In Favor of Diploid Daylilies

In many kinds of plants both diploid and polyploid forms occur. The normal gardener need not be concerned about whether the plants are diploid or tetraploid. The basic or original plant would be diploid with two sets of chromosomes, one set from the male parent and one set from the female parent. Polyploid plants (with more than two sets of chromosomes) in nature tend to become more abundant as one moves away from the equator. Polyploids tend, in general, to be somewhat hardier, larger more vigorous plants with larger, thicker flowers. One might expect that breeders dealing with polyploid plants would have the upper hand in producing plants with new characteristics, but this is not necessarily true.

Plant Ploidy:

A daylily gamete would normally have 11 chromosomes. Chromosomes are numbered from one to eleven. Any particular nuclear gene is located on one of the eleven chromosomes. For example, a gene responsible for chlorophyll production might be located on chromosome seven (I am not sure what its location actually is). Each chromosome may have thousands of genes, most of which would not be involved in phenotype characteristics which we would easily see. Most produce enzymes involved in metabolism, photosynthesis, and structural products. In diploid plants each cell would have two copies of each gene (22 chromosomes). In tetraploid plants each cell possess four copies of each gene, one on each of the particular chromosomes.

Number of sets of chromosomes per nucleus.

1. Monoploid (single set -11 chromosomes)

1. Haploid (11 chromosomes in a diploid, 22 in a tetraploid)

2. Diploid (22 chromosomes)

3. Triploid (33 chromosomes)

4. Tetraploid (44 chromosomes) Polyploid

6. Hexaploid (66 chromosomes)

8. Octoploid (88 chromosomes)

Monoploid, hexaploid and octoploids have not yet been reported in Hemerocallis

Ploidy and gametes

Diploid plants (22 chromosomes) will have their haploid/monoploid gamete containing 11 chromosomes. Tetraploid plants (44 chromosomes) will have their gametes containing a haploid number of 22 chromosomes. Note: Haploid refers to the gamete situation and the nuclei containing half the normal number chromosomes. Triploid plants would not, normally, form functional gamete because half the chromosomes come from the pollen parent and half come from the pod parent. Thirty-three can not be divided in half to get a complete set of chromosomes (11 chromosomes or a multiple of eleven). Once in a great while normal meiosis partially fails and a complete set (11 or 22) chromosomes will end up in a gamete or complete failure of meiosis will occur and the unreduced gamete will end up with 33 chromosomes. Either result may produce a functional gamete. The ploidy of the offspring, then, may end up being diploid, triploid, tetraploid or even pentaploid (five sets of chromosomes), if fertilized with a diploid gamete from a tetraploid plant.

Genetic Variation of Diploids and Tetraploids

When comparing diploids and tetraploids it is important to consider the amount of variation in the plants. The more different genes or allele variation present in a plant, the more variation one might expect in the offspring. The amount of variation is determined by the variation found in the parental plants. Tetraploids have been reported to have lower variability then diploids. This would likely be due to the fact that many of the original diploids have not been treated to double their chromosome number. For hybridizers this means that more variation may be expected in diploid crosses than in tetraploid crosses because there are more different alleles available to influence the phenotype/appearance. To increase the genetic diversity of diploids even further one might recommend crossing today’s diploids with the original diploid species.

Diploids Genetics:

Simple genetics involving a single pair of genes can be illustrated with a capital A, standing for a dominant gene and a lower case a standing for a recessive gene. The genotype of a diploid could be said to be an AA (homozygous for the dominant genes), Aa (one dominant gene and one recessive gene or heterozygous) or aa (homozygous for the recessive genes). We could say that the gene, we are calling capital A causes chlorophyll production and the gene lowercase a does not cause chlorophyll production. Therefore, genotypes AA and Aa would cause the leave to be green and genotype aa would be colorless, or more accurately yellow, because of the carotenoids and zanthophylls, instead of green. With complete dominance, the amount of chlorophyll and growth produced would be the same in genotypes AA and Aa. In incomplete or partial dominance the genotype AA would produce more chlorophyll and therefore be healthier than genotype Aa with less chlorophyll. As there is a difference in the health/speed of growth in daylilies, we might want to look at the genetics of the offspring of parents that produce one or more chlorophyll-less plants. I have noticed that I sometimes get colorless or yellow seedlings in my beds. These plants do not live long. I suspect that from a cross that produces these colorless seedlings some of the green seedlings would be homozygous dominant and some of the seedlings would be heterozygous for the chlorophyll gene. If the gene demonstrates partial dominance the leaves of heterozygous plants would have less chlorophyll and therefore grow slower.

Complete Dominance

Notice that the resulting genotypic ratio is 1AA: 2Aa: 1aa

The phenotypic ratio would be 3 green: 1 white with complete dominance

With partial dominance the phenotypic ratio would be 1 green: 2 light green: 1 white.

