We have identified relict olivine and pyroxene grains in chondrules from ordinary and carbonaceous chondrites that probably did not crystallize in situ. Some of these olivines are clear, but others contain fine‐grained Fe,Ni (‘dusty olivines’) and resemble previously described occurrences in ordinary chondrites. There are also chondrules in which all olivine is dusty. We conclude: (1) not all relict olivines are dusty, (2) not all dusty olivines crystallized outside the chondrule in which they occur,and (3) some dusty olivines were produced during chondrule formation by a reduction process that affected the whole chondrule. The occurrence of dusty olivines and relict pyroxenes and olivines in chondrules from carbonaceous as well as ordinary chondrites supports the argument that chondrules from all chondrites had similar origins and histories. We propose that chondrules and mineral fragments were transported across f(O2) gradients in the solar nebula while they were hot, or were reheated in an environment different from the one in which they formed. Partially molten chondrules sometimes incorporated mineral grains or chondrules with different redox states, producing compound chondrules and chondrules containing anomalous grains. Dusty olivines may also have formed when hot chondrules were transported to regions of lower oxygen fugacity.