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.
Table A-2 (continued) Structural Formula Name 12 1 11 ~ 2 Chrysene 9 ~ 3 7 2 1 Benzo[a]pyrene ~,'Z ~ 3 5 sW6 Benz[e]pyrene ~Perylene 9 ~ 7 8 ~ S 1^ ~ .- I ~ 3 Molecular Weight 228.0939 0/+ 252.0939 ++ 252.0939 0/+ 252.0939 0 Benzo[j~fluoranthene 252.0939 ++ A-11 Carcinogenic Activity ~... ..
Table A-2 (continued) Structural Formula Name 2 1'~s 9 8 7 2 44 18 1l ~ 13 Molecular Carcinogenic Weight Activity Benz~e]acephenan- thrylene (Benzotbifluoranthene 252.0939 ++ Benzo~k~fluoranthene 252.0939 2 7H-Dibenzotc,g~car- 267.1048 2 5 7H-Dibenzota,g]car- 267.1048 + bazole 11 2 4 -b Dlbenzo[a,i]car- 267.1048 A-12 A; ++ + - . . .
Table A-2 (continued) Structural Formula 3 ~2 ~ 7 12 2 3 4 12 2 34 8 7 6 2 12 ~ 4 IO ~ 8 9 12 1 1l ~ 2 Molecular Carcinogenic ___ Weight Activity Benzo[~]naphtho[l, 2- f ] - 279 a 1048 + qulno. . 1ne Dibenz[a,j]acridine 279. 1048 + Dibenz[c,hiacridine 279.1048 + 13 Dibenz[a,h]acridine 279.1048 + Benzo[ghi]perylene 276.0939 + A-13
Table A-2 ~ continued ~ Structural Formula Name . _ 1l 12 9~3 _.. 9~3 1~3 12 0~8 2 I `~3 'I 12 ION ,7 9 8 3 2~4 1~6 7 2 19 11 R0 Molecular We ight u~uenzo [clef ,mno I-276.0939 chrysene ~ Anthanthrene ~ Indeno ~ 1, 2, 3-cd ~ pyrene276.0939 + Dibenz [a ,h ~ anthracene ~ 278.1096 + Benzotcichrysene278.1096 + Benzo ~ g ~ chrysene A-14 Carc inogenic Activity lo 278.1096 +
Table A-2 (continued) Molecular Carcinogenic Structural Formula Name Weight Activity 2 3 8 10[ ~ 3 8 7 6 2 21 ~ 4 - 2 I,> ~ 14 12 2 ~o~l. 8 7 6 Picene 278.1096 0 14 1 13 ~ 2 Benzotb]chrysene 278.1096 0 Benzotb~triphenylene 278.1096 (Dibenz~a,c~anthracene) + Pentaphene 278.1096 0 (Dibenzo[b,h]phen anthrene) Dibenz[a,j]anthracene 278.1096 + A-15
Table A-2 (continued) S tructural Formula Name |2 ~ t0~3 Molecular We ight - Coronene 300.0939 2 14 ~Benzo[rat]pentaphene 302.1096 ++ 11 ~ (Dibenzo[a,i]pyrene) 9 8 7 13 14 1 12~2 Dibenzo [b, de f ] chrysene 302. 1096 IO ~ (Dibenzo[a,h~pyrene) ~ 9 ~ 5 1346 12~8 11 10 9 11~2 a NA = not available. Care inogenic Ac t ivity 0/+ ++ Dibenzo[def,p]chrysene 302.1096 ++ ( Dibenzo [ a, 1 ] pyrene ~ Naphtho[ 1,2 ,3,4-def]- 302. 1096 ++ chryeene ( Dibenzo [ a, e Jpyrene A-16 a. .. . .