Out of 1000 offspring we would therefore expect nearly 250 green: 500 light Green: 250 white with incomplete dominance. Note: due to the randomness of chromosomal separation during meiosis, while we statistically expect a 1:2:1 ratio, it would be much more likely to get a ratio close to 1:2:1 than the exact expected ratio.

In daylilies, the red pigment, cyanidin, will usually often show as an reddish-orange because of the presence of carotinoids (yellow pigments) in the flowers. If we can simplify the discussion and assume that there are no carotinoids present, then in red flowers we can call the "red" gene "R" and the recessive gene "r". We wouldn’t know it, but if a mutation of a dominant gene to a recessive gene occurs then:

In order that you can be 50% sure of getting a white flower and be sure that the mutated allele actually shows in the phenotype at least four offspring from each cross must be examined. Four each from the three selfed Rr x Rr and four each from the three selfed RR x RR. Remember we cannot tell the genotypes from each other when they all have red flowers so we really don’t know whether we cross heterozygous to heterozygous and homozygous to homozygous. In summary, to be able to view the phenotype caused by a new mutation, to would take seven crosses, assuming six seeds per pod. One starting pod and then checking the offspring of the six resulting seedlings. Fifty percent sure is not very sure. To be more certain, one must produce many more offspring.

If the red anthocyanin (cyanidin) in the daylily flower color were regulated by a single (it may not be a single gene) partially dominant gene, then:

Example: R= functional gene producing a red pigment - - - r = non-functional pigment, or gene produces a no colored pigment

Pink flower X white flower		OR		Red Flower X white flower
Rr X rr				RR X rr

With the tremendous variation we see in daylily color, it is obvious that many genes determine anthocyanin production, which controls the color of the flower. Different genes, or at least, similar genes located in different locations on the chromosome are probably responsible for the coloring of different parts of the flower (petal edge, eye, throat, and various markings. Some of the alleles may be dominant or recessive; some may show partial dominance or co-dominance inheritance. Because there are many genes involved with the expression of flower color or many other characteristics, it may be somewhat difficult to see when a new characteristic or variation has occurred. Remember, with daylilies we typically deal only with a comparatively low number of seedlings per cross it may be difficult to even pick up when a new phenotype (characteristic) shows up. The more seedlings produced from a single cross the better. For good data with diploid plants we might want to consider the production of 20 to 50 seedlings from each cross. (It’s easy to consider it but very difficult in practice)

Tetraploids Genetics:

Tetraploid daylilies have four sets of chromosomes or two pairs of chromosomes instead of the diploid one pair. In order for a recessive mutation to show up in the phenotype, the allele must be homozygous recessive (yyyy) and present on all four chromosomes. With diploids the recessive allele (yy) must be present only on the two chromosomes present. In daylilies a light pink or melon pigment is changed to a yellow pigment called beta-carotene. As long as the enzyme functions correctly, the allele will be dominant (Y) and produce the yellow pigment. If the enzyme doesn’t function correctly then the allele would be considered to be recessive (y) and the pigment produced would be melon in color.

First Cross (recessive mutation [y] occurs in aa dominant gene[Y])

Second Cross (self of flower with unseen mutation)

F₂ Third Cross (self of all the previous offspring, some of which would have the unseen mutations – all of the offspring must also be crossed because there is no way to determine which plants carry the mutation).

We will look at the result of just one of these crosses.

YYyy x YYyy

All seeds must be planted and the flower color examined.

Complete Dominance - Phenotypic Ratio = 35:1 (one out of every 36 crosses should possess the genotype yyyy and phenotype of a melon colored flower – if all the right plants were used in the crosses, and there is no way of knowing so all the plants must be selfed). This would be expected but it would not always happen. We all know that in a family with four children we would expect two to be girls and two to be boys. To be sure that the mutation is seen, many more selfs must be made.

Partial dominance means just that. Each dominant allele will cause some of the enzyme to be produced. Each dominant allele will produce an increase in the phenotypic effect, such as increased height, flower size, color, eye size, etc.

YYYY - yellow - phenotype when all genes are having an effect on appearance

YYYy – off yellow - phenotype when three of the four genes are affecting the appearance

YYyy – yellow (melon tinge) - phenotype when half the genes are affecting the appearance

Yyyy – yellow/melon mix phenotype when one of the four genes is affecting the appearance

yyyy – melon - phenotype when none of the four genes are affecting the appearance

In summary, if we assume that each seed pod contain six seeds then in order to show a recessive mutation in diploids it would take six different selfings of flowers. With tetraploids it would take nearly 600 selfings to have a recessive character show up. Many more than 600 seedlings from selfed hybrids must be examined to determine, for sure, if a new recessive mutation has occurred in tetraploid daylilies.

