the cytogenetics surveillance program has been in effect, more than 5000 cultures on more than 4000 individuals have been available for study. R&S 041107 106 Genetic Damage in Man A description of our cytogenetic study of vinyl chloride- exposed workers will "serve"'tfo' illustrate the potential useful ness of our surveillance program. Our cytogenetic study group was composed of 209 currently employed persons who had worked in the vinyl chloride plant for periods ranging between one and 332 months - an average of 48 months - at the time of the study. Findings were compared to cytogenetic data from a group of 295 preemployment examinees. The records selected for in clusion in the control group were matched to those of the study group, insofar as possible, for sex, number of cells analyzed, and time period during which the culture was initia ted. A detailed evaluation of the results of our vinyl chlo ride cytogenetics study is now in press (Picciano et at. , 1977). The major finding of the study, however, can be summa rized simply: that no cytogenetic differences of significance were found between the group of vinyl chloride-exposed workers and the preemployment control group. Table IV shows the compari son of the percentage findings for chromatid and chromosome breaks; rings, dicentrics, and exchanges; and the proportion of abnormal cells. Table V displays the results of Chi-square analysis for differences in the distribution of chromatid breaks between the two groups. Both groups were divided into those showing 0-to-5-percent aberrations and those showing greater-than-5-percent aberrations; the vinyl chloride workers were separated into those with estimated average exposure levels of less-than-one part per million (ppm), from l-to-5 ppm, and greater-than-5 ppm. The same analysis was used to i i devaluate for differences in the distribution of chromosome breaks (Table VI) and the proportion of abnormal cells (Table VII). We would like to emphasize that our work-force has been * exposed to relatively low levels of vinyl chloride and that we believe this has a definite bearing on our results. It is to be hoped that our demonstration of lack of significant % TABLE IV. Cytogenetic Study of 209 Workers Eogposed to Vinyl Chloride .1 i, H H .... Workers Controls 4 a<v Mi ii Number of cultures 209 295 !I 1 *i -T! i; u $t Number of cells Chromatid breaks Chromosome breaks 10,483 2.4% 1.0% 14,761 3.6% 1.1% 33 if) n Rinas, dicentrics, and exchanges 0.4% 0.2% o Abnormal cells 3. 7% 4. 5% O 00 f- i ? in Man rride3ef"l?r rrked one c the group in- Is nitia- nlo- r ammaranee orkers mpariome tion guare d into ag orkers o-5 to ome Table been hat t is t to Is ?5 n 3,6% 1,1% 0,2% 4. 5% - Chromosome Damage 107 TABLE V. Distribution of Chromatid Aberrations Related .to Vinyl,Chloride (VCl Exposure Exposure to VC 1 <1 ppm 1-5 ppm >5 ppm Controls Number in group Percentage of group with 0-5% aberrations 209 70 90 98 7? 41 80 295 75 Percentage of group with >5% aberrations 10 23 20 25 x2rsr 7-7S (p^O.OS) Exposure levels are estimates based on .calculations for specific job c lassifications; prior exposure for individuals may have been higher or lowerj exposure levels for individuals are known to vary within job classification TABLE VI. Distribution of Chromosome Aberrations Related to Vinyl Chloride (VC) Exposure Number in group Percentage of group with 0-5% aberrations Percentage of group with >5% aberrations Exposure to VC <1 ppm 1-5 ppm >5 ppm Controls 209 70 98 - 41 295 96 94 95 94 4 6 5 6 x2f3r (Pr^0.95) effect at these levels will be useful in establishing levels of exposure that are genetically safe. In 1974-75, a group of five reports appeared in the lit erature concerning chromosome aberrations in vinyl chloride and polyvinyl chloride workers (Table Villa) . Three of the study teams concluded that an increased rate of chromosomal aberra tions was associated with vinyl chloride exposure (Ducatman et al, r 1975; Funes-Cravioto et al. , 1975; Purchase et at., 1975), while two studies - one of them a preliminary investi- R&S 041109 108 Genetic Damage in Man TABLE VII. Distribution of Abnormal Cells Related to Vinyl Chloride (VC-) -Exposure - ....... ........... Number in group Percentage of group with 0-5% aberrations Percentage of group with >5% aberrations Exposure to VC <1 ppm 1-5 ppm > 5 ppm Controls 209 70 98 41 295 84 71 73 70 16 29 27 30 X2= 5.97 (P-- 0.12) TABLE Villa. "Published Cytogenetic Studies of Workers Exposed to Vinyl Chloride Authors Cases reported Exposures Workers Controls Type Magnitude IDJ J U Ducatman et al. 11 10 PVC Estimated Positive to exceed 500 ppm Funes-Cravioto et al. 7 3 PVC Decreased Positive in recent years to 20-30 ppm Purchase et al. 56 24 PVC Not given Positive Kilian and Picciano 203 108 VC 1969-74: average 5 ppm; 1974-75: 1-2 ppm Negative Fleia and Thiess 10 4 PVC Until Negative 1974: >100 ppm; 1975: 1025 ppm .