TABLE A-3 Nitroarenes Detected in Diesel-Exhaust Particulate Extracts: Molecular Formulas and Molecular Weights Struc ture Molecular Molecular No. Name Formula Weight Mononitroarenes: 1 Nitroindene C9H7NO2 161.16 2 Nitroacenaphthylene C12H7NO2 197.19 3 N~troacenaphthene C12HgNO2 199.21 4 N~trobiphenyl C12H9NO2 199.21 5 Nitrofluorene C13H9NO2 211.22 6 Nitromethylacenaphthylene C13H9NO2 21L.22 7 Nitromethylacenap~thene C13HllNO2 213.24 8 Nitromethylbiphenyl C13HllNO2 213.24 9 Nitroanthracene C14HgNO2 223.23 10 Nitrophenanthrene C14H9NO2 223.23 11 Nitromethylflourene C14HllNO2 225.25 12 Nitromethylanthracene ClsHLlNO2 237.26 13 Nitromethylphenanthrene C15HLlNO2 237.26 14 Nitrotrimethylnap~thylene C13H13NO2 215.25 15 N~trofluoranthene C16HgNO2 247.25 16 N~tropyrene C16HgNO2 247.25 17 Nitro(C2-alkyl~anthracene C16H13NO2 251.29 18 N,tro(C2-alkyl~phenanthrene C16H13N~2 251.29 19 N~trobenzofluorene C17HllNO2 261.28 20 Nitromethylfluoranthrene C17HllNO2 261.28 21 Nitromethylpyrene C17H12NO2 262.29 22 Nitro(C3-alkyl~anthracene C17H15NO2 265.31 23 Nitro(C3-alkyl~phenanthrene C17H15NO2 265.31 24 Nitrochrysene C18HllNO2 273.29 25 Nitrobenzanthracene C18HllNO2 273.29 26 Nitronaphthacene C18HllNO2 273.29 27 Nitrotriphenylene C18HllNO2 273.29 28 Nitromethylchrysene Cl9H13NO2 287.32 29 Nitromethylbenzanthracene Cl9H13NO2 287.32 30 Nitromethyltriphenylene ClgH13NO2 287.32 31 Nitrobenzopyrene C20HllNO2 297.31 32 N~troperylene C20HllNO2 297.31 33 N~trobenzofluoranthene C20HllNO2 297.31 Polynitroarenes: 34 Dinitromethylnaphthylene CllHgN2O4 233.20 35 Dinitrofluorene C13H8N2O4 256.22 36 Dinitromethylbiphenyl C13HloN2o4 258.23 37 Din~trophenantl~rene C14H8N2O4 268.23 38 D~-nitropyrene C16H8N2O4 292.25 39 Tr~nitropyrene C16H7N3O6 337.25 40 Trinitro(C5-alkyl)fluorene C18H17N3O6 371.35 41 Dinitro(C6-alkyl~fluorene Cl9HlgN2O4 339.37 42. Dinitro(C4-alkyl~pyrene C20H16N2O4 348.36 A-17 =. ...
Table A-3 (continued) Struc ture Molecular Molecular NO.a Name Formula Weight . Nitro-oxyarenes: - 43 NitronapUthaquinone CloH5NO4 203.L5 44 Nitrodihydroxynap~thalene CloH8NO4 206.18 45 Nitronaphthalic acid CloH8NO4 206.18 46 Nitrofluorenone C13H7NO3 225.20 47 Nitroanthrone C14H9NO3 239.23 48 Nitrophenanthrone C14H9NO3 239.23 49 Nitroanthraquinone C14H7NO4 253.21 50 Nitrohydroxymethyliluorene C14HllNO3 241.25 51 Nitrofluoranthone C16H~NO3 262.24 52 Nitrofluoranthenequinone C16H7NO4 277.24 53 Nitropyrenequinone C16H8NO4 278.24 54 Nitropyrone C16HgNO3 263.25 55 Nitrodimethylanthracene C17H12NO3 278.29 carboxaldehyde 56 Nitrodimethylphenanthrene C17Hl2No3 278.29 carboxaldehyde Other nitrogen compounds: 57 - Benzocinnoline C12HgN2 180.21 58 Methylbenzocinnoline C13HloN2 194.24 59 Phenyloaphthylamine C 16H13N 219. 29 6 0 ( C2-Alkyl ) phony 1 naph thylamine C 1 SHE 7N 247 . 34 aStructure numbers refer to structures in Table A-4. A-18 ~.
