Principles of Plant Genetics and Breeding

540 CHAPTER 33
sustainable way. New cultivars must give more yield of saleable product at less cost of production. They must have inbuilt resistances
to pests and diseases, and increased water and mineral use efficiency, that will allow a reduced use of pesticides and fungicides
and better use of water and fertilizers. Finally, they must help meet consumer demands for convenience foods, improved
nutritional and health benefits, improved flavour, and novel products. In contrast, in Asia and Africa there is a need for increased
and stable potato production to meet increased demand for food. New cultivars must deliver higher yields under low inputs, disease
and pest attacks, and environmental stresses such as heat, cold, drought, and salinity. If possible, they should also have
improved nutritional and health properties, but the greatest need is to raise fresh weight yields from a world average of 17 t/ha to
European and North American levels of 45 t/ha (Lang 2001).
Breeding finished cultivars
Parents
Potato breeding worldwide has traditionally involved making crosses between pairs of parents with complementary features and
this is still the main route to new cultivars. The aim has been to generate genetic variation on which to practice phenotypic selection
over a number of vegetatitive generations, for clones with as many desirable characteristics as possible for release as new cultivars.
The choice of parents is all important as breeding can never simply be a numbers game. Crossing the 3,200 cultivars in the
world catalog in all possible combinations would generate 5,118,400 progenies, and raising 500 seedlings of each would give a
staggering total of 2,559,200,000 for evaluation – an impossible task. In contrast, a phenotypic assessment of 3,200 cultivars is
feasible, and so is a genotypic assessment of diversity with molecular markers. Hence breeders can now think in terms of capturing
allelic diversity in a smaller core set of parents and of using association (linkage disequilibrium) genetics to choose parents
genotypically as well as phenotypically (Simko 2004). They can also use genetic distance based on molecular markers to complement
co-ancestry/pedigree analysis (Sun et al. 2003) in order to avoid closely related parents, and hence inbreeding depression,
and to ensure genetic variation for continued progress.
As genetic knowledge accumulates, it will be possible to choose parents for use in pair crosses so that one or both parents have the
desired major genes and alleles of large effect at quantitative trait loci (QTLs). Major genes have been mapped for flesh and skin
color, for tuber shape and eye depth, and for resistance to late blight, nematodes, potato viruses X, Y, and A, and wart. QTLs of
large effect have been mapped for maturity and resistance to late blight, potato cyst nematodes, and potato leaf roll virus (PLRV).
In contrast, many economically important traits still appear to be complex polygenic traits and these include tuber dormancy, dry
matter and starch content, fry color, resistance to Erwinia blackleg and tuber soft rot, tuberization, and yield. For these traits,
breeders will still have to rely on phenotypic data and use knowledge of offspring–midparent regressions to determine crossing
strategy. A statistically significant regression is evidence of heritable variation, and the slope of the regression line is a measure of
heritability. With a highly heritable trait like fry color, the midparent value is a good predictor of the mean performance of the offspring
and a few carefully chosen crosses can be made (Bradshaw et al. 2000). In contrast, with only a moderately heritable trait
such as yield, offspring mean is less predictable and more crosses need to be made to ensure that they include the best possible.
Early generations
The program at the Scottish Crop Research Institute (SCRI) before 1982 was typical in its handling of the early generations
(Bradshaw & Mackay 1994). Visual selection reduced the number of potential cultivars from 100,000 (200 crosses × 500
seedlings) in the seedling generation in the glasshouse to 40,000 spaced plants at a high grade seed site in the first clonal generation,
then to 4,000 four-plant plots at the seed site in the second clonal generation, and finally to 1,000 clones in replicated yield
trials at a ware site in the third clonal generation. Several independent reviews concluded that such intense early generation
visual selection was very ineffective (Bradshaw & Mackay 1994).
A potato breeding strategy (Table 1) has been developed at SCRI that avoids intense early generation visual selection between
seedlings in a glasshouse and spaced plants at a seed site (Bradshaw et al. 2003). Once pair crosses have been made, progeny
tests are used to discard whole progenies before starting conventional within-progeny selection at the unreplicated small-plot
stage. Clones are also visually selected from the best progenies for use as parents in the next cycle of crosses whilst they are multiplied
to provide enough tubers for assessment of their yield and quality. Midparent values, as well as progeny tests, are then used
to select between the resultant crosses. Material from other breeding programs can be included in the parental assessments and
used in the next cycle of crosses if superior. Finally, in seeking new cultivars, the number of clones on which to practice selection
can be increased by sowing more true seed of the best progenies, but without selection until the small-plot stage. The theoretical
superiority of this strategy lies in being able to practice between-cross selection for a number of economically important traits
within 1 or 2 years of making crosses, something that is not possible on individuals as seedlings in the glasshouse or spaced plants
at the seed site. At SCRI, seedling progeny tests are used for resistance to late blight, resistance to the white potato cyst nematode
(Globodera pallida Stone), and tuber yield and appearance, as visually assessed by breeders. Tuber progeny tests are used for fry
color and a second visual assessment of tuber yield and appearance. The use of progeny tests for key traits also means that full-sib
family selection can be operated on a 3-year cycle for these traits, an improvement on the practice of clonal selection after a further
six vegetative generations (i.e., not using potential cultivars as new parents until they are entered into official National List Trials).