-*i' t-.V .''21 .J3S & -ffi jfi.; * 3L ige-in Man 1 to ; >5% S <ers Effect ositive ositive ositive egative egative Chromosome Damage 109 TABLE VUIb. Conditions (Neoplastic or with Neoplastic Association) Displaying Chromosome Anomalies^ Chronic Myelocytic Leukemia Acute Myelocytic Leukemia Chronic Lymphatic Leukemia Polycythemia Vera Myeloid Metaplasia Erythro leukemia Sideroblastic Anemia Meningioma Burkitt's Lymphoma Retinoblastoma in D-, Ataxia-Telangiectasia Fanconi 's Anemia Bloom1 s Syndrome Kostmann's Infantile Agranulocytosis Glutathione Reductase Deficiency Down 's Syndrome Turner's Syndrome Klinefelter's Syndrome Wi Ims ' Tumor-Aniridia Syndrome etion .Syndrome 1 Source: MuIvihill3 1975 gation from our laboratory - appeared to show no differences of statistical significance between exposed and control groups (Fleig and Thiess, 1974; Kilian et al. , 1975). These studies tend to indicate a dose-related response since the exposure levels of polyvinyl chloride workers to the vinyl chloride monomer are, generally, much higher than the exposure of vinyl chloride manufacturing workers (Occupational Safety and Health Administration, 1974). The indication of a dose-related effect was reaffirmed by further study of vinyl chloride workers at Imperial Chemical Industries in which increased aberration rates were significantly correlated to length and intensity of exposure (Purchase et al. , 1976). It is known that increased rates of chromosomal breakage, as have been demonstrated in cases of exposure to high levels of vinyl chloride, have occurred following x-ray treatment of ankylosing spondylitis (Buckton et al. , 1962), in atomic bomb survivors (Bloom et al. , 1970), radium dial painters (Vaughan, 1962). persons exposed to Thorotrast (Fischer et at., 1966), and workers exposed to lead (Garza Chapa et al. , 1976), benze ne (Vigliani and Forni, 1976), uranium dust, and plutonium. Increased rates of chromosomal breakage are also seen in a variety of diseases (Table VUIb) in which an increased risk of cancer is also evident (Mulvihill, 1975). Evidence of a relationship between chromosomal abnormali ties and increased risk of neoplasia has continued to accumu late, although the exact nature of this relationship is far from clear. It is because of this evidence that we believe cytogenetic monitoring can be an extremely valuable tool in the detection of those environmental situations that may be R&S 041111 no Genetic Damage in Man associated with an increased risk of cancer and, perhaps, reproductive problems for the chemically exposed worker as well-. A cytogenetic program and periodic medical evaluation, however, are not enough to ensure comprehensive surveillance of worker groups at risk. Epidemiology programs, both conven tional and special, are also essential. Regarding our epide miology efforts, we are fortunate to be reasonably close to the University of Texas Medical Branch at Galveston. Under the direction of University of Texas epidemiologist. Dr. Patricia Buffler, a unique epidemiology program was started a few years ago; the heart of this program is the cooperative efforts put forth by both our own industrial concern and academia. Univer sity of Texas employees actually work at the plant site on a continuous basis to identify and process necessary records for later analysis at the medical school. Complete confidentiality is maintained through the use of various coding systems. Epidemiologic analysis of our vinyl chloride workforce showed that the employees, past and present, fell into one of three major assignment areas (Table IX): the vinyl chloride plant itself, the maintenance units, and the developmental laboratory. Dozens of specific job classifications were iden tified, and we were able to arrive at fairly firm estimates of actual or