12.2 Measuring cigarette smoke constituents

Cigarette smoke is a complex mixture of thousands of chemicals. It has long been understood as having particulate and vapour phases.1 The particulate phase can be understood technically as those smoke particles that are large enough to be captured on the filter of a smoking machine (which samples mainstream smoke and is used to measure the 'tar', nicotine and carbon monoxide 'yields' (or 'emissions') of cigarettes). The vapour phase consists of smaller particles and gases that pass through a smoking machine filter (and generally are also not captured by cigarette filters). 'Tar' in its technical sense is the collected mass of larger smoke particles captured on the filter when a cigarette is machine smoked. However 'tar' also refers non-technically to the brown stains seen on the end of cigarette filters and on smokers' fingers.

Measuring the 'yield' of 'tar' captured by a smoking machine filter was a long standing means of comparing the cigarette smoke of different brands – both by the tobacco industry and by public health authorities. While measuring 'tar yields' was a practice developed by the tobacco industry for 'quality control' purposes, it was also long assumed to be a useful means for comparing the relative harmfulness of different brands. In the 1950s it was found that when 'tar' dissolved in acetone was painted on mouse skin, tumours developed. Further, there was a dose-response relationship between the amount of 'tar' to which mice were exposed and the frequency with which tumours developed.1, 2 It appeared to follow logically that cigarettes yielding less 'tar' would be less harmful. Accordingly, it was believed that setting regulatory limits on 'tar' yields and/ or encouraging smokers to switch to brands with lower 'tar' yields would produce relative health benefits for those smokers who were unwilling or unable to quit.

Nicotine yields have also long been used for making comparisons between brands. Public health authorities first became interested in measuring the nicotine yields of cigarettes on the grounds that nicotine is the primary addictive ingredient of cigarette smoke and that reducing smokers' nicotine exposures would lessen their addictions and facilitate quitting.

Carbon monoxide yields were the third measure to join tar and nicotine yields. This is because carbon monoxide is likely to be a particularly important cause of the cardiovascular damage caused by smoking.

The tar, nicotine and carbon monoxide figures that were printed on the side of Australian cigarette packs until March 2006 are derived from the standard ISO (International Standards Organization) cigarette yield test, in which a cigarette is machine smoked with a 35ml puff of 2 seconds duration, once per minute, until it has been smoked down to a 30mm butt. All cigarettes are smoked in the same manner for this test, regardless of whether they are 'full strength', 'light' or 'ultra-light.'

Since the 1990s, a number of other smoking machine test protocols have been introduced. For present purposes, the most important of these is the ISO Intensive Condition test (also frequently referred to as the Canadian Intensive Condition test), in which the machine takes a 55ml puff of 2 seconds duration, once every 30 seconds (and any perforations in the side of the filter–known as filter ventilation–are taped over).

Other more recent developments in measuring smoke constituents have included smoking machines which quantify sidestream smoke yields as well as mainstream smoke yields and testing for a much larger number of specific smoke constituents than 'tar', nicotine and carbon monoxide. Where both mainstream and sidestream smoke measurements and smoke constituents other than 'tar', nicotine and carbon monoxide are involved, it has become more usual to refer to 'emissions' than 'yields' but the two terms remain largely interchangeable.

There were a number of assumptions underlying the presentation of standard ISO tar, nicotine and carbon monoxide yield figures as useful risk information for either regulators or consumers. One assumption is that 'tar' does not vary significantly in composition. Thus, a milligram of tar from one cigarette will contain roughly the same amounts of carcinogens and cardiovascular/ respiratory toxicants as a milligram of tar from any other cigarette. Given that cigarette smoke contains hundreds or even thousands of carcinogens and cardiovascular/ respiratory toxicants, it is plausible that variation in concentrations of specific harmful substances will occur against a background of similarity and the assumption is legitimate for practical purposes. However, more recent thinking on the possibilities for regulating the harmfulness of cigarettes has questioned this assumption.3,4 There is evidence that concentrations in smoke of certain known carcinogens vary substantially between brands. It may follow then, that there are possibilities for reducing the harmfulness of cigarettes by setting limits on the concentrations of particular harmful smoke constituents that are known to be highly variable.

However, the most serious problem with presenting standard ISO yield figures as a means for comparing the relative harmfulness of cigarettes lies with the assumption that individual smokers will always take the same volume of smoke from any cigarette. In fact, the constant for the vast majority of smokers is their target nicotine intake and not the volume of smoke they take from each cigarette.5 Addicted smokers have a target nicotine intake from each cigarette, in order to receive rewarding sensations. The majority of addicted smokers appear to require somewhere between 0.9mg and 1.4mg of nicotine from each cigarette for it to be satisfying.6 Smokers unconsciously change smoking parameters such as puff size and time taken between puffs when they change brands, in order to achieve their target nicotine intakes.7,8 These changes in smoking behaviour are known as compensatory smoking. Whether one looks at individual smokers 'down-switching' to brands with lower tar, nicotine and carbon monoxide yields or looks at entire populations of smokers, one finds that, as standard ISO tar and nicotine yields decrease, parameters like puff size and total number of puffs taken per cigarette increase. The recognition of compensatory smoking set in train the regulatory changes that have seen the end of tar, nicotine and carbon monoxide yield figures being printed on the packs of Australian cigarettes, along with the banning of 'light' and 'mild' brand variety descriptors.

