as is indicated by samples from atmospheric monitoring stations. These are not samples taken within 2 feet of an exhaust pipe or in the immediate vicinity of traffic, but samples from stations reasonable and realistically located.

Finally, the question on tolerance is also unsettled, in my opinion. An interesting study by Gillick in England appeared to indicate that tolerance could develop.

I should like now to emphasize a point which has plagued much of our discussion today. And that is the importance of time-the time of exposure, the time of occurrence of response to exposure, the duration of exposure, the rate of change of exposure-all of these should be incorporated into experimental systems and into research programs. For example, I cite the work that Dr. Spicer reported. He talked about peak levels and mean levels. He did not talk about the rate of change of levels, and this is a parameter, so to speak, and a way of looking at air pollution exposures which may be a great deal more relevant than we have thought.

Also there was very little discussion on what is very important, to my way of thinking, the time relationship from all available data with respect to possible carcinogenic exposure. From all of the data that we already have, there is reason to believe that 10 to 30 years may elapse after exposures which have an effect on carcinogenesis before the carcinogenic result occurs. I want to make this plea to all who are engaged in research and to all who interpret research: pay a little more attention to the complexities introduced into research and its interpretation because of the factor of time.

Grace Talbott (for the San Francisco Bay Area Air Pollution Control District). Persons with chronic respiratory disease, especially those with emphysema, chronic breathlessness, chronic bronchitis, allergic asthma, or any other type of asthma, are victims of increased discomfort and disability when exposed to air pollutants. More specific research on pollutant irritants and their effects on the health of people who have chronic respiratory diseases should be developed by the combined effort of those separately involved in such research, and the results disseminated by the U.S. Public Health Service.

Francis Silver. Which straw really broke the camel's back is a foolish kind of question. I'm an engineer. Engineers deal with multiple causal conditions as a matter of routine, and even the thought of a single cause seems a little foolish. We quan

tify each of our various causes and stresses and summate them. I have been attempting for several years to do as the old-fashioned family physician or general practitioner used to, go into the home and make suggestions such as Dr. Randolph spoke of. And because of the increasing complexity of medicine he has to a considerable extent withdrawn from the home. In addition, the complexity of the home, engineeringwise and chemistrywise, has increased to such an extent that he is not able to cope with it. I have been attempting to go into the home as a graduate gas engineer and then study some of these environmental things and make recommendations to the housewife along the line that Dr. Randolph has made. I quite agree with him that these chemical conditions in the home are very grave and in many cases I think it would be more serious than either general air pollution in the outside air or smoking.

Now I shall read the question I sent in. Because of the nature of their work, gas and mining engineers have usually received very careful training in the direct evaluation of health effects in humans exposed to toxic gases. One of the basic principles that gas and mining experience over the centuries demonstrates is that effects such as can be observed in oneself and others in many of our city streets should form a basis for corrective action if serious chronic effects are to be avoided. Why is so little attention paid in air pollution work to such basic gas engineering and mining engineering principles, which have been established over centuries of experience? Present delaying practices in air pollution control seem to a gas engineer to be quite out of order. They seem similar to asking a structural engineer to use a safety factor so low as to insure that his bridge will collapse.

Sterner. Yes. Mr. Silver, I believe that, with respect to the more acute conditions, the principles of gas and mining engineering and of industrial hygiene can be applied and will be applied in the reasonably near future in establishing levels which are associated with overt evidence of injury. However, when we drop to a lower level, when we're concerned about chronic effects, the problem becomes infinitely more complicated. Look at the experience again in occupational exposures. Where a disease condition develops only after a long interval from the time of exposure, and from very small amounts of the etiologic agent, as for example the bladder tumors associated with betanaphthylamine, the problem of establishing with any reasonable

degree of certainty the lowest levels which were effective in producing the disease in human cases was almost impossible. The corollary problem of determining safe or permissible levels was even more difficult, so that many of the companies gave up the manufacture of betanaphthylamine, even though it was a very useful industrial chemical.