TABLE A-4 Struc Lures of Nitroarenesa H H N02 ¢ 1 N 0 2~3 AS N02~1 9J12~1 7/ 22 o 11 No2~3 H H ~7 o ., No2~3 11 o ~9 N02~ 27,30 No7~1 I 2,6 ~ H No2~3 5.L1 N O 2 10,1~.1l,~ it, N02 16, 2 N 0 2 2`,2 N O2-~`J 31 - C~5 H ~3 59,60 Hi; A-19 . ~ H H l ,H _ _ 1 NORM ~ ~ a,7 o C- OH N02~ - N02~ J NO 2 15,20 No2~3 N02 32 ^. ... .
NO 2~--3 NO2--~X63 9 25 29 o en. NO,-~ ,,,: 26 · ~ _~ 54 it' NO2 NO2-~ ( C~3)3 14 Nit 3 N O -~3 N O 2-~ ~ o H )2 lo 57,58 o ~C`oH NO2 ~5 NOW 0 48 ,, N O2-Id 54 o 11 NO 2- H ~7 o ll N 0 2- ~6 O ,0 N 0 2- ~N 0 2-- o " O C-H " NO2- ~NO2- 55 , ~53 ~6 "Numbers under structures refer to compounds listed in Table A-3. A-20 51 hi. . .
APPENDIX B POLYCYCLIC AROMATIC HYDROCARBONS IN THE AMBIENT ATMOSPHERE Compound Unsubstituted: Biphenyl Naphthalene Anthracene Phenanthrene Benz~aJanthracene Dibenz~ac~anthracene Benzotc~phenanthrene Benzota~fluorene Benzotbifluorene Dihydrobenzo~a,b , and c]£1uorenes Fluoranthene Benzotbifluoranthene Benzo~j~fluoranthene Benzo~kifluoranthene Benzotghi~fluoranthene Pyrene Benzota~pyrene Benzoteipyrene Anthanthrene (dibenzo~cdjk~pyrene) Dibenzopyrenes (4 isomers) Indeno(1,2,3-cd~pyrene Chrysene Perylene BenzotghilperyLene Coronene Picene Benzotc~phenanthrene Benzotb~chrysene Benzotc~tetraphene Hexahydrochrysene Dihydrobenzo~ciphenanthrene Dihydrobenz~aJanthracene Dihydrochrysene Benzacenaphthylene Binaphthyl (3 isomers) Quarterphenyl DiphenylacenapUthalene Ambient concen tration ? ng/m3 References a 0.05-0.35 0.07-6.15 0.04-25 0.5-22 0.03-4.5 0.04-1.0 0.8 O. 1-1. l 0.03-0.9 0.1-41 0.1-7.4 0.2-4.4 0.14-20 O . 9-9 . l 0.1-35 0.1-75 0.1-42 0.1-6 a,b 1-12.8 0.2-39 0.1-5 0.2-46 0.~-48 a a a a a a a a b b b b B-1 6 6 6 6 6 6 6 1.6 6 6 6 6 6 6 6 6 6 4,6 6 6 6 6 6 1 ll 4 4 4 4
Compound Alkyl-substituted: Methylanthracene 0.22-0.66 6 1-, 2-, 3-, and 9-Methylphenan threnes b 4 1-~1ethylpyrene 0.01-0.15 6 1-, 2-, and 4-Metoylpyrenes b 4 EthylanthraceneC a 1,4 EthylphenanthreneC 1, ~ Methylfluoranthene (5 isomers) a 1,4 Methylbenz~alanthracened a 1 Methylchrysened a 1 MethYlbenzotbkifluoranthene a 1 Methylbenzo~ae~pyrene a 1 Methylbenzopyrenes or benzo fluoranthenes (5 