Donald E. Watson




Bio-Medical Division


Comparison of the costs due to environmental pollution caused by fossil-fueled and nuclear power sources is complicated by the small overlap of comparable effects. This report examines the potential carcinogenic activity of pollutants emitted by fossil-fueled sources By using the example of cigarette smoking, it is predicted that carcinogenic, and more importantly, co-carcinogenic substances emitted by fossil-fueled sources are present in ambient air in sufficient amounts to present a risk of carcinogenesis. It is emphasized that the anticipated carcinogenic activity cannot be established by epidemiological studies in human populations, and that research should be directed toward understanding mechanisms of carcinogenesis is with the goal of discovering methods of preventing or curing cancer.


Man-made air pollution has increased substantially in recent decades, adding to the air concentrations of naturally occurring gases and particles as well as providing species of pollutants not normally present in air. The total pollutant exposure of man and other life, plant and animal, results partially from the combustion of fossil fuel which provides the large amounts of electrical power demanded by affluent consumers. Electrical power demands in America are increasing at a rate of about 7% annually; by 1990, current needs will be quadrupled. If all of these power demands were to be supplied by fossil fuel-burning plants, the potential pollution sources would increase commensurately. To avoid a four-fold increase in the total air burden of pollutants, the effectiveness of air pollution control measures for each source must be increased by an average of a factor of four.

Since the expense of controlling source emissions increases roughly exponentially with the effectiveness of the control measures, there is an economically practical limit to the measures that can be applied to ameliorate air pollution. This limit should be determined by the evaluation of all costs of pollution, including effects on human health and economic losses to the agricultural industry. Data for the comparison of these costs for the alternative forms of electrical power production, fossil fuel and nuclear, must be provided. Since the forms of pollution from these sources are different, an optimum ratio of power source types might be established to provide electrical power production with a tolerable total pollution burden.

This brief and highly selective survey is an attempt to describe potential health effects of pollution generated by the combustion of fossil fuels in terms comparable with the effects attributed to ionizing radiation. The comparison is necessarily very limited since there are many well-known costs resulting from fossil fuel pollution which have no counterpart in radiation pollution. For example, current ambient air pollution levels in California are blamed for agricultural economic losses of at least $45 million annually. Sulfur dioxide, nitrogen dioxide, and photochemically produced ozone and peroxyacetyl nitrate (PAN), all resulting from the combustion of fossil fuels, produce the most common lesions in vegetation.2

Detrimental eye and nasal irritation, exacerbation of chronic cough and triggering of asthmatic attacks have all been reported by persons exposed to ambient polluted air.3 Though none of these conditions is in itself life-threatening to a reasonably healthy individual, the magnitude of discomfort and economic loss in the total population is large because of the extensiveness of the problem, particularly in high-density urban areas. These costs of air pollution due to fossil fuel combustion have no analogs in terms of low-dose environmental exposure to ionizing radiation. Therefore, any comparison of harmful levels of pollution from fossil fuel and nuclear energy sources which ignores the economic losses and non-lethal harmful effects of fossil fuel-generated pollution is invalid. Attempts to establish equivalence between economic loss, for example, and cancer rate are entirely subjective; no universally acceptable equivalence between the values of money and life can be established by objective criteria. Perhaps guidelines for acceptable costs in terms of human life can be taken from precedents already set by society. For example, the convenience and comfort of the private automobile are considered acceptable benefits in America despite a cost of about 50,000 lives a year. Nearly 50,000 deaths from lung cancer are reported annually, most of them arising from the putative beneficial use of tobacco. Many more lives are ended prematurely through effects of smoking on the cardiovascular and pulmonary systems. There are many other instances of proven acceptability for events or activities associated with known costs or risks.4

The emphasis of this paper is on the poorly understood effects which may accompany long-term exposure to low doses of several pollutants acting alone or in concert with other agents, particularly regarding the initiation of cancer. Experiments designed to determine the long-term low-dose effects of substances are difficult to execute and complex to analyze. Some of the results described in this report obtain from early steps taken by scientists to provide information about these effects.