I should think then that, certainly with respect to the episodes of smog and of pollution which are associated with an increased incidence of clearly identifiable adverse health effects, we shall be able to come up and I hope, in the not too distant future with levels which will be realistic. I'm much more dubious that we shall get these levels with a reasonable degree of reliability for the longterm effects. Now this doesn't mean that I'm not in favor of doing something about air pollution anyway, because I think there is enough evidence to do something about it. But I think we shouldn't delude ourselves with the idea that we're doing this because there is clear and convincing evidence of serious and substantial injury to health. I think we have ample reasons to do it on other grounds. E. J. Cassell. Dr. Spicer, both the British and the Americans smoke, but the British usually open doors and don't have central heating. And, although it's a little hot in our rooms, I'd rather have it that way than to have the doors open. I didn't really mean to be the last of a dozen people to comment on complexity, but perhaps the fact that many of us have so commented makes it appear that this in itself is as much a problem as the nature of the effects of air pollution. Dr. Spicer's paper and the work of Dr. Gocke at Seton Hall and our own work and that of many other people have shown that we can amass huge quantities of data which appear to relate to the people we're studying but which we're unable to relate well to the air pollution information which we also have gotten. I think it's harder to analyze data when you neither know what the cause is nor what effect you're looking for.

I also think that we have to begin to recognize for a change that the methods which we use to gather data may not be inadequate after all, that the data which we've gotten may be really quite good, that the pulmonary function data with all its imperfections may be really quite good, that illness data may be quite good, and that air pollution data with all its imperfections may be quite good, but that the methods for analysis are really totally inadequate. And that it's time we began to come together to discuss exclusively the best method of

analysis of the reams of data which we've all gathered; I am quite sure that some studies have failed completely because of inability to handle the information obtained.

I also believe that Dr. Goldsmith's point about time is exactly pertinent to this, because that becomes one other factor which is impossible to incorporate under most methods of analysis. For example, if Dr. Kotin's relationship between influenza and carcinoma of the lung in mice were also true in human beings, then we should have about now the age-case ratio of cancer of the lung which we find in smokers who got influenza in 1918 to 1920, and about now we should be very distressed about it and start to do something about it. And pretty soon, the case rate would fall off as the susceptibles began to be used up and we would call this the effect of what we've done to the atmosphere, when in actuality it was only the effect of beginning to use up this particular set of susceptibles.

So, in summary, I think that we really have to begin to work consciously on much more sophisticated methods of analysis, methods that are the equal of the data we've obtained.

Spicer. Thank you, Dr. Cassell. Now, 15 seconds per panel member to answer one key question: Do you believe that air pollution is one of the causes of chronic respiratory disease? Dr. Goldsmith?

Goldsmith. Yes.

Spicer. Dr. Sterner?

Sterner. Yes; but if I modified it as I should like to, I'm afraid that I'd take more than my 15 seconds. Spicer. Dr. Dohan?

Dohan. We discussed this last night and the question is what you mean by cause. But in a general way I would say yes.

Spicer. Dr. Kotin?

Kotin. I have nothing that would improve on Dr. Dohan's answer.

Spicer. Dr. Wynder?

Wynder. I would say that in special settings air pollution, as I indicated, can contribute to respiratory disease.

Spicer. Dr. Landsberg?
Landsberg. I pass.
Spicer. Dr. Mitchell?
Mitchell. I pass.
Spicer. Dr. Askew?

Askew. Both the individual and the community contribution of air pollution to chronic disease, I think, will grow and be self-evident in time.

CONCLUDING REMARKS

ROGER S. MITCHELL

The Webb Institute for Medical Research Denver, Colo.

My personal interest in air pollution and its relation to human health has centered around the apparent part it plays in the pathogenesis of the chronic bronchitis-emphysema syndrome. Morphologic studies of 126 lungs, conducted at the Webb-Waring Institute for Medical Research in Denver, have shown that approximately 10 percent of our patients dying with clinical evidence of chronic obstructive airway disease and its complications have full-blown chronic bronchitis as the major cause of the clinical picture and death; the other 90 percent have sufficient alveolar wall destruction (i.e., emphysema) to explain their symptoms and death. In London, on the other hand, the findings are almost exactly the reverse (L. Reid): approximately 90 percent of a similar series of cases have enough chronic bronchitis but apparently not enough emphysema to kill them, and only about 10 percent have apparently died directly as a result of their emphysema.

This striking discrepancy can be explained, I believe, by the differences in the quality of the air in Colorado and in London. In London air, SO has been quite prevalent, due to the English custom of burning soft coal in each family hearth instead. of heating homes centrally. SO, is seldom found in U.S. air pollution, except in those areas where sulfur-containing coal is still burned in large quantities without proper safeguards.

SO2, even in very low concentrations, temporarily paralyzes the cilia and thus probably increases the retention of all other pollutants (including cigarette smoke). Other factors, such as colder temperatures and higher humidity, probably play an additional part in explaining the difference in the autopsy findings between the United States and the United Kingdom.