isomers) b 4 4H-Cyclopenta~def~phenanthrene b 4 Methyl 4H-cyclopenta~def~phen anthrene b 4 ~ ~ ~ , thy ~ 4H-cyclopentaldef~phenanthrene (5 isomers) b 4 Ethylmethyl 4H-cyclopenta~def] phenanthrene b 4 Ethylmethyl anthracene or phenan threne b ~ 4 Ethylpyrene or fluoranthene (4 isomers) b 4 Ethylmethylpyrene or fluoranthene (3 isomers) b , 4 Methylbenzotciphenanthrene b 4 Methylbenzotghi~fluoranthene b 4 Ethy~chrysene or benz~aJanthracene (7 isomers) b 4 Methylbinaphthyl (4 isomers) b 4 Methyldibenzanthracene b 4 N-Hetero (aza): _ Acridine 0.04 6 Hethylacridine 0.007 6 Benz~aJacridine 0.9 6 Benz~c~acridine 0.1-1.5 6 Dibenz~aj~acridine 0.04 6 Dibenz~ahiacrid ine 0 . 08-0 . 1 6 Carbazole 1.9 6 Quinol ine 0. 02-0 .6 6 Me thylquino 1 ine 0.03 6 2 , 6-D imethylquinol ine 0 .03 6 Dimethylquinolines 0. 04-0.09 6 Ethylquinol ines 0. 01-0 .02 6 03 Alkylquinolines 0.01 6 B-2 Ambient concen tration, ng/m3 References ~...
Compound Benzo~fJquinoline Benzoth~quinoline 11-Indenot1,2b~quinoline Phenanthridine Isoquinoline Methylisoquinolines Dimethylisoquinolines Ethylisoquinolines C3 Alkylisoquinolines Benz~f~isoquinolines 4-Azafluorene 4-Azapyrene and isomers 1-Azafluoranthene Benzotcicinnoline 2-Methylindole Benzota] carbazole Benzotc] carbazole Phenoxazine C; Alkylquinolines Me thylphenanthr id ine s Methylbenzoquinolines Me thylbenzoisoquino 1 ines Azabenzofluorenes Methylazapyrenes Me-thylazafluoranthenes Azabenz~aJanthracene Azachrysenes Azabenzopyrenes Azabenzofluoranthenes Dibenzoquinolines Dibenzoisoq~inolines Quinones: 9,lO-Anthraquinone Benzotaipyrene 6,12-quinone Benzota~pyrene 1,6-quinone Benzo~aipyrene 3,6-quinone Dibenzo~b,def~chrysene 7,L4-quinone Phenaten-l-one Benzanthrone Perinaphthanone Carboxylic acids: Nap~thalene carboxylic acid Phenanthrene carboxylic acid Anthracene carboxylic acid Pyrene carboxylic acid Ambient concen- tration, ng/m3 0.01-0.2 0.01-0.3 0.1 0.02 0.14-0.18 0.17-0.31 0.06 0.07-0.16 0.03 0.03-0.11 0.005 0.02-13 trace-3 1.0 .0 a a a a a a a a a a a a a a a b b b b b 0.3-17 0.6-48 a a B-3 References 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 1 l 1 l ll 5,6 6 1 .. ....