potential exposure for most of them. The estimates were based on both personnel and area monitoring; Table X shows a few of them. It needs to be noted that actual exposur es will vary from person to person within the same job classi fications and that accidental, short-term exposures of some workers to concentrations in excess of the mandatory standard probably occur from time to time. Documentation of such inci dents is usually difficult. The current U.S. standard for - JO CD O M TABLE IX. Vinyl Chloride (VC) Cohort for Historical--Pro spective Study of Mortality Experience. Worker Classifications VC production plant Operators Loaders Supervisors Quality control laboratory Maintenance workers Electricians Boiler-makers Pipefitters Others Developmental laboratory iar. -Z SF \r. - ;e in Man .on, .ance >nven>ide; to ler the :ricia / years :s put Jniveron a rds for :iality )rce me of ide :al idenites of ja f* " X :posur:lassi;ome mdard inci>r \l-Proations Chromosome Damage III TABLE X. Estimated Exposure to Vinyl Chloride (VC) for Vinyl Chloride-Related Job-Classifications Job classification Estimated exposure in parts per millionj as time-weighted-average 1973-1974 1960-1972 Before 1960 R&D engineer Production engineer Quality control laboratory Instrument man Painter Machinist Material handler Production foreman Control A OP. (Vinyl) Control B OP. (Oxy 8 Chloride) Control C OP. (Chlorination) 1.7 1. 7 8. 7 1.3 0. 7 2.4 0. 5 1.3 2.8 0. 7 2.6 4.4 4.4 11.4 4.0 0. 7 5. 1 0. 5 4. 0 5.5 0.7 5.3 8.2 8.2 15.2 7.8 0. 7 8.9 0. 5 7.8 9.3 0.7 9.1 TABLE XI. Vinyl Chloride (VC) Workers: U.S. and Dow Standards for Exposure 1948-1958 1959 1959-1967 1968-1973 Before 1974 1974 1974 1975 Dow Dow DoWj Texas Division Dow3 - Texas Division Other U.S. VC producers Other U.S. VC producers Dowj Texas Division Dow3 Texas Division No specified standard; exposure levels unknown 50 ppm} averaged concentration> established as air quality goal Estimated average exposures: below 50 ppm Estimated average exposure: below 5 ppm No specified standard; exposure levels unknown; some very high Threshold limit value of 500 ppm recommended by ACGIH1 as industry guideline; voluntary compliance Estimated average exposure; 1-2 ppm Estimated average exposure: below 1 ppm 1American Conference of Governmental Industrial Hygienists R&S 041113 112 Genetic Damage in Man TABLE XII. Vinyl Chloride (VC) Cohort for HistoricalProspective Study of Mortality Experience Cohort membership: Types of exposure: On Dow payroll Actual exposurej or potential exposure Exposed at least 4 consequtive work- weeks Continuous Intermittent l i i i i TABLE XIII. Vinyl Chloride (VC) Cohort for HistoricalProspective Study of Mortality Experience. Cohort Membership Currently employed Formerly with VC unit Presently with VC unit Present. status determined 100% Formerly employed Retired Terminated Disabled Dead Present status determined 100% vinyl chloride is 1 ppm as a time--weighted--average over an 8- hour day. Table XI -illustrates, the changes in the standard for vinyl chloride that have occurred over the years. In defining our vinyl, chloride cohort for epidemiologic study, it was decided to include (Table XII)- only those on the Dow payroll with documented exposure, or significant potential for exposure, to vinyl chloride for at least a four-week-con secutive period. We were able to identify a total of 533 per sons meeting these criteria; this group is composed of persons who are working now, or who have worked previously, in the vinyl chloride plant and those who worked with vinyl chloride 33 ft (/) in the past, but are no longer employed by Dow in any depart ment. The present status of our currently employed workers o (Table XIII) was relatively easy to ascertain, but it took con siderable effort to arrive at the 100% follow-up figure for former employees. It is our belief, however, that close to 100% follpw-up is necessary for proper assessment of the risks that may be faced by any group of chemical workers. The vinyl chloride cytogenetics study is one of several Tr involving Texas Division employees. A special epidemiologic  in Mon il- \ Chromosome Damage 113 TABLE XIV. Reproductive Consequences of Vinyl Chtovi.de Exposure:' Survey of ~WdrkmeWt~s~ wives'" a? flo co o 4* sure Participation Number of women Percentage of group Ol Study group Control group 205 