2 had a similar fate in that portions of chromosome 2 are also found on chromosomes 7, 10, and perhaps 4 (Table 1). More extensive evaluation of these duplications will provide an indication as to whether there has been any bias in rearrangements. For example, there is a strong bias for paracentric inversions, as opposed to translocations and pericentric inversions, between potato and tomato. It was reasoned that the bias toward paracentric inversions reflects the relatively low effect of paracentric inversions on fitness (Bonierbale et al., 1988). Additional studies of chromosomal duplications in maize could provide additional insights into the kind of rearrangements that are most evolutionarily stable.

Is maize typical with regard to its polyploid history and prevalent chromosomal duplication? There is no doubt that polyploidy is common in plants, with up to 70% of angiosperms owing their history to polyploidy (Masterson, 1994; Stebbins, 1950). Furthermore, genetic maps demonstrate that a great number of species contain chromosomal duplications. Even species with streamlined genomes contain chromosomal duplications; for example, rice has a large duplication between chromosomes 11 and 12 (Harushima et al., 1998) and Arabidopsis also has at least one large chromosomal duplication (Mayer et al., 1999). Other plant genomes with chromosomal duplications include sorghum (Chittenden et al., 1994), cotton (Reinisch et al., 1994), soybean (Shoemaker et al., 1996), and Brassica species (Bohuon et al., 1996; Cavell et al., 1998). Some of these genomes are degenerate polyploids like maize, but others may owe their chromosomal duplications to independent segmental events.

It is important to note that chromosomal duplications are usually inferred from genetic maps, but most (if not all) genetic maps are based on low copy-number markers. Low copy-number markers are systematically biased against detecting duplicated chromosomal segments, and hence the extent of chromosomal duplication is likely grossly underestimated for most plant taxa. In addition, the resolution of most genetic maps is low, such that relatively small areas of chromosomal duplication cannot be detected. The result is that we do not have a realistic understanding of either the extent to which chromosomes are duplicated or the extent to which genomes contain functional redundancies. We can, however, look to Arabidopsis sequence data as preliminary examples of the extent of chromosomal duplication. Based on the sequences of chromosomes 2 and 4 (Lin et al., 1999; Mayer et al., 1999), it is estimated that 10–20% of the low-copy regions of the Arabidopsis genome lie within duplicated chromosomal regions (Mayer et al., 1999). Given that the Arabidopsis genome is streamlined, this percentage is undoubtedly much higher in