Compound Phenols: Hydroxyanthracene Hydroxyphenanthrene Hydroxypyrene Hydroxyfluoranthene S-Hetero: - Benzothiazole Dibenzothiophene Methyldibenzothiophenes (3 isomers) Ethyldibenzothiophene Benzo~defidibenzothiophene Naphthobenzothiophenes (3 isomers) Methylnaphthobenzathiophenes (3 isomers) Nitro derivatives: 1-Nitropyrene 3-Nitrofluoranthene 5-Nitroacenaphthene 6-Nitrobenzota~pyrene Ambient conce~- tration, ng/m References a 0.014-0.02 b b b b b b b b b l 2 4 4 4 4 4 4 7 3 aConcentration reported in micrograms per gram of part iculate matter or micrograms per gram of benzene-soluble fraction, but not in nanograms pe cubic meter. Compound identified, but no concentration reported. CEight isomers of ethylanthracene/ethylphenanthrene identified. drive isomers of methylbenz~aJanthracene/methylchrysene identified. B-4 me_
REFERENCES 3. 5. Cautreels, W., and K. Van Cau~enberghe. Determination of organic compounds in airborne particulate matter by gas chromatography-mass spectrometry. Atmos. Environ. 10:447-457, 1976. 2. Dong, M. W., D. C. Locke, and D. Hoffmann. Characterization of aza arenes in basic organic portion of suspended particulate matter. Environ. Sci. Technol. 11:612-618, 1977. lager, J. Detection and characterization of nitro derivatives of some polycyclic aromatic hydrocarbons by fluorescence quenching after thin-layer chromatography. Application to air pollution analysis. J. Chromatogr. 152:575-578, 1978. Lee, M. L., M. Novotny, and K.~. Bartle. Gas chromatography/mass spectrometic and nuclear magnetic resonance determination of poly nuclear aromatic hydrocarbons in airborne particulates. Anal. Chem. 48:1566-1572, 1976. Pierce, R. C., and M. Katz. Chromatographic isolation and spectral analysis of polycyclic quinones. Application to air pollution analysis. Environ. Sci. Technol. 10 :45-51, 1976. Santodonato, J., P. Howard, D. Basu, S. Lande, J. K. Selkirk, and Pe Sheehe. Health Assessment Document for Polycyclic Organic Matter. EPA-600/9-79-008. Research Triangle Park, N.C.: U. S. Environmental Protection Agency, Office of Health and Environmental Assessment, Environmental Criteria and Assessment Office, 1979.  pp. (preprint) Tokiwa, H., R. Nakagawa, and Y. Ohnishi. Mutagenic assay of aromatic nitro compounds with Salmonella typhimurium. Mutat. Res. 91: 321-325, 1981. B-5
APPENDIX C HUMAN-CANCER RISK ASSESSMENT* Malcolm C. Pike Epidemiologic studies, animal carcinogenesis experiments, and in vitro mutagenesis and transformation assays all provide data relevant to the assessment of the human-cancer risk from exposure to PAHs. The data used in this assessment are in the main taken from epidemi- ologic studies, because they refer directly to man. It is recognized that an alternative approach would have been the extrapolation of experi- mental animal data to humans, but the epidemiologic approach offers two advantages : the avoidance of interspecies extrapolation and the derivation of results from exposures not too different from that suffered by the general population. Epidemiologic studies often suffer from various inadequacies, such as imprecise dose measurements and poor measurement of confounding factors, and exposure is invariably to ~ complex mixture of PAHs and other chemicals. Extrapolation to other complex mixtures therefore inevitably involves making assumptions, and evidence from in vitro and in vivo experiments must be sought to provide a rational basis for these assumptions. At present, the two sources of human exposure to PAHs on which data appear reliable are work around coke ovens and cigarette-smoking. The major known human cancer associated with exposure to chemical mixtures containing PAHs is undoubtedly lung cancer. Although cigarette-smoking is of overwhelming importance as a cause of lung cancers ~ 40 and cigarette smoke does contain PAHs, this appendix is concerned with cigarette-smoking only insofar as the information derived from epidemiologic study of the smoking population is essential in measuring the health effects that might be expected when Germans are exposed to other PAH-containing mixtures. The quantitative relationship between cigarette-smoking and lung cancer has been thoroughly explored in many epidemiologic studies-° and is well understood.8~2 However, it is still far from established that the PAR content of cigarette smoke is responsible for the development of lung cancer. Epidemiologic data (mainly occupational) on the relationship *Quantitative risk assessment is ~ developing, rather than a precise, science. The numerical estimates in this appendix are based on a series of assumptions. The use of different assumptions or extrapolations from animal data could lead to very d i f ferent cone lus ions. The calculated risk values at ambient concentrations are not meant to be absolute indicators of risk, but rather to indicate the region between the upper bounds of risk and the lower bound of zero risk. C-l
