Over the last 15 years, interest in plant virus evolution has re-emerged, as shown by the increasing number of papers published on this subject. In recent times, research in plant virus evolution has been viewed from a molecular, rather than populational, standpoint, and there is a need for work aimed at understanding the processes involved in plant virus evolution. However, accumulated data from analyses of experimental and natural populations of plant viruses are beginning to delineate some trends that often run contrary to accepted opinion: (1) high mutation rates are not necessarily adaptive, as a large fraction of the mutations are deleterious or lethal; (2) in spite of high potential for genetic variation, populations of plant viruses are not highly variable, and genetic stability is the rule rather than the exception; (3) the degree of constriction of genetic variation in virus-encoded proteins is similar to that in their eukaryotic hosts and vectors; and (4) in spite of huge census sizes of plant virus populations, selection is not the sole factor that shapes their evolution, and genetic drift may be important. Here, we review recent advances in understanding plant virus evolution, and describe the experimental and analytical methods most suited to this purpose.