Smoking and Tobacco Control Monograph No. 7 
of carcinogens and tumor promoters in cigarette smoke led to a gradual 
change in the design and composition of commercial cigarettes in North 
America, Western Europe, and other developed countries (Hoffmann and 
Hoffmann, 1994a; Jarvis and Russell, 1985). The modifications were intended 
to reduce both the toxicity and the carcinogenic potential of the cigarette 
smoke. Although research on the changing cigarette was pursued in several 
countries, this chapter deals primarily with the developments relating to 
U.S. cigarettes between 1954 and 1993. 
At the basis of all analytical assessments of smoke composition lies the 
standardization of machine smoking methods, first suggested for empirical 
cigarette smoking in Europe (Pfyl, 1933; Pyriki, 1934). In the United States, 
Bradford and colleagues (1936) developed a procedure for cigarette smoking 
on the basis of "arbitrarily selected" parameters of a 35-mL puff volume, 
a 2-second puff duration, and one puff per minute. The only goal of this 
method was to offer a means for comparing the smoke yields of various 
types of cigarettes; there was no intent to simulate human smoking patterns. 
The influences on smoke yields and composition that are exerted by the 
overall physical characteristics of a cigarette — including its length and the 
butt length to which it is smoked, its circumference, whether it is filtered 
or nonfiltered, and the effects of the puff volume, puff frequency, and puff 
duration; the type and cut of tobacco used as a filler; the properties of the 
wrapper; and the mode of precipitation of the condensate — were described 
in many research papers during the 1960's (Wynder and Hoffmann, 1967). 
For regulatory purposes, Pillsbury and colleagues (1969) adapted in principle 
the method of Bradford and coworkers (1936) and made some refinements 
to establish what became known as the FTC method; the smoking parameters 
were still a 35-mL puff volume, a 2-second puff duration, and a 1-puff-per- 
minute frequency. What was new was the definition of the butt length to 
which a cigarette was to be smoked. Butt lengths were set to be 23 mm for 
plain cigarettes and length of the filter plus overwrap with an additional 
3 mm for filter cigarettes. CORESTA, the International Organization for 
Research on Tobacco, developed a comparable method that is widely used 
in most of the developed countries (CORESTA, 1991-1993). 
This chapter describes the analytical data obtained with the FTC method, 
although many studies (Russell, 1980; Herning et al., 1981; Kozlowski et al., 
1982; Fagerstrdm, 1982; Haley et al., 1985; Byrd et al., 1994) have shown 
that the standardized machine smoking method does not reflect the smoking 
habits of consumers of filter cigarettes. This is especially so for filter cigarettes 
with low and ultralow smoke yields, because smokers of such cigarettes 
tend to inhale more deeply and draw puffs more frequently to satisfy a 
physiologically conditioned need for nicotine (U.S. Department of Health 
and Human Services, 1988). 
Figure 1 presents the sales-weighted average tar and nicotine deliveries 
of all U.S. domestic brands for the years 1954 through 1993 (Hoffmann and 
Hoffmann, 1994a). Fhis figure also shows the major changes in the makeup 
of U.S. cigarettes, such as the introduction of filter tips, porous cigarette 
16 