between exposure to other PAH-containing mixtures and lung cancer are much less precise. The lung-cancer risks (as well as the risks of cancer at other sites) associated with such exposures have, in fact, always been measured in relation to lung-cancer rates in the "nonexposed," and - cigarette-smoking has been responsible for some 90% of the lung cancers in these "nonexposed. "9 To measure the risk, rather than the relative risk? associated with these other exposures, it is essential to understand the lung-cancer risk associated wi th cigare t te-smoking . DEFINITIONS_ The incidence rate of a disease is the number of cases of the disease _ that are diagnosed during a specified period per specified unit of population.2 The mortality rate of a disease is the number of deaths from the disease during a specified period per specified unit of popuLa- tion. In epidemiologic studies, the unit of time is usually a year and the unit of population is usually 100, 000. All incidence (and mortality) rates quoted here for man use a period of a year, but the unit of popula- tion is 1, unless otherwise stated. If we write the incidence rate without qualification (e. g., I), it is assumed to refer to the standard condition of 1 yr and 1 person. The incidence rate is often affected by many factors, particularly age, and, if the incidence rate is for some particular subgroup, this is stated in referring to the incidence rate, and the symbol, I, for incidence rate is qualified in some way, e.g., I(t) for the incidence rate for a person of age t. For cancers associated with a substantial cure rate or a long time between diagnosis and death, the incidence and mortality rates may be very different. For lung cancer--the major cancer discussed in this chapter--the distinction is not so important, because some 75t of newly diagnosed lung-cancer patients are dead within a year and some 90t within 3 yr. The lifetime risk of a disease is the probability of being diagnosed as having the disease by age 70 (a "lifetime") in the absence of other causes of death. This measure has been found particularly useful in comparing human data and experimental-animal data and forms the basis of current methods of extrapolating animal data to man.3 The lifetime risk is virtually identical with the cumulative incidence rate (to age 70) used by the International Agency for Research on Cancer.3'39 CIGARETTE-SMOKING AS A SOURCE OF PAR EXPOSURE Much of what has been learned about the quantitative relationship between cigarette-smoking and lung cancer over the last 30 yr may be summarized by the statement, "The excess lung-cancer incidence of a smoker, compared with a nonsmoker, is proportional to the number of cigarettes smoked per day and to the duration of smoking raised to the C-2
power 4.5~8,40 If we write the excess incidence--or single-cause incidence 4--of a smoker aged t years who started smoking at age w years and who smokes c cigarettes per day as Ic~t,w), that statement may be expressed mathematically as Ic~t,w) = scat - w)4 5, (1) where the constant a is approximately 1.0 x 10 11 for U.K. smokers.8 For U.S. smokers the constant a must be decreased by 25-50%.12,13,19 33 The reasons for this include the use of different tobaccos in the two countries and the mode of cigarette-smoking--in particular, British smokers tend to smoke their cigarettes down to a considerably shorter butt.7343 Similar reasons probably explain the exis fence of a range. We may express the lung-cancer risks from cigarettes in the usual risk-assessment terms27 of "lifetime risk" by using Equation 1. "Lifetime" is taken as 70 yr, and exposure is taken as starting at birth. If exposure is to c cigarettes per day, Equation 1 shows that the lung-cancer rate at age t will be Ic~t,O' = act4 5. (2) The lifetime risk (cumulative incidence) can be shown to be CIC(T) = 1 - expt-ac(705 5/5.5~. - (3) The lifetime lung-cancer risk associated with one U.K. cigarette per day is 2,524 per 100,000, or 2.527. The lung-cancer risks associated with smoking depend strongly on age at which one started to smoke, i.