Experimental Health Effects of the Air Pollutants

This section deals with representative species of air pollutants emitted by fossil fuel-burning sources. Automobiles are the largest single source of these compounds nationwide, but power plants add significant amounts to the total burden. Emphasis is on a few of the potential synergistic or potentiating carcinogenic effects of some of the common pollutants. These effects may result from tissue injury, inhibition of cellular or molecular repair mechanisms, increased efficiency of distribution of carcinogenic material to susceptible tissue, promotion of photodynamic damage, increased vulnerability of cells to penetration by foreign substances, direct carcinogenic stimuli adding to the carcinogenic activity of radiation, and/or other cocarcinogenic effects. No attempt to distinguish mechanisms is made in this review.

As a frame of reference for evaluating concentrations in the experimental work reviewed here, Table 1 lists the current California air quality standards for carbon monoxide, sulfur dioxide, oxidant, nitrogen dioxide, hydrogen sulfide and particulates.5 The criterion for most of the air quality standards is the threshold for known health the most sensitive individuals. The effects of some pollutant gases can be augmented several times by the presence of particles. This interactive phenomenon should be considered while evaluating the animal experiment results reviewed below.

Table 1. Current air quality standards (from ref. 5)
CO 20 ppm/8 hr.
SO2 0.04 ppm/24 hr,
Particulates 60 micrograms/M3
Oxidant 0.10 ppm/hr
NO2 0.25 ppm/hr,
Organic No standard
H2S 0.03 ppm/hr.
Visibility 10 miles


Ozone comprises about 80-90% of the air pollutants reported as "oxidant." It is a product of photochemical reactions involving oxides of nitrogen, hydrocarbons and airborne particles. The current California standard for oxidant is 0.1 parts per million (ppm) averaged for 1 hour. In the experiments reviewed here, doses range from about this value to a few orders of magnitude higher.

Acute and chronic tissue damage caused by ozone may increase the susceptibility to cancer initiation by carcinogens. Chronic bronchitis, bronchiolitis, emphysema, and pneumonitis were observed in small animals exposed to 1 ppm ozone for 1 year.6 Ozone increases the rate of protein cross-linkage formation in lung tissue, exemplifying a molecular effect which could have substantial influence in carcinogenesis.7 An interesting result of exposure to very low doses of ozone is the increase in susceptibility to infection in mice. The minimum effective dose to produce significant mortality is 0.08 ppm for 3 hours.8 Normally, when mice are given streptococcus group C, fewer than 4% of the bacterial cells are viable after 4 hours. However, when animals are pretreated with 1 ppm or more of ozone, the number of cells at 4 hours exceeds the original number. The magnitude of the effect depends on the dose given. Coffin9 suggests several possible modes of action:

The first possibility seems to be ruled out by the finding that intraperitoneal infection is not affected by ozone; the second, by the observation that killing in situ is the means of eliminating bacteria. Concentrations of ozone are too low to account for the focal lesions. However, macrophages are affected by ozone and NO2. Also, the lysosomal enzyme activity in bronchial washings is reduced by ozone exposure. The enzymes are beta glucuronidase, acid phosphatase and lysozyme. Evidence of an overall ozone-induced reduction in the ability of the lung to adapt to insult is abundant in these studies.

Ozone is responsible for increasing the incidence of lung tumors (adenomas) in tumor-susceptible rats. The incidence of tumors in rats exposed to 1 ppm ozone for 15 months increased from 38% to 85%.10 Since overall mortality was not increased, it was concluded that ozone-induced gross tissue damage was not an intermediary for the increased tumorigenicity.

