daylily hybridization

daylily hybrid
daylily hybrid
daylily hybrid


Sculpted daylilies are mostly in shades of yellow. It should be simple to hybridize a flower that is pink and sculpted. However, it has proven to be difficult. They do exist, for instance, 'Coquina Key' - but I want to use these traits in a example to help understand how two linked genetic traits (yellow and carved) can be unlinked even with the added complication of tetraploidy.

It's in fact much more complicated than crossing a sculpted daylily to a pink one and crossing one's fingers. I suspect the combination does not occur because the gene for sculpted flowers (which I assume is recessive, I have to verify this experimentally) is located on the same chromosome as the gene for the yellow color. That means that yellow and sculpted travel together on the gametes (pollen and ovule). Tetraploidy definitely is NOT your friend, here. But all sculpted flowers are tetraploid, so we'll make do with what we've got.

You need to begin with genetically homozygous parents if you want to have any clue about what you're going to get down the generations.

Our goal is: (ywhiteywhiteywhiteywhitessclptdssclptdssclptdssclptd)PPPP a "non-yellow non-smooth pink"

We're crossing tetraploids "yellow non-smooth non-pink" to a "non-yellow smooth non-pink" - that's right, a boring, non-carved WHITE. We'll ignore the genes for pink for now.

F1: (YYYYssss) X (ywhiteywhiteywhiteywhiteSSSS)

All offspring will be:

(YYywhiteywhiteSSssclptdssclptd) "(heterozygous yellow/non-yellow) (heterozygous smooth/non-smooth)"

They will all look yellow, most will be sculpted. Hey wait... that's not good! But you do have your STUD plants. You hope that the magic of recombination, possibly a rare event in this case, will occur and result in the following chromosome: "ys" - If you can get a non-yellow non-smooth chromosome, you're in business and the door is opened to many other colors.

Now you take your stud plant which has this genotype:

(YYywhiteywhiteSSssclptdssclptd) - choose yellow flowers that are not too bold, and with some hints of sculpting, but not extreme. Use these for sibling crosses.

Normal non-recombinant gametes:
Lucky recombinant gametes(rare):

Now you have to cross your stud plant to another one that will help you distinguish between the normal non-recombinants and the lucky recombinants. There is no easy way - because unfortunately we are working with tetraploids. You'd be very lucky to correlate these genotypes with phenotypes, ie, guess what the genotype may be according to the appearance of the flower. Here is the miserable table assuming a 10% recombination rate, from a random pool of sibling crosses from the F1 above:
YYssclptdssclptd 45% Y ywhite Sssclptd 5% Y ywhite Sssclptd 5% ywhiteywhite SS45%
YYssclptdssclptd 45% YYYY ssclptdssclptd ssclptdssclptd YYYywhite Sssclptdssclptdssclptd YYYywhite Sssclptdssclptdssclptd YY ywhiteywhite SSssclptdssclptd
Yywhite Sssclptd 5% YYYywhite Sssclptdssclptdssclptd YYywhiteywhite SSssclptdssclptd YYywhiteywhite SSssclptdssclptd Yywhiteywhiteywhite SSSssclptd
Y ywhite Sssclptd 5% YYYywhite Sssclptdssclptdssclptd YYywhiteywhite SSssclptdssclptd YYywhiteywhite SSssclptdssclptd Yywhiteywhiteywhite SSSssclptd
ywhiteywhiteSS 45% YYywhiteywhite SSssclptdssclptd Yywhiteywhiteywhite SSSssclptd Yywhiteywhiteywhite SSSssclptd ywhiteywhite ywhiteywhite SSSS
The genotypes outlined in purple will have TWO recombinant chromosomes, and will segregate in your favor, that is, these siblings (unfortunately phenotypically indistinguishable from the others in yellow), will have these alleles travelling in tandem on the same chromosome, a new allelic combination:
  • ywhiteywhitessclptdssclptd - the one we want!
  • YYSS - though we don't care about this half of the equation.
Some other genotypes will have ONE recombinant chromosome, and ONE non-recombinant chromosome. They too can give the desired result, although the probability is cut by half compared to the two recombinant chromosome situation.

If you make random sibling crosses, and the recombination rate is 10% as per the above example, here are the daunting odds:

45% 5% 5% 45%
45% 20.3% 2.25% 2.25% 20.3%
5% 2.25% 0.25% 0.25% 2.25%
5% 2.25% 0.25% 0.25% 2.25%
45% 20.3% 2.25% 2.25% 20.3%

This is terrible news: only a total of 1% of your crosses (those among the rarest category, double recombinants) will even have a shot at getting 1/16 of their offspring having the desired genotype of the much sought after white, carved plant

In other words, you'd need to screen through 1600 seedlings, on average, to hit a white sculpted flower if the recombination rate between these two genes is 10%.

If the genes are really close, say, with only 1% recombination rate, you'd need to plant 16,000 seedlings. Difficult, but not impossible! If you don't do the right crosses as shown above, chances are greatly reduced.

The good news is, once you have a white sculpted flower, you may begin to outcross and introduce your favorite colors, as white is pretty much recessive to everything else. You have separated "carved" from "yellow" and the colors of the rainbow are your only limit. The gene for sculpted will still be recessive, but you'll see it in your new line of F2 sibling crosses from the white carved parent to the non-carved pink, purple, white, blue, white... White will be recessive too, but that's in fact an advantage that will make it easier for the other colors to show through.

E-mail Chantal.