~., on duration of exposure (see Figure C-1. The increase in lung-cancer incidence rate of a smoker at age 60 who started to smoke at age 20 is proportional to 404 5; if he Sad started at age 15, the extra rate would be proportional to 45 . Starting to smoke 5 Or earlier has thus increased the extra lung-cancer rate by 70% [~45/40) 5], or roughly 14Z for each year. To make valid comparisons between groups of persons exposed to different concentra- tions of PAH-containing mixtures (e.g., different occupational groups), we must therefore know their comparative smoking habits, not only in terms of number of cigarettes smoked per day, but also in terms of age at starting to smoke. For a smoker of c cigarettes/d starting at age w and stopping at age s, the extra lung-cancer incidence rate at age t is Ic s~t,w) = acts - w)4 5. (4) Equation 4 states that the lung-cancer incidence rate associated with cigarette-smokin§ y~m4~ns constant at the value it had reached when smoking stopped. ~ ~ If a person aged 60 who has smoked 30 cigarettes/d from age 20 to 40 (30 pack-yr in total) is compared with a C-3
person at the same age (60) who has smoked 15 cigarettes/d from age 20 to 60 (also 30 pack-yr in total), calculations using Equation 4 show that the latter person will have more than 11 times the lung-cancer incidence rate of the fonder. Thus, to understand quantitatively the effect of exposure to a PAH-containing mixture, one must know not only the total cumulative exposure, but also the time during which it is accumulated. Hoffmann et al.17 pointed out that the major carcinogenic activity of cigarette smoke. resides in the particulate phase (the tar) and that there is good experimental evidence that cigarettes with lower tar yields are less tumorigenic to both hamster larynx and mouse skin. Lower-tar cigarettes have also been shown to be less tumorigenic to man in all epidemiologic studies that have investigated this question. Case-control studies have found that people who smoke filter-tip cigarettes (in effect, lower-tar cigarettes) have lower lung-cancer incidence rates than smokers of plain cigarettes at the same frequency,l,44~45 and Hammond et al.14 found, in the American Cancer Society (ACS) cohort study, that persons smoking low-tar cigarettes had lower risk of lung cancer than smokers of high-tar cigarettes (matched for numbers of cigarettes smoked per day). Table C-1 shows the results of the ACS study: the lung-cancer mortality ratios are clearly not decreased in men in proportion to tar content, but they are nearly so in women. The Latter finding suggests that the added lung-cancer risk is close to being simply proportional to tar content and that the failure to find a proportional reduction in men arises from the male smokers' having switched from high-tar to low-tar cigarettes. As Hammond et al. 1 stated: "Cigarettes with reduced tar and nicotine were not introduced until the mid 1950' s. a . . Almost all of the male cigarette smokers and the great majority of the female cigarette smokers in our study began smoking cigarettes tong before that date. Therefore the subjects classified here as low [tar] cigarette smokers were, with few exceptions, persons who smoked high ~ tar] or medium [tar] cigarettes for many years and then switched to low [tar] cigarettes." These results substantiate the linear dose-response assumption of Equation 1. EXPOSURES TO OTHER SOURCES OF PAM-CONTAINING MIXTURES Large-scale studies of benzo[a]pyrene in the air of the United States were conducted between 1958 and 1959 by Sawicki et al.3 The range of3 BaP concentrations in urban air was from less than 1 to around 60 ng/m and the median was roughly 6 ng/m . In contrast BaP concentrations in nonurban air were almost always less than 1 ng/m3, with a median of 0.4 ng/m . BaP concentrations have since decreased by 1969, the median BaP concentration in urban air was less than 2 ng/m .2 However3 some cities were still experiencing average annual BaP concentrations of nearly 10 ng/m3. BaP is not a perfect indicator of either PAR in the air or its carcinogenicity,35 and it accounts for a much Smaller fraction of the carcinogenicity of cigarettes than of air.il,4 It should be emphasized C-4