A possible mode of action of ozone in tumorigenesis is through radiomimetic properties; the following list was compiled by Stokinger and Coffin:11 1) As a free radical, ozone can initiate chains of free radical reactions. 2) Chemical agents which protect against radiation also protect against ozone.12 The most effective of these are SH, S-S compounds and iodine, all free radical scavengers. 3) Ozone and x-rays produce chromosomal abnormalities and the effects are essentially additive.13 4) Both O3 and uv irradiation produce retardation of deoxygenation of oxyhemoglobin in capillaries; the dose-response effect is possibly without a threshold.14 5) O3 may produce protection against acute lethal x-ray doses.15 6) Repeated exposure to O3 leads to premature aging.16 7) The avoidance reaction in mice simulates that effect of ionizing radiation.17


Oxides of nitrogen result from the high temperatures achieved in the efficient burning of fossil fuels. In general, the more efficient the combustion, the lower the emitted hydrocarbons and the higher the nitrogen oxide emission, whether in power plants operating on coal, oil or natural gas or in automobiles. The oxides of nitrogen enter into photochemical reactions producing "smog." The California air quality standard for NO2 is 0.25 ppm for one hour. Experimental doses reviewed below are beyond this standard. Species differences described appear to be too unpredictable to allow application of the quantitative results of the experiments to predict human effects.

As in the case of ozone, the possibilities of delayed major health effects from low doses administered chronically arise from interactive effects.

High concentrations of NO2 (100 ppm) cause a rapid burst of epithelial proliferation in animals, peaking about 24 hours after a 6 hour exposure. The effect, greatest in the major bronchi, returns to normal after about 4 days.18 Even though the acute dose in these experiments is much higher than is likely to be associated with air pollution levels, the cell proliferation response is of significance as a potential accelerator of carcinogenic activity if the response is proportional to the concentrations of NO2 in the low range.

Increased susceptibility to infection is associated with the inhalation of NO2 at low levels. The threshold for lethal enhancement of infectibility by K. pneumoniae in mice is 3.5 ppm for 2 hours.18 Extended exposure of mice to 0.5 ppm of NO2 caused increase susceptibility to infection after 90 days. Mice were more sensitive to this effect of NO2 than either hamsters or squirrel monkeys, but these species also showed increased infectibility with NO2 exposure. These results were attributed to the diminution in the ability of the lung to clear microorganisms. The ability of NO2 (and ozone) to interfere with the complex clearing mechanism of the lung suggests an action capable of contributing to carcinogenesis through multiple mechanical and chemical influences.

As in the case of ozone there is evidence that NO2 may accelerate the rate of tumorigenesis in tumor-susceptible mice, although the data are not statistically convincing. The effect has been studied with long term exposure to 5 ppm and to brief exposures to massive concentrations of NO219,20


The major source of sulfur compounds in the air is the combustion of coal and residue fuel oil. Power plants contribute the bulk of this. Hydrogen sulfide, mercaptans and sulfur dioxide are the most significant sulfur compounds emitted into the atmosphere. This review concentrates on potential health effects of SO2. The California standard for SO2 is 0.04 ppm for 24 hours or 0.5 ppm for one hour.

SO2 is a soluble gas, absorbed soon after inhalation on the membranes of the naso-pharynx unless there is a high concentration of particles capable of adsorbing the gas and delivering it to deeper pulmonary tissues.

In susceptible individuals the Levels of SO2 required to produce fatal complications to pre-existing illness may be quite low. A report on a retrospective epidemiological study in New York City comparing daily death rate with SO2 concentrations on meteorologically similar days showed an increase of about 10-12 deaths/day for concentrations of SO2 of about 0.2-0.4 ppm averaged for 24 hours.21 This study used SO2 as the measure of air pollution, but no conclusion can be drawn as to the role of SO2 in causing the excess deaths, since no other specific pollutant was tested for correlation with death rate. The more general conclusion is that high levels of air pollution, as indicated by SO2 levels, may be a necessary condition, though not necessarily a sufficient condition, to account for the increased death rate observed.

Pathological chances induced by SO2 inhalation include pneumonitis, tracheitis, bronchitis, and ulceration with increased mucosal cellular proliferation. All of these lesions are potential promoters of carcinogenesis.

