Cotton Breeding for Fiber Quality Improvement - Industrial Crops: Breeding for BioEnergy and Bioproducts - page 200

DNA methylation, it is likely more information about the cotton epigenome will be
uncovered over the next decade, and ways of exploiting this in breeding or even
management will be explored.
While there are questions about whether the overall genetic diversity of modern
cotton cultivars is narrowing and limiting genetic gain, most would accept that
breeding programs should ensure they maximize genetic diversity within their
material. Maintaining access to germplasm resources will therefore be important
in cotton breeding. Breeders will need to actively introgress interesting and impor-
tant traits from non-adapted germplasm including wild cotton relatives, as well as
access lines or cultivars from other breeding programs that may have similar goals.
Utilization of Germplasm Resources in Modern Cotton
Improvement
Traditionally, and in increasing order of complexity and effort, there are three gene
pools, primary, secondary, and tertiary, that can be accessed for genetic improve-
ment of a crop like cotton. Given the dominance of G. hirsutum in the world
production, most of the global breeding effort is toward that species. The primary
gene pool is other tetraploid cottons with the same A and D genome complement
that are sexually compatible with G. hirsutum . For example, considerable time and
effort has gone into interbreeding G. hirsutum and G. barbadense cultivars to move
agronomic traits in both directions improving productivity and/or quality of culti-
vars in both species [ 8 , 19 , 20 ], although those efforts appear to have mostly run
their course. Race cottons, often domesticated but largely unimproved G. hirsutum
or G. barbadense accessions, also offer a potential source of new variability for the
cultivated species, but their use has been hampered by a number of factors including
their photoperiodicity that makes breeding difficult in more temperate regions away
from the equator. This has been partly alleviated by converting them to a
day-neutral habit [ 37 ], although they are still poorly utilized because of severe
yield drag. It remains to be seen whether the promise of marker-assisted selection
(MAS) (see sections “ Molecular approaches to yield and quality enhancement ” and
“ Breeding strategies and integration of new biotechnology ”) will improve upon the
current level of trait introgression between and within these different species. Most
breeders are reluctant to sacrifice the high level of linkage drag, aberrant pheno-
types, and often fertility issues associated with genomic breakdown when crossing
these two species (even between elite cultivars) or the loss in agronomic perfor-
mance when crossing elite cultivars with wild or unimproved material. There is
justification for ongoing research on wide introgression. The wild tetraploid cottons
( G. tomentosum , G. mustelinum , and G. darwinii ) have also been relatively
underutilized in cotton breeding because of these same issues, but there is consid-
erable scope for using them in developing intermediate breeding stocks to produce
material with enhanced pest, disease, or stress tolerance not available from existing