I have also been interested in the relationship between the quantity of black pigment deposited and the amount of lung damage found in human lungs. Our studies have shown black pigment to be present in the lungs of some individuals as early as age 11, and in all individuals we have studied so far past the age of 25. The quantity of pigment varies roughly with the amount of alveolar wall damage, and its microscopic location is usually, but not always, immediately adjacent to the areas of damage. The quantity of pigment and damage could also be shown to have a rough relationship to the quantity of cigarette smoking; data on exposure to other air pollutants were too uncertain to be evaluated satisfactorily.

We have heard during this conference of a number of ways in which polluted air can cause human disease. It is by no means a simple picture. Multiple factors are almost certainly at work.

The most common pollutants are the relatively large particles-visible as smoke and usually malodorous; they are essentially without biologic activity. They can do harm, however, by overburdening the self-cleansing mechanisms of the tracheobronchial tree and thus tending to delay the excretion and prolong the adverse effects of pathogenic particles.

Next, we have the very tiny and usually biologically active particles-invisible and often without odor. SO, and NO2 are the classic examples of this group. They can damage tissue acutely.

From experimental evidence not reviewed today (H. Boren), we know that the heavier, inert, and largely carbonaceous pollutants can probably also act as carriers of the smaller biologically active particles and thus may prolong and increase their adverse effects.

Then we have some extremely small, highly reactive substances, the so-called free radicals, of which ozone and peroxyacetyl nitrate (PAN) are examples. Such substances have been shown to interfere with the essential enzyme processes of plants in remarkably tiny concentrations. Their effects on humans, however, other than irritative, are still poorly understood.

Finally, we also have noxious gases such as carbon monoxide. The prolonged inhalation of sublethal doses of carbon monoxide may not have a significant chronic adverse effect by itself, but in the presence of other noxious inhalants and especially for patients with chronic pulmonary or cardiac disease, the added effect of retained carbon monoxide may become significant. Lung tissue cells may well be damaged or killed by such combinations of noxious agents, when singly their effects are of no serious consequence.

It has been stressed that pollutants may be irritant, poisonous, or pathogenic; pathogenicity, of course, includes carcinogenic activity.

The importance of associated infection has been properly stressed. Polluted air has been found to contain an increased number of pathogenic microorganisms, and respiratory tissues, which are irritated and/or damaged by the inhalation of pollutants, are demonstrably more susceptible to both viral and bacterial infections. Chronic respiratory infection in turn makes the individual more susceptible to later exposure to the air pollutants.

In any discussion of air pollution and health, the problem of individual air pollution-tobacco smoking should be included, as we have consistently heard this morning. Tobacco smoke contains numerous irritants, numerous carcinogens, and a poison, carbon monoxide. It also contains unidentified substances which, like SO2, temporarily inhibit ciliary action. It is thus quite obvious that community and individual air pollution are effective partners in crime.

Armed with our present knowledge, I think we can construct a reasonably sound hypothesis regarding the pathogenesis of our two major chronic

respiratory diseases, bronchogenic carcinoma and the chronic bronchitis-emphysema syndrome. Of course, the causes of these two diseases are not yet fully understood, but we can be reasonably certain that more than one factor is involved in each disease.

First, perhaps, the self-cleansing mechanism is inhibited by SO2, or by something in tobacco smoke and/or by the need to remove large quantities of inhaled inert carbonaceous material. Chronic irritation is caused by consequent excessive retention of various biologically active pollutants. Chronic irritation is frequently followed by chronic and/or recurrent infection. The respiratory mucosa is damaged, and this leads to deeper penetration of pathogenic substances, possibly prolonged in action by inert carbonaceous carriers.

In addition to this sequence of events, we are also almost surely dealing with differences in individual susceptibility as well as differences in degree of exposure to the pollutants. Individual differences probably include differences in susceptibility to infection, differences in hepatic function (as we heard today), or even inherent differences in the individual's self-cleansing mechanisms.

I have omitted reference to a host of other effects of air pollution, such as the influence upon the bronchospastic disorders, the influence upon the weather, and the influence upon the psychology of the population of a city.

In conclusion, it is my opinion that polluted air (from both community and individual sources) is one of the causes of at least two major respiratory diseases, bronchogenic carcinoma and the chronic bronchitis-emphysema syndrome. The complex mechanisms by which this happens are by no means clear. In spite of the inadequacy of our present knowledge, the available facts are, in my opinion, quite sufficient to justify an all-out campaign to control or eliminate air pollution throughout the United States. And last but not least, one distinctly possible and happy dividend of the abatement of community air pollution would be a probable decrease in the hazards of cigarette smoking.