SO2 and benz(a)pyrene (BaP) given together have a carcinogenic effect greater than BaP alone. Rats and hamsters were given BaP and SO2 by inhalation. The rate of SO2 given was 10 ppm SO2 for 30 hours a week, and the average daily concentration of BaP was 0.104 mg/M.3 The animals exposed to both SO2 and BaP developed cancer at a higher rate than those of control groups.22


A wide variety of elemental species is represented in particles. Health effects of inhaled particles may arise from direct toxicity of the particle itself or from interactive effects among particles and other substances. Properties or combinations involving particles which may determine effects are 1) the degree of adsorption or absorption of a vapor on a particle, 2) the degree of chemical or catalytic interaction of the vapor and particle, 3) the relative desorption rate on the biological surface, and 4) the toxicity of the combination.23 A few examples of experimental models are presented here, illustrating various effects that several particle species exhibit in animal tissues.

The mechanism of transport by particles is illustrated by experiments using mixtures of gases and particles. When nitrogen dioxide is administered with particles of carbon (about 2 micron diameter), focal destructive lesions are produced in the lung which are not produced by inhalation of either carbon or NO2 alone.24 It is pointed out that the amount of NO2 adsorbed on a 2-micron particle is enough to produce a one-normal solution of nitric acid of the volume of a cell.

Airway resistance in guinea pigs is increased by the administration of sulfur dioxide with sodium chloride particles (0.04 micron median diameter). Control animals were given sulfur dioxide alone in concentrations ranging from 2 to 250 ppm. The accentuating effect of the particles was greatest (27:1) in the low ranges of concentration.26

Inhalation of ferric oxide (Fe2O3 particles (1 micron) combined with BaP or the systemic carcinogen, diethylnitrosamine (DEN) accelerates the rate of development of squamous cell carcinoma of the bronchi, trachea, bronchioles and larynx.27 There is a several-fold increase in the cancer rate compared with the rate in animals given either carcinogen alone. In this case, the effect of the particles appears to be due to a local pulmonary co-carcinogenic action.

Particles of zinc ammonium sulfate are irritants to the pulmonary tree if administered alone. Mixed with sulfur dioxide, the response of airway resistance in guinea pigs is more than additive.28 Since zinc ammonium sulfate and zinc sulfate accounted for 58% and 21% of the soluble particulate in the 1948 air pollution episode in Donora, PA,29 these particles may have been major contributors to the health effects in that episode.

The variety of tissue responses to particulates alone or mixed with other pollutants can be enormous. One can appreciate the complexity of the problem by contemplating the plethora of physical and chemical properties characterizing particles composed of the following elements, all detected by neutron activation analysis of sea salt particles,30 and urban air samples:31 Ce, La, Sm, Eu, Tb, Yb, Hf, Sc, Cr, Fe, Co, Cu, Sn, Ga, Se, In, Sb, Cd, Au, and Hg (sea salt particles), and Al, S, Ca, Ti, V, Cu, Na, Mg, Cl, Mn, Br, In, I, K, Zn, As, Ga, Sb, La, Sm, Eu, W, Au, Sc, Cr, Fe, Co, Ni, Se, Ag, Ce, Hg, and Th (urban air particles).


The importance of organic particles lies in the fact that several carcinogenic compounds have been measured in ambient air particles.32,33 The best example of the effects of inhaling low concentrations of organic particles over extended periods is that of smoking cigarettes. Salient features of the smoking effects are reviewed here as they relate to potential dangers of air pollution.

Recently, beagles have been reported to develop squamous cell carcinoma of the bronchi after prolonged periods of inhalation of cigarette smoke.34 This is the first time this disease has been produced convincingly in experimental animals, but the epidemiological evidence associating lung cancer with smoking in humans35 is so convincing that there is little reasonable doubt of causality.

The median particle size of tobacco smoke is about 0.5 micron. Major classes of compounds in the particulate phase are given in Table 2.


