Smoking and Tobacco Control Monograph No. 7 
taken on the basis of total recovery. This study found a strong relationship 
between yield and nicotine intake per cigarette per day that was totally 
different from any other study's findings. There are serious methodological 
concerns that might affect these conclusions. First, there were only 
33 subjects, and recruitment procedures were unclear. In contrast, data 
already presented involving 2,000 individuals have shown a weak or no 
relationship between cotinine and FTC nicotine yields, so there is a problem 
of generalizability of the Byrd data. Second, an examination of particular 
data in the 1-mg tar group results in an average intake of 9 mg nicotine. 
However, the studies of Gori and Lynch (1985) and Benowitz and colleagues 
(1986b) showed an average cotinine concentration of 200 ng/mL for large 
groups of smokers of the same ultralow-yield cigarettes. A cotinine level 
of 200 ng/mL would correspond to an average daily intake of 15 or 16 mg, 
not the 9 mg reported by Byrd and colleagues (1995). Thus, even if only 
one group is studied, the subjects are not representative of the much larger 
numbers of subjects that have been studied by other investigators. 
CARBON Gori and Lynch (1985) have provided data on carbon monoxide levels in 
MONOXIDE a large group of smokers of cigarettes of different yields (Figure 7). Their 
AND FTC study and other studies have found virtually no relationship between 
YIELD carbon monoxide levels and FTC yields, even for the ultralow group. 
Thus, FTC carbon monoxide yield appears to be of no value in predicting 
human carbon monoxide exposure. 
TAR-TO-NICOTINE The tar-to-nicotine ratio also must be considered. Some authors 
RATIO have argued that even if there is only a small reduction of 
nicotine, because the machine tar-to-nicotine ratios are lower for low-yield 
cigarettes, there is a disproportionately greater overall health benefit due 
to reduced tar exposure (Russell et al., 1986). The question is whether 
machine-determined tar-to-nicotine ratios predict ratios of exposure in 
human smokers. 
The author attempted to examine this question by studying mutagenic 
activity of urine by Ames test. This test involves culturing salmonella bacteria 
that are unable to generate histidine and therefore cannot grow. However, 
if the bacteria are mutated so that they can make histidine, they can grow. 
Growth can be quantitated by the number of colonies on a culture plate, 
and the number of colonies can be used as a measure of mutagenic activity 
of chemicals that were added to the salmonella culture before incubation. 
It is well known that the urine of cigarette smokers is mutagenic, presumably 
reflecting exposure to chemicals found primarily in cigarette smoke tar 
(Yamasaki and Ames, 1977). 
Figure 8 shows urine mutagenicity data from one individual smoking 
his or her own brand of cigarettes who switched to a Camel (1 mg nicotine) 
cigarette, then switched to a I'rue (0.4 mg nicotine) cigarette, and followed 
with a period of no smoking (al)stinence). Urine mutagenicity was fairly 
stable for the individual, and there was no difference between smoking the 
Camel and True cigarettes. 
104 
