Around 1900, the "tin" in tin cans contained about 12% lead, which leached into fruits and vegetables. You have to wonder if there's any way of correcting for that. [Source: "The Great A&P" by Marc Levinson, page 50.][Edit: read the paper cited and they make no mention of this]
Do you suspect that lead to be distributed in a way that would confound the results? In the absence of a theory, I'd expect lead to be fairly evenly distributed amongst cities in that particular way.
