This paper explores two issues that have received limited attention in the exchange rate pass-through literature. First, are the pass-through estimates sensitive to the choice of the exchange rate index? Second, are pass-through estimates asymmetric with respect to the sign of exchange rate changes? Using data for 87 industries, we find that the answer to both questions is yes. J-test results indicate that the "Major" exchange rate index produced by the Board of Governors of the Federal Reserve System tends to fit the data better than two alternative indexes. With respect to asymmetry, we find that pass-through is both more likely and greater in magnitude when the dollar is depreciating than when it is appreciating.