While standard ISO tar and nicotine yields do not provide a useful guide to smokers' intakes, they do provide a rough measure of the effort required for any particular smoker to gain her/his target nicotine dose 8 Below certain yield levels, many smokers will find that the effort required to gain their target nicotine dose has become excessive and cigarettes in these yield ranges will no longer be acceptable to them. Accordingly, many heavily addicted smokers are unable to successfully 'down-switch.' However, the fact that certain groups of smokers will not accept brands that have tar and nicotine yields below certain levels is an entirely separate matter to smokers being able to reduce their intakes by switching to brands with lower tar, nicotine and carbon monoxide yields. Where smokers have been able to 'down-switch' successfully, they almost certainly have not reduced their intakes of nicotine or other harmful smoke constituents. However, there is evidence that a substantial proportion of Australian smokers do not understand compensatory smoking and continue to believe that the cigarette brand varieties which were previously labelled as having lower tar, nicotine and carbon monoxide yields are less harmful.9 The likely reason for the persistence of this belief is that smokers can easily tell that when they smoke brands that previously were labelled as having low tar, nicotine and carbon monoxide yields, the smoke tastes weaker and is less harsh and irritating. However, they cannot easily tell that they are responding to more dilute smoke by taking in larger volumes of it.7 In other words, cigarettes which are manufactured to produce dilute smoke but to facilitate smokers gaining their target nicotine doses provide a compelling illusion of reduced intakes that does not appear to require the support of misleading tar, nicotine and carbon monoxide figures.

Relevant news and research

For recent news items and research on this topic, click here. ( Last updated October 2021)


1. Wynder E and Hoffmann D. Tobacco and tobacco smoke: studies in experimental carcinogenesis. New York: Academic Press, 1967.

2. Hoffmann D, Djordjevic, MV and Brunnemann, KD. Changes in cigarette design and composition over time and how they influence the yields of smoke constituents., In: The FTC Cigarette Test Method for Determining Tar, Nicotine, and Carbon Monoxide Yields of U.S. Cigarettes. Smoking and Tobacco Control Monograph 7. Bethesda, MD: U.S. Department of Health and Human Services, Public Health Service, National Institutes of Health., 1996. 15-37.

3. Gray N. The modern cigarette, an unregulated disaster. Medical Journal Australia 2007;187(9):502-3. Available from: http://www.ncbi.nlm.nih.gov/pubmed/17979612

4. Scientific Advisory Committee on Tobacco. The scientific basis of tobacco product regulation: report of a WHO study group. Technical Report Series, no. 945. Geneva: World Health Organization, 2007.

5. Benowitz N. Compensatory smoking of low yield cigarettes. In: Risks associated with smoking cigarettes with low machine-measured yields of tar and nicotine. Bethesda, MD: U.S. Dept. of Health and Human Services, Public Health Service, National Institutes of Health, National Cancer Institute, 2001. 39-63.

6. Jarvis M, Boreham R, Primatesta P, Feyerabend C and Bryant A. Nicotine yield from machine-smoked cigarettes and nicotine intakes in smokers: evidence from a representative population survey 2001. Journal of the National Cancer Institute 2001;93(2):134-8.

7. Kozlowski L and O'Connor R. Cigarette filter ventilation is a defective design because of misleading taste, bigger puffs, and blocked vents. Tobacco Control 2002;11(suppl.1):i40-i50. Available from: http://tc.bmjjournals.com/cgi/content/abstract/11/suppl_1/i40

8. Kozlowski L, O'Connor R and Sweeney C. Cigarette design. In: Risks associated smoking cigarettes with low machine-measured yields of tar and nicotine Smoking and Tobacco Control Monograph. Bethesda, Maryland: US Department of Health and Human Services, Public Health Service, National Institutes of Health 2001. Available from: http://cancercontrol.cancer.gov/tcrb/monographs/13/

9. Borland R, Fong GT, Yong HH, Cummings KM, Hammond D, King B, et al. What happened to smokers' beliefs about light cigarettes when "light/mild" brand descriptors were banned in the UK? Findings from the International Tobacco Control (ITC) Four Country Survey. Tobacco Control 2008;17(4):256-62. Available from: http://tobaccocontrol.bmj.com/cgi/content/abstract/17/4/256