General Considerations

When starting a backcrossing program, there are a few items that a breeder needs to consider.

First of all, backcrossing is most easily conducted if the character being added is easily selectable. To be easily selected it needs to be: simply inherited (although the backcross approach may be applied to quantitative traits, especially if molecular markers are used to select the gene of interest); dominant; and easily recognized in the hybrid plant. An example of an easily selected trait is awns (recessive trait) on wheat heads (Fig. 7) and a less easily selected trait is dough strength (affected by the environment and generally under the control of many different genes; Fig. 8). See the Cereal Quality video below for some examples of grain qualities (no audio on the video).

Figure 7. Awnless wheat (left) is easy to detect and select from wheat with awns (right).

Figure 8. A more difficult trait to select for is the bread-making quality of wheat. Mixograph curves are used for determining the strength of dough which affects this trait (0 is considered weak and 7 is considered strong). In this situation, 4 is most desirable. Baenziger, P. Stephen; Shelton, D. R.; Shipman, M. J.; and Graybosch, R. A., "Breeding for end-use quality: Reflections on the Nebraska experience" (2001). Agronomy & Horticulture -- Faculty Publications. 26. https://digitalcommons.unl.edu/agronomyfacpub/26.

There are also some absolute requirements for a successful backcross program.

First is a satisfactory recurrent parent. Good candidates for the recurrent parents are those varieties or inbred lines that have maintained their importance for many years despite the release of numerous ’improved’ varieties intended to replace them. A recurrent parent that has maintained its importance is one that remains high yielding in comparison to other lines over time. Remember that it often takes 5 to 7 years to develop a backcross line. In that 5 to 7 years, new cultivars are developed. If the new cultivars with higher yield are better than the recurrent parent with the new trait, no one will grow the recurrent parent with the new trait. Thus the time and effort to back cross the new trait into the recurrent parent would be wasted except as a parent to develop other new cultivars.

Secondly required is a useful gene that is unaffected by environment or being in a new genetic background. High heritability of the character being transferred is important for phenotypic selection. Transfer is easiest when the character can be identified readily in hybrid populations by visual selection, by simple tests, or by linked molecular markers. There has been some success in dealing with quantitative traits through backcrossing; however, like all other methods it depends on the ability of the breeder to distinguish between genetic and environmental variability, and to select those individuals that are desirable for genetic reasons. The general agricultural worth of the donor parent need not be of great concern. Selection of the donor, (nonrecurrent parent) is almost exclusively on the basis that it exhibits the character in a particularly intense form. This high gene intensity is important since often the presence of modifier genes in the new genetic background (recurrent parent) cause some intensity to be lost even though the most stringent selection has been practiced throughout the backcrossing program.

A final absolute requirement is a sufficient number of backcrosses must be used to reconstitute the recurrent parent to a high degree if that is the goal. Recovery of the recurrent parent is a function of the number of backcrosses and the effectiveness of selection for recurrent parent type in the early generations. Usually 6 backcrosses with selection for type in the early generations has proved sufficient. However, in wide crosses and/or with undesirable linkages a greater number of backcrosses may be necessary. Remember that in practice the objective is to improve the recurrent parent by only one trait. It should be noted that since the recurrent parent is already a proven variety or line, it is usually not necessary to conduct extensive performance trials once satisfactory introduction of the desired character has been achieved. Previous data on the recurrent parent will indicate the benefits of the backcross line assuming the added gene or linked genes are not detrimental.

The last consideration a breeder is aware of is the influence of environmental conditions on a backcross program. Provided the expression of the character being transferred is sufficient for selection, backcrossing can be conducted in any environment. For example, several generations may be grown per year in the greenhouse (Fig. 9).  Again, tightly linked molecular markers that are not affected by the environment to the gene of interest can be extremely useful in ensuring the backcross procedure can be done in any environment.

Figure 9. Wheat breeding in a greenhouse. Wheat heads are bagged to control the pollination process according to the backcross breeding plan. Photo by P. Baenziger, 2009.