Table 2. Major classes of components of particulate phase of tobacco smoke and their general effects on health (ref. 35)
Acids 7.7 - 12.8 Irritant
Glycerol, Glycol, Alcohol 5.3 - 8.3 Poss. irritant
Aldehydes, Ketones 8.5 Some irritation
Aliphatic Hydrocarbons 4.9 Some irritation
Aromatic Hydrocarbons 0.44 Some carcinogenesis
Phenols 1.0 - 3.8 Irritant, Co-carcinogenesis
Water 27 -

Notice that the particulate fraction includes the aromatic hydrocarbons. Carcinogenic hydrocarbons isolated from tobacco smoke are tabulated in Table 3.


Table 3. Carcinogenic polynuclear hydrocarbons in cigarette smoke (Ref. 35)
Carcinogenic Activity
Nanograms per 20 cigarettes
Benz(a)pyrene ++++ 320
Dibenz(a,i)pyrene ++++ 0.4 - 200
Dibenz(a,h)anthracene ++ 80
Benz(c)phenanthrene + ?
Dibenz(a,j)acridine + 54
Dibenz(a,h)acridine + 2
7H-Dibenz(c,g)carbazole + 14
The gas phase components of tobacco smoke are given in Table 4 for reference.
Table 4. Gas phase components of tobacco smoke (Ref. 35)
Conc. ppm
"Safe" ppm/day
Toxicity to Lung
CO 42,000 100 Unknown
CO2 92,000 -- None
Methane, Ethane, etc. 87,000 500 None
Acetylene, Ethylene, etc. 31,000 5,000 None
Formaldehyde 30 5 Irritant
Acetaldehyde 3,200 200 Irritant
Acrolein 150 0.5 Irritant
Methanol 700 -- Irritant
Acetone 1,100 200 Irritant
Methyl Ethyl Ketone 500 2.5 Irritant
Ammonia 300 150 Irritant
NO2 250 5 Irritant
Methyl Nitrite 200 -- Unknown
H2S 40 20 Irritant
HCN 1,600 10 Resp. Enzyme Poison
Methyl Chloride 1,200 100 Unknown
Several compounds isolated from tobacco smoke, including benzo(a)pyrene (BaP) and dibenzo(a,i)pyrene, are very potent inducers of cancer. BaP is the most abundant and one of the most potent of the tobacco smoke carcinogens (Table 3). It has been used as an experimental carcinogen so extensively that there is a large body of literature on its effect alone and in concert with other agents.

Most of the carcinogens in tobacco smoke are not found in the tobacco leaf, but are produced by pyrolysis of other compounds at high combustion temperatures (800-900oC). For example, stigmasterol is pyrolyzed to BaP, and pyridine and nicotine are pyrolyzed to dibenzo(a,j)acridine and dibenzo(a,h)acridine.

It is significant that BaP is also produced by the combustion of coal, fuel oil, and to a lesser degree, natural gas, and by the incineration of waste material.36 Urban air concentrations of BaP and other polynuclear compounds are given in Table 5. Obviously, fossil fuel-burning electrical power sources can be implicated in the total air burden of these substances.

Table 5. Concentrations of some large organic compounds in the average American urban atmosphere. (Ref. 33)
Airborne particulate, nanograms/M3
Airborne particulate, nanograms/day
Benz(a)Pyrene 5.7 17.0
Benz(e)Pyrene 5 150
Benz(f)Quinoline 0.2 6
Benz(h)Quinoline 0.3 9
Benzo(a)Acridine 0.2 6
Benzo(c)Acridine 0.6 18
11H-Indeno(1, 2-b)Quinoline 0.1 3
Dibenz(a,h)Acridine 0.08 2.4
Dibenz(a,j)Acridine 0.04 1.2
Benz(a)Anthracene 4 120
Fluoranthene 4 120
Pyrene 5 150
Perylene 0.7 21
Benz(g,h,i)Perylene 8 240
Anthanthrene 0.26 7.8
Coronene 2 60


The combustion of fossil fuels is responsible for the atmospheric release of numerous substances which have detrimental effects on plant and animal health. Those pollutants which are released in largest amounts, such as nitrogen dioxide, sulfur dioxide, carbon monoxide and particulates, have demonstrable acute and chronic toxicity if administered in sufficiently large doses. Photochemical end products, including ozone and numerous free radicals) are also harmful if copiously inhaled. Local, state and federal air quality standards are designed to limit exposure of the general population to such high doses.

