which genes are introgressed via bridging species, are highly resistant to the reniform nematode (Stewart and Robbins, 1995), and the most resistant of G. arboreum accessions suppress nematode reproduction 95% or more compared to susceptible G. hirsutum. As the extreme case, G. longicalyx, from which genes can be transferred only with great difficulty, is virtually immune. This apparent inverse relationship between compatibility and resistance within Gossypium greatly confounds strategies and funding for developing resistant cultivars.
Discussions among public and private cotton breeders in the United States (personal communication, R. L. Nichols, Cotton Incorporated) revealed that once genetic introgression of nematode resistance into G. hirsutum is achieved and breeding lines carrying nematode resistance are thereby made available, it will be essential to also provide the cotton seed industry with molecular, or readily observable phenotypic, markers for nematode resistance. Without markers for resistance, seed companies would not undertake breeding programs aimed at incorporating nematode resistance, because of the resources it would require of their breeders to conduct nematode reproduction assays. As a result, marker discovery becomes an essential component of resistance introgression research by public researchers.
A related factor is that a single gene, or a cluster of genes that is inherited simply and thus that can be monitored with one or a small number of molecular markers, is generally more desirable from the standpoint of costs of cultivar development than are multigenic sources of resistance. Because genetic recombination rates generally decrease with increasing genetic distance, introgression of resistance from some alien sources of resistance could involve small or large haplotypes inherited as a single alien segment, that would be trackable within progeny by a single marker. If resistance is due to an R-gene that is part of an R-gene cluster, then the patterns of resistance to certain other pathogens and pests might be inadvertently affected. In cotton, genes for resistance to Fusarium wilt, Verticillium wilt, bacterial blight, and the cotton root-knot nematode [Meloidogyne incognita (Kofoid & White) Chitwood race 3] would be of particular concern. Also, if the reniform nematode R- gene is embedded in a relatively large haplotype that encompasses genes that deleteriously affect other traits of economic importance, introgression of the R-gene might deleteriously affect overall performance.
Resistance from G. longicalyx and G. arboreum appears to be inherited as a single gene (Avila et al., 2006; Robinson et al., 2007). In contrast, inheritance of resistance from the G. barbadense accession, TX 110 (PI 163608), appears to be polygenic (J. L. Starr, personal communication). Thus, marker-assisted manipulation of TX 110-derived resistance might require the discovery and simultaneous monitoring of multiple markers during cultivar development. With increasing numbers of loci, the likelihood increases of encountering complications from R-gene clustering and/or haplotype based negative traits. Multigenic resistance also means that only a small number of progeny in a segregating population are likely to retain the full level seen in parents. However, multigenic resistance to pathogens in general, is more durable than monogenic resistance, because it is more difficult for pathogen populations to overcome resistance based on multiple genes.
There are several nematode-crop systems where nematode-resistant cultivars have been deployed. Nematodes in general do not overcome resistance as fast as fungi and bacteria, probably because parasitic nematodes reproduce much more slowly than fungi and bacteria, but eventual development of resistance-breaking nematode populations is common, especially among sexually reproducing species, such as the reniform nematode (Niblack et al., 2002; Plantard and Porte, 2004; Cook, 2004; Bakker et al., 1993; Roberts, 1995; Williamson, 1999; Castagnone-Sereno, 2002; Roberts, 2002; Robbins et al., 2001). An