Smoking and Tobacco Control Monograph No. 7 
The machines are calibrated to take one puff of 2-second duration and 
35-mL volume every minute. Cigarettes are smoked to a butt length of 
23 mm or the length of the overwrap plus 3 mm, whichever is longer. 
When the cigarette has been smoked down to the prescribed length, it 
burns through a string that has been placed on that mark; this causes a 
microswitch to be flipped, which in turn disconnects that particular port 
of the smoking machine. (Although this seems like a fairly unsophisticated 
way of terminating the test, more sophisticated methods — such as infrared 
detectors and thermal sensors — have been tried and rejected over the years.) 
Five cigarettes of each variety are smoked, one at a time, using the same 
filter holder.^ (A total of 100 cigarettes of each variety are smoked to get the 
official tar, nicotine, and carbon monoxide ratings.) After the smoke from 
those five cigarettes has been filtered through each filter pad, the holder is 
removed and weighed. The difference between the weight of the holder 
before and after the smoking process divided by the number of cigarettes 
smoked is the total particulate matter collected from the cigarette smoke. 
The filter pad is then extracted with a solvent,^ and the moisture content 
is determined by injecting a measured amount of the extract into a gas 
chromatograph and comparing the resulting peak against the standard curve. 
Ratings for the three constituents reported by the Commission are then 
determined as follows: 
• Nicotine: As with moisture, a specified amount of the extract from the 
filter pad is injected into a gas chromatograph, and the resulting peak 
is compared against the standard curve.^ 
• Carbon monoxide: The gas collected in the plastic bag is passed 
through an infrared detector to determine carbon monoxide levels. 
• Tar: Tar level is determined by subtracting water and nicotine levels 
from total particulate matter. 
Tar and carbon monoxide figures are rounded up or down to the nearest 
milligram, while nicotine figures are rounded to the nearest 10th of a 
milligram. Varieties with tar and carbon monoxide results below 0.5 mg 
per cigarette or nicotine results below 0.05 mg are reported as <0.5 mg or 
<0.05 mg, respectively, because the FTC test method is not sensitive enough 
to report these components at lower levels. 
Although the ratings are based on 100 cigarettes, at least 150 (and 
preferably 200) cigarettes of each variety are needed for the test to ensure 
‘ To make certain that the machine is working properly, at least 4 of the 20 ports are reserved on each run for 
"monitor" cigarettes — cigarettes with known yields for tar, nicotine, and carbon monoxide. 
* the solution contains extractant and internal standards: 2-propanol containing 1 mg anethole per mL as 
an internal standard for nicotine and 20 mg ethanol per ml, as an internal standard for water. 
* Ultraviolet spectroscopy was used to determine nicotine until 1980, when it was replaced by gas 
chromatography. 
10 
