Method 8082A states in section 18.104.22.168, "For each Aroclor, the set of 3-5 peaks should include at least one peak that is unique to that Aroclor". Since AR1016, AR1232, AR1242, and AR1248 all have exactly the same peaks (just in different height/area ratios), how do you find a peak that is unique to each one? The same case exists for AR1260 and AR1262.
The method states at least one peak that is unique to that Aroclor should be included. However, since many of the Aroclors have the same peaks, but with different area ratios, it is not practical to do so on samples that contain similar congeners. For PCBs that do have different congeners, for example 1016 and 1260, unique peaks can be chosen.
Choosing peaks that have the best response and the least inferences is recommended. Pattern matching is probably the most important part of the PCB analysis, which can become difficult if more than one Aroclor is present or if weathering has occurred. By choosing five peaks or more for quantitation, the analyst can choose the peaks that correspond closely with one another (in concentration) and remove the values that are affected by inferences and weathering.