Significant danger of carcinogenesis not adequately covered by the current air quality standards exists from certain pollutants resulting from the combustion of fossil fuels. This speculative statement derives from two basic assumptions: First, that components of tobacco smoke are responsible for initiating the excess number of cases of cancer associated with smoking cigarettes, and second, that polluted air contains compounds which are identical to or similar in action to the carcinogenic and cocarcinogenic components of tobacco smoke. Attempts to demonstrate quantitative similarities between the two forms of inhaled pollution are severely limited by the paucity of information about interactive effects of multiple co-carcinogenic factors present in each case. For example, the carcinogenic activity of total tobacco smoke tar (when painted on mouse skin) is about 40 times that of benz(a)pyrene alone. 35 Details of the mechanism of this augmentation are lacking, but phenol and its derivatives, long-chain fatty acids and esters, and hydroperoxides of unsaturated compounds are suspected as co-factors.

Corresponding co-carcinogenic activities of some of the common air pollutants discussed earlier in this paper cannot be quantitatively compared with those of tobacco smoke.

From the data of Tables 3 and 5, the dose of BaP inhaled by an individual can be calculated for smoking a pack of cigarettes (0.32 microgram/day) and from breathing urban air at a rate of 30 cubic meters/day (0.17 microgram/day). The dose rate for this carcinogen from breathing urban air is of the order of that taken by a light smoker (one-half pack a day). The lung cancer mortality ratio for smokers in this category and others can be seen in Table 6.

Table 6. Lung cancer for male smokers. Mortality ratios (smokers/non-smokers) by age and smoking parameters (Ref. 35)
Current Cigarettes/day 1-9 6.17 3.53 5.32 4.60
10-19 3.90 8.77 9.62 7.48
20-39 9.37 13.82 17.62 13.14
40+ 7.67 17.47 29.84 16.61
Age began smoking 25+ 2.77 3.39 3.38 3.21
22-24 5.83 11.11 12.11 9.72
15-19 8.71 13.06 19.37 12.81
<15 12.80 15.81 16.76 15.10

The latent period for lung cancer manifestation is not well defined, but the median latency appears to be of the order of 20 years. With present knowledge, it is not possible to predict expected rates of cancer induction in a population exposed to polluted urban air by using the smoking statistics, since co-carcinogenic action is a major variable in both cases. Moreover, the value of the total integrated dose of carcinogens and all co-carcinogens to an individual living constantly in a polluted environment for 20 to 30 years cannot be reliably estimated. Therefore, expected rates of cancer induction cannot be reasonably established for predictive models for epidemiological studies involving the multivariate systems of population distribution and air pollution levels over a long exposure period.

Cancer initiation may result from a single event or a combination of events. The probability of this occurrence may be minuscule on a short time scale with low concentrations of reactants, but as exposure times and pollutant concentrations increase, cancer initiation will occur with increasing frequency. If the carcinogenic reaction is irreversible, there will be no threshold below which the population as a whole is not vulnerable.

The problem presented by these possibilities is subtle, due to the very low probabilities of occurrence, but important because of the catastrophic consequence to an individual if cancer is initiated. The probability of occurrence of a carcinogenic event is enhanced by increasing the abundance of environmental carcinogens or co-carcinogens. Cocarcinogenic substances account for increasing the susceptibility of biological tissues or molecules by mechanisms which are poorly understood at present.

One plausible mechanism of co-carcinogenic action is the stimulation of cell division or repair processes by irritants or general cellular toxins in the presence of the primary carcinogen. Clearly, the complexity of air pollution study is increased substantially when it is recognized that trace amounts of certain species of particles may amplify by many times the potential carcinogenic activity of polluted air. Acceptable air quality standards for specific pollutants may be set only when enough is known about their modes of action and interaction.

