Plant Breeding Terms and Theory
R. A. Smith - Severn, MD
The following eight-part discussion focuses on plant breeding topics that may have some applicability to one’s hosta hybridizing program. It is meant to introduce standard plant improvement theories often employed with crop plants, some common terms of reference and documentation annotation styles, and briefly present some concepts related to handling gene variants (alleles) that may affect the outward appearances (phenotype) and approaching a goal-oriented hosta breeding program[i]. This set of plant breeding topics was originally drafted in response to discussions on the Hosta Seed Growers forum and has subsequently been edited for the Online Journal. My hope is that some will find the terms of reference and the theoretical ideas presented both informative and thought provoking when designing hosta breeding goals and specific protocols to achieve those goals. Enjoy.

Part I:
Plant Mating Strategies
There are five main plant mating strategies to consider, from random to crossing like types with each other. This is a good place to start since I’ll be using a couple of these terms later in this article.  As originally described by Wright (in 1921)[ii], there are five primary plant breeding strategies and for hostas, many hybridizers are employing genetic disassortative mating or phenotypic disassortative mating – that is, the best ways to push the envelope for new plant forms. As background, the 5 types of plant mating Wright described (for plants that can be hybridized and/or selfed):
1) Random mating (all plants have equal chances of being crossed with others, bloom times are assumed to be the same)
2) Genetic assortative mating (means hybridizing with like background plants -- used for inline breeding to fix the favorable multilocus genotypes and establish homozygosis)
3) Phenotypic assortative mating (crossing two plants that have a common look although not a common background -- to hopefully concentrate extreme phenotypic expressions of the trait sought)
4) Genetic disassortative mating (less closely related than random mating plants -- mainly employed to develop plants with a broader genetic base for future breeding projects)
5) Phenotypic disassortative mating (different linage and different outward appearance -- a wide cross to maintain genetic diversity)

Using any of the non-random methods above affects the overall genetic correlations among relatives, as well as the genetic structure of the overall group.  By employing influence onto a group of like plants that can openly cross with each other, that is, a population considered to be in equilibrium, we influence the overall population and can achieve results well beyond what is normally seen for a hosta species.  Additionally, by interacting with hostas to influence the progeny, we also throw off the “Hardy-Weinberg equilibrium[iii]” which in theory states that under random mating, with equal chance that any pod and pollen can combine, that traits may move between plants but overall as a large plant population, the various genes stay in a equilibrium for the overall population.  This is why field of species hostas may have variability but it is limited to the transfer of a set number of genes causing that variability. By isolating extreme specimens and applying genetic disassortative mating, the hosta community has seen some truly superior and unique phenotypes.

I should comment here that hosta leaf variegation is controlled by cytoplasmic inheritance -- that is, via chloroplasts and the distribution of DNA into gametes following nuclear division is unpredictable and not equitable thus why we use the streaked plants as the maternal parent when seeking to produce variegated offspring as the maternal parent is the one that contributes the chloroplasts. This said, I have read that in at least one major crop breeding test (on tobacco), about 1 in 10,000 plants did pick up an extrachromosonal trait via pollen so there may be exceptions with other plants as well.
Part II:
Gene Actions

According to one of the leading distinguished professors in the field of plant breeding, George Acquaah (Principles of Plant Genetics and Breeding, 2007), there are 4 primary gene actions to consider when working with plants: 1) dominance, 2) additive, 3) overdominance, and 4) epistatic. Additionally, Acquaah describes in his text books how some genes are major genes for a trait while others play more of a minor role (e.g., polygenes)[iv].  For an excellent coverage of these topics, I’d also recommend the “Fundamentals of Plant Genetics and Breeding” text by Welsh, 1990 reprint[v]. Of the text books on this topic I’ve read (12), the two referenced above provide excellent plant breeding general topic coverage of these concepts. Let’s briefly look at these types of gene actions a bit more.

Dominance: Mendel's work highlighted traits that exhibited complete dominance --the all on or off type approach. This gene action equates to what is often called dominant or recessive genes --the theory being, that in the right combination, one or the other trait shows full dominance. Building on Gregor Mendel's work on dominance theories, plant breeders over the past 100 years discovered that sometimes genes may partially mask a trait. This is referred to as incomplete dominance or partial dominance.  The classic example is a red and a white flowering annual; crossing with each other results in some red, some pink (blended), and some white progeny with 1/2 being pink (a 1:2:1 ratio versus the expected 3:1 ratio (red to white) with complete dominate expressions). Another concept to mention here is one of codominance -- two traits can be expressed completely in the progeny (to make a new look or phenotype altogether)[vi]. In simple terms, incomplete dominance likely produces a blended phenotype while codominance produces unique and separate phenotypes (Acquaah, 2007) with both traits being fully expressed (not blended). In hostas, to identify and then cross plants that have different looks (phenotypic disassortative) with the hope for both leaf traits to be fully expressed would be a worthwhile goal.


Additive: This deals with the concept that each gene enhances the overall expression of a trait (up and beyond what is expressed in either parent), then it is said to be additive gene actions. This concept’s basis is that various gene variants (alleles) involved in trait expression each add one unit of the trait, assuming that the plants that have good General Combining Ability (GCA). For instance, if 4 different units are combined (2 from each parent), then the progeny may have up to 4 total units of the trait expressed (and thus, greater trait expression than either parent that only had the 2 units each). For example, in wheat, the red kernel colors can range from dark red to white based on 3 loci (the location of a gene on a chromosome) that can be added together to make a continuous range between these two colors (ratios being 1:6:15:20:15:6:1 from white to medium red to dark red -- Nilsson-Ehle 1911[vii]). If I had to take a guess with hostas, it appears that the puckered leaf trait may behave in this way.  (Please let me know if your experience holds true as this initial observation is based on only very limited examples.) I’ve also made the assumption that the amount of red/purple pigments expressed in hosta petioles is an additive trait -- this will take some dedicated hybrid crossing experiments and a few years to fully explore.

Overdominance: This exists when each allele at a locus produces a separate effect on the phenotype, and their combined effect exceeds the independent effect of the alleles. In other words, the heterozygous offspring have greater expression of a trait (say vigor) up and beyond either of the parents. When identified, the parents are said to have Specific Combining Ability (SCA)[viii]. I’ve seen this primarily discussed in various texts as manifesting in hybrid vigor (also referred to as “heterosis”).


Epistatic:  Often described as non-allelic gene interactions. That is, variants of genes (alleles) in one location have an effect on alleles in a distant location which can lead to the masking of a gene's expression. When something odd shows up (or does not show but was expected) in your plant breeding program, think about this concept as a possible underlying cause.  In hostas, I often wonder if the yellow color is sometimes subject to this effect (especially with H. kikutii types) in which the yellow color is likely often overridden by a dominant epistatic green condition that normally is not a factor for other species.


 *For this discussion, I’m going to ignore the challenges posed by multiple copies of the chromosome numbers, for instance, polyploidy: when entire chromosome sets are repeated or aneuploidy when only partial extra chromosomes exist.  For coverage of these more complex topics, I recommend “Principles of Cultivar Development, Volume 1, Theory and Technique”, Walter R. Fehr, 1987 as a good reference to check out from your local library.

Editor's Note:  In order to preserve the author's precise formatting, including proper end notes, the complete article has been converted into PDF and Flip-Book formats.  Please click on one of the links to continue with page 5 of "Plant Breeding Terms and Theory."

Next Article
Previous Article
Contents Page