As almost all gene mutations produce recessive (usually non functional) genes, the chances of having the phenotype (appearance) change after one cross is zero. With the second self-cross there is also a zero chance of the characteristic showing up in tetraploids, but a one in four chance in diploids. In tetraploids, with the third self, assuming all possible offspring develop, there would be one chance in 36 (just in the one "correct" cross) or a one in 144 chance that the character would show up when all the offspring of the second cross are examined.

Remember, with complete dominance, in order to see the recessive phenotype all crosses must be made. In tetraploid crosses, over two hundred offspring must be examined in order to be fifty percent sure that you can pick up a recessive mutation if it occurred

It is important to self plants once in a while. This is the best way to find new recessive mutations. One should also realize that the more homozygous recessive genes present are in a plant, the less vigorous the plant is likely to be (homozygosity tends to be increased by selfing). An increase in gene recessive homozygosity will allow us to see new mutations, but it will also cause daylilies to be less vigorous and more susceptible to disease. To counteract this trend toward homzygosity, it is important to realize that, while we talk about the decreased number of selfed crosses needed in diploids to show the recessive characteristics, compared to tetraploids, we don’t necessarily recommend actually making all these crosses. Doing so would decrease the vigor of the plants. What we are really saying is that crossing diploids, in general, is more efficient than crossing tetraploids. The more crossing is done with other diploids, the better the chance of having a recessive character show up. If someone were to find interesting characteristics due to repeated selfing, to prevent homozygosity problems, one might then cross the plants with desired characteristics with unrelated plants. Using the right crosses, this would tend to retain the characteristic, at the same time, increase heterozygosity.

In partial dominance inheritance, the dominant alleles (dominant genes at a particular location on a chromosome) are functional but do not produce enough protein/enzyme to have the maximum phenotypic effect. This means that a changed gene (mutated) in a diploid plant may be seen be a vigilant observer. In tetraploid plants, with twice the number of genes involved with any one characteristic, it will be considerable more difficult to differentiate between a mutated gene and a normal one. It might be possible to recognize the fact that the yellow pigment is not quite as strong, but it would be very difficult.

If a totally dominant mutation occurs, it will immediately show up. In daylilies, however, I suspect that some characteristics may show partial dominance. In tetraploids, any affect on appearance (phenotype) of a mutation will be slight and great care must be taken to see it. In tetraploids, with more genes, it may be much more difficult to differentiate between flowers/plants which are normal and those with new mutations than it would be in diploids.

To take advantage of the diploids ability to show new mutations and the tetraploids ability to show greater growth and size, one might suggest that more hybridizers specialize in diploids. When these diploids demonstrate improved potential they can then be changed to tetraploids.

Some questions:

How many of us make multiple attempts of the same cross? Learning about the genetics of daylilies will take many self-pollination attempts to produce a hundred seedlings or more with very careful record keeping. We must know the exact numbers of individuals showing every characteristic out of each cross. We may then determine which genes show complete dominance, incomplete dominance/blending inheritance and multiple gene effects.

As the characteristics change when changing a diploid daylily to a tetraploid daylily the two flowers will sometimes appear quite different. As an example, look at the appearance of Gossard’s diploid Radiation Biohazard compared to the tetraploid form. Should these two forms have different names? One of the problems in doing this would be the tremendous increase in names used. On the other hand, I suspect that there are more differences between some of the diploid/tetraploid varieties than there are in many of the named siblings some breeders offer today.

Has anyone come across any characteristics in daylilies, which appear to be passed on only from the pod parent? There are genes in most organisms located in the mitochondria or chloroplasts, which are passed on by the female and not the male. We might expect the same thing in daylilies.

1. Norton, J. 1972. Hemerocallis Journal 26 (3) This information is old and is here merely to be an example.

2. Tomkins, J.P., Wood, T,C, Barnes, L.S., Westman, A. and Wing, R.A. 2001. Evaluation of genetic variation in the daylily (Hemerocallis spp.) using AFLP markers. Theoretical and Applied Genetics, 102:489-496.

3. Bisset, K. Specrophotometry, 1976. Chromatography and Genetics of Hemerocallis Pigments. Dissertation, Florida State Univ. Bisset actually found seven different forms – dark green, pale green, yellow green, lemon yellow, pale yellow, cream and ashen white. Three forms had, at least, some chlorophyll and four had no chlorophyll but differed in the amounts of xanthophylls and carotinoids. The last (ashen white) apparently had no chlorophyll or carotinoids present. The dark green and pale green forms tended to survive, but one might surmise that the pale green forms probably did not grow as well. Hybridizers might do well to eliminate the pale green seedlings even though they may grow, more or less, normally.

4. We should also be aware that the sharpness or clarity of the red pigment is influenced by a dominant gene causing the color to be drab or muddy.

5. According to Bisset, there is a separate gene for the pink color.