The findings of the advisory committee to the surgeon general on smoking and health35 indicate the need for further study on the molecular and cellular effects of chronic exposure to environmental pollutants. The conservative position for health scientists is to assume that current environmental exposure to carcinogens will account for future cases of cancer. The environmental factors in cancer induction, whether man-made or natural, are unquestionably responsible for some degree of the disease. It is important at this stage to determine causality rather than to wait for further evidence of effect.

The AEC-funded research into the effects of radionuclides in the biosphere must be supplemented by research into the analogous environmental results of fossil fuel combustion. It is unrealistic to hope that the coal or petroleum industries would adequately fund research which might document harmful effects of their products. The resources of the national laboratories should be directed toward a programmatic resolution of these complex and far-reaching problems. Rapidly increasing demands for electrical power generation must be met with responsible planning which takes into consideration all long-range environmental consequences.

The need for information detailing the costs and hazards of both nuclear and fossil fuel power generation is critical now.


1 Yocom, J. E. and R. 0. McCaldin, "Effects of Air Pollution on Materials and the Economy," in Air Pollution, Vol. I, Ed. A. C. Stern (1968), p. 617.

2. Brandt, C. S. and W. W. Heck, "Effects of Air Pollutants on Vegetation," in Air Pollution, Vol. I, Ed. A. C. Stern (1968), p. 401.

3. Goldsmith, J. R., "Effects of Air Pollution on Human Health," in Air Pollution, Vol. I, Ed. A. C. Stern (1968), p. 547.

4. Starr, C., "Social Benefits versus Technological Risk," Science 165, 1232 (1969).

5. California Board of Air Resources. Air Pollution Control in California, Annual Report (1969).

6. Stokinger, H. E., "Evaluation of the Hazards of Ozone and Oxides of Nitrogen: Factors Modifying Toxicity," A.M.A. Arch. Ind. Health 15, 181 (1957).

7. Buell, G. C., Y., Tokiwa, and P. K. Mueller, "Potential Crosslinking Agents in Lung Tissue. Formation and Isolation After In Vivo Exposure in Ozone," Arch. Environ. Health 10, 213 (1965).

8. Coffin, D. L., E. J. Blommer, D. E. Gardner, and R. Holtzman, Third Annual Conference on Atmospheric Contamination in Confined Spaces (Dayton, Ohio, May, 1967).

9. Coffin, D. L., "Study of the Mechanisms of the Alteration of Susceptibility to Infection Conferred by Oxidant Air Pollutants," in Inhalation Carcinogenesis, (A.E.C. No. Conf-691001, April, 1970), p. 259.

10. Stokinger, H. E., in Air Pollution, Vol. I, Ed. A. C. Stern (1968).

11. Stokinger, H. E. and D. L. Coffin, "Biologic Effects of Air Pollutants," in Air Pollution, Vol. I , Ed. A. C. Stern, (1968), p. 445.

12. Fairchild, E. J., S. D. Murphy, and H. E. Stokinger, "Protection by Sulfur Compounds Against the Air Pollutants Ozone and Nitrogen Dioxide," Science 130, 861 (1959).

13. Fetner, R. H., Res. Engr. (Georgia Tech. Eng. Expt. Sta.) 11 (1960).

14. Brinkman, R. and H. B. Lamberts, "Direct Registration of an Instantaneous X-ray Effect in Rats and Man," Nature 181, 504 (1958).

15. Hattori, K., N. Kato, M. Kinoshita, S. Kimoshita, and T. Sunada, "Possible Effect of Ozone in Mice Against Whole-Body X-Irradiation," Nature 196, 1220 (1963).

16. Scheel, L. D., 0. J. Dobrogorski, J. T. Mountain, J. L. Svirbely, and H. E. Stokinger, "Physiologic, Biochemical, Immunologic and Pathologic Changes Following Ozone Exposure," J. Appl. Physiol. 14, 67 (1959).

17. Peterson, D. C. and H. L. Andrews, "The Role of Ozone in Radiation Avoidance in the Mouse," Radiat. Res. 19, 331 (1963).

