We conducted an assessment of North American heavy-duty engine emission test results for biodiesel from 49 experimental studies, including both engine dynamometer and vehicle test results. Comparison with a commercial database showed that the engines in the emissions database are not representative of the existing North American in-use fleet as of 2007; more than 50% of the tested engines were of 1995 or earlier vintage. Nevertheless, the results show that the use of a common biodiesel blend (B20) consistently reduces emissions of particulate matter, hydrocarbons, and carbon monoxide by 10-20%. Tests with B20 show varying effects on oxides of nitrogen (NO x ). If results for pre-1992 two-cycle 6V-92TA(E) engines (which represent 0.2% of the 2007 in-use fleet but 28% of the engines tested) are removed, then there is no statistical evidence that the average NO x emissions from B0 and B20 are different (p value of 0.50 for an estimated average increase of 1%). Several researchers have used changes in engine calibration to eliminate any NO x penalty associated with B20 (in engines that show an increase in NO x with B20), while still maintaining the advantages of B20 in reducing other pollutants. The emissions effect of B20 on heavyduty diesel truck emissions did not show any correlation with model year or type of fuel injection equipment.