88% 144 81% iclI)hip i 8- 1 for jic the itial :on-- >er-- rsons Z - ride irt- " 1 . con>r * isks a' c TABLE XV. Comprehensive Surveillance Program for All Workers Ex*uosed to Toxic Chemicals Periodic Evaluations History and Physical Examination Laboratory Tests Computerize Findings Mutagenic Evaluation Cytogenetic Monitoring Routine Exposures Accidental Over-Exposures Other Mutagenicity Tests Conventional Epidemiology Mortality Morbidity Special Epidemiology Reproductive Studies Hospital Records .. Vital Statistics study is now underway to investigate the possibility that in creased rates of fetal loss or birth defects in offspring are experienced by wives of vinyl chloride workmen. We have had a gratifying response from the women - both in the study group and a control group - who were asked to participate. The study group was composed of women married to currently employed Dow workmen identified as having had at least two consecutive months of exposure to vinyl chloride. The comparison group was made up of wives of currently employed workmen who did not work in production, a laboratory, or the maintenance depart ment. As shown in Table XIV, almost 350 women, 88 per cent of the study group, and 81 per cent of the control group, agreed to discuss these personal matters with the trained. University ! i ft u I 3 i t* HiW - r. H4 Genetic Damage in Man of Texas interviewersThe~lnterviewing is complete, the data are being analyzed, and we hope to have a report of the final results in the literature soon. Our major contribution to this conference is to demon strate the feasibility of studying the whole person and that person's family through coordinated, long-term investigations (Table XV). Life-disease processes will be recognized and re corded by medical monitoring and insurance programs, mutagenic exposures and somatic consequences will be recognized by app ropriate test methods, and conventional and special epidemio logic studies will be used to evaluate occupational disease and germinal mutation problems. It is only through such a co ordinated effort that the hazards of chemical exposure to the human species can be accurately evaluated. Such a program not only benefits the individual worker and family, but society as a whole. i t * 1 |i i t 1 t i t DD fio C/> O 4> a REFERENCES Bloom, A.D., Nakagoma, Y., Awa, A.A., and Neriiski, S. (1970). Am. J. Public Health. 60, 641. Buckton, K.E., Jacobs, P.A., Court Brown, W.M., and Doll, R. (1962). Lancet. 2, 676. Ducatman, A., Hirschhorn, K. and Selikoff, I.J. (1975). Muta tion Res. Zl3 163. Fischer, P., Golob, E., Kunze-Muhl, E., Haira, A.B., Dudley, R.A., Mullner, T., Parr, R.M., Radiat. Res. 29, 505. Fleig, I., and Thiess, A.M. (1974). Praeventimed. 9,260. and Vetter, Arbeitsmed. H. (1966). Sozialmed. Funes-Cravioto, F., Lambert, B., Lindstein J., Ehrenberg, L., i Natarajan, A.T., and Golkar, S. (1975). Lancet. 2, 459. Garza Chapa, R., Leal, C.H., Alvarez, M., and Sanchez, F.J. (1976). In "Abstracts V International Congress of Human Genetics" (s. Armendares and R. Lisker, eds), no. 325, Excerpta Medica, Amsterdam. German, J. (1972) . Progr. med. Genet. 83 61. Kilian, D.J., and Picciano, D. (1976). In "Chemical Mutagens: Principles and Methods for Their Detection", Vol. 4 (A. Hollaender, ed.), p. 321. Plenum, New York. Kilian, D.J., Picciano, D.J., and Jacobson, C.B. (1975). Ann. N.Y. Acad. Sci. 269, 41. Kilian, D.J. Moreland, F.M., Benge, M.C., Legator, M.S. and X: ir } ;- Whorton, E.B., Jr. (1977a). Mutation Res.j in press. br- - Kilian, D.J.> Moreland', F.M., Benge, M.C., Legator, M.S. and . ip ^ Whorton, E.B., Jr. (1977b) . In "Handbook of Mutagen Testing" (B.J. Kilbey, ed.), Elsevier/North Holland, Amsterdam. t i Chromosome Damage IIS Mulvihill, J.J. (1975). In "Persons at High Risk of Cancer, An Approach to Cancer Etiology and Control" (J.F. Fraumeni, Jr., ed.), p. 3. Academic Press, New York. Occupational Safety and Health Administration (1974). Federal Register. 39, 35889. Picciano, D.J., Flake, R.E., Gay, P.C., and Kilian, D.J. (1977). J. Occup. Med., in press. Purchase, I.F.H., Richardson, C., and Anderson, D. (1976). Proc. Roy. Soc. Med. 69, 32. Purchase, I.F.H., Richardson, Lancet. 2, 410. Vaughan, J. (1962). Int. Rev. Vigliani, E.C., and Forni, A. C., and Anderson, D. (1975). Exp. Path. 1, 243. (1976). Environ. Res. 11 , 122. R&S 041117