18. Ehrlich, R., M. C. Henry, and J. Fenters, "Influence of Nitrogen Dioxide on Resistance to Respiratory Infections," in Inhalation Carcinogenesis (A.E.C. No. Conf-691001, April 1970).

19. Wagner, W. D., B. R. Duncan, P. G Wright, and H. E. Stokinger, "Experimental Study of Threshold Limit of NO2 " Arch. Environ. Health 10, 455 (1965).

20. Schuman, L., Cited in Air Pollution, Vol. I, Ed. A. C. Stern (1968), by H.E. Stokinger and D. L. Coffin, p. 473.

21. Glasser, M. and L. Greenburg, Air Pollution, Mortality, and Weather; New York City, 1960-1964. In press (1969).

22. Laskin, S., M. Kuschner, and R. T. Drew, "Studies in Pulmonary Carcinogenesis," in Inhalation Carcinogenesis (A.E.C. Conf-691001, April 1970).

23. Goetz, A., "Study of the Properties of Aerosols, with Particular Reference to the Nature of the Air-Particle Interface," (U. S. Public Health Service, SAPH-69557, Part B, 1958).

24. Boren, H. G., "Carbon as a Carrier Mechanism for Irritant Gases," Arch. Environ. Health 8, 127 (1964).

25. Crocker, T. T., cited by H. G. Boren, Arch. Environ. Health 8, 130 (1964).

26. Amdur, M. O., "'The Physiological Response of Guinea Pigs to Atmospheric Pollutants," Int. J. Air Pollut. 1, 170 (1959).

27. Montesano, R., U. Saffiotti, and P. Shubik, "The Role of Topical and Systematic Factors in Experimental Respiratory Carcinogenesis," in Inhalation Carcinogenesis (A.E.C. No. Conf-691001, April 1970).

28. Amdur, M. O. and M. Corn, "The Irritant Potency of Zinc Ammonium Sulfate of Different Particle Sizes," Am. Ind. Hyg. Ass. J. 24, 326 (1963).

29. Hemeon, W. C. L., "Estimation of Health Hazards from Air Pollution," A.M.A. Arch. Ind. Health 11, 397 (1955).

30. Dudey, N. D., L. E. Ross, V. E. Noshkin, "Application of Activation Analysis and Ge(Li) Detection Techniques for the Determination of Stable Elements in Marine Aerosols," in Proceedings 1968 International Conference "Modern Trends in Activation Analysis" (Gaithersburg, Maryland, 1969) p. 55.

31. Dams, R., J. A. Robbins, K. A. Rahn, and J. W. Winchester, "Sampling Aerosols for Nondestructive Neutron Activation Analysis," (in press) - submitted to Anal. Chem. (1970).

32. Sawicki, E., T. R. Hauser, W. C. Elbert, F. T. Fox, and J. E. Meeker, "Polynuclear Aromatic Hydrocarbon Composition of the Atmosphere in Some Large American Cities," Am. Ind. Hyg. Ass. J. 23, 137 (1962).

33. Sawicki, E., S. P. McPherson, T. W. Stanley, J. Meeker, and W. C. Elbert, "Quantitative Composition of the Urban Atmosphere in Terms of Polynuclear AZA Hetrocyclic Compounds and Aliphatic and Polynuclear Aromatic Hydrocarbons," Int. J. Air Water Pollut. 9, 515 (1965).

34. Hammond, E. C. and O. Auerbach, cited in the New York Times, Feb. 6, 1970 and Feb. 8, 1970.

35. U. S. Public Health Service. Smoking and Health. Report of the Advisory Committee to the Surgeon General of the Public Health Service (U. S. Dept. of Health, Education and Welfare, Washington, D. C., P. H. S. Publication No. 1103, 1964).

36. Hangebrauck, R. P., D. J. Von Lehmden, and J. E. Meeker, "Emissions of Polynuclear Hydrocarbons and other Pollutants from Heat-Generation and Incineration Processes," J. Air Pollut. Control Ass. 14, 267 (1964).