Lady Duncan - Hexaploid Hybrid Rugosa?

HMF lists Lady Duncan as a hexaploid - can this be true? Parentage is listed as: Rosa wichuraiana Cr


In my view this is not very plausible, to get a hexaploid out of this diploid x diploid crossing.

The other thing … what would happen, if there are suddenly 6 sets of chromosomes in the cells?

They would simply be downregulated after s while, to tetraploid or even diploid chomosome sets.

Those hexaploid combinations of rose chromosomes that normally had been diploid or even tetraploid are not very stable, as I have heard.

So, - this would be really very odd, if it really existed within Lady Duncan’s genome.



I agree with Arno in that it is not very likely to happen, but it can. Joan Monteith found a hexaploid plant of R.woodsii growing wild, the ploidy was confirmed by David. R.woodsii is normally a diploid.

Soren, Do you know of a source for Lady Duncan? There aren’t any sources listed on HMF.


It would be interesting to get a Lady Duncan and do the cytology. That 6x likely R. woodsii has very large pollen the volume expected for 4n pollen (12x)! Perhaps that is a trait that was inherited by this plant from a diploid parent. Perhaps the diploid parent had a megaspore or microspore mother cell with duplicated chromosomes (4x) that just became the gamete instead of going through the reductional and equational divisions to produce 4 nuclei all 1x. Then that gamete pollinated from a gamete by a tetraploid or from another diploid or itself with a 2n gamete for 6x all together in the end.

I have a 6x seedling from a diploid polyantha hybrid. I suspected it was polyploid because of its leaf and flower characteristics and after counting its chromosomes was really surprised it was 6x instead of 4x.

I guess odd things happen from time to time. If I can get a Lady Duncan I would love to do a root tip squash on it.



I’d like to try a fertile hexaploid seed parent with banksia hybrid pollen if anyone comes up with a good one. Thanks

Hi Paul

I have searched high and low for this rose - but no success so far - but I will certainly try.

Another question - most Rugosa like tetraploid?


Randy Hudges crossed a diploid Calocarpa (seed parent - 7 chromosomes) with a hexaploid R. Nutkana (21 chromosomes in pollen ). This should give a tetraploid.

Randy shared some of the seeds with me:

I have a successful cross of the above pictured tetraploid with my tetraploid acicularis:

This past year I got hips from the following crosses:

(217 (Calocarpa X Nutkana)) X (224(Royal Edward OP))

(217 (Calocarpa X Nutkana)) X Carefree Sunshine

(217 (Calocarpa X Nutkana)) X Carefree Sunshine

And also from:

(413 ((Calocarpa X Nutkana) X (acicularis OP))) X (mixed red, yellow florist’s hybrid teas)

I am just about to start putting last summers seeds in petri dishes.

Hi Soren,

The rugosas that come to mind that might be tetraploid are Will Alderman and George Will. I beleive there is some questions as to whether they are tetraploid or not though.

Otherwise, David has produced a mixoploid rugosa where cells in two of the three layers have been doubled. The layer where the gamates are produced has been doubled so there is promise of tetraploid offspring from it.


Have you thought about R.moyesii or it’s hybrids like Highdownensis or Germanium. I’ve wondered if Highdownensis and Germanium are hexoploid also. I would like to use hexoploid pollen on R.glauca and thought these would be a good choice.


I have thought about moyesii. I naturally would prefer something with repeat. It’s hard enough getting repeat in banksia. I suppose it could be a long range goal.

Paul, what are you expecting from hexaploid pollen on R. glauca? I am not following.


Very interesting…

So a possible course of action could be (R. Moyesii X R. Rugosa) X (R. Moyesii X R. Rugosa)

Giving a tetraploid with 50% Rugosa blood

Has anybody tried this?


That’s going to be tough to find, a hexaploid with repeat bloom. Maybe a chromosome doubled triploid? Might be worth looking into.

Otherwise all hexaploids that I know of are once bloom. And like you say, getting repeat bloom while using them in your breeding program is going to be a long term goal.

Hi Paul

Wouldn’t the suggested cross (R. Moyesii X R. Rugosa) X (R. Moyesii X R. Rugosa) have some repeat bloom potential?


Hi you!

I did a few crossings with Rosa moyesii pollen and other diploid seed partners, last year, but so far only few functionally (seed settings) came through, maybe because of the different sections I used, or because of different cell biology too.

David writes:

“I guess odd things happen from time to time. If I can get a Lady Duncan I would love to do a root tip squash on it.”

Ok, the problem is to understand more of these odd things and why they happen sometimes more often and straight like this. :slight_smile:

In the section Cinnamomea, where also Rosa woodsii derives from, there exist diploid and tetraploid Rosa majalis idiotypes, as I am informed.

So, your idea with a triploid crossing that doubles the chromosomes is plausible to me - for at least this section.

But I’ve never heard of a stable hexaploid synstylae idiotype, thats really bizzare to me.

(By the way: the canina / rubiginosa hexaploids are part of another story … I won’t discuss it here, further.)

As this Lady Duncan also has a Cinnamomea lineage, - … perhaps it is a true pearl, fished out of the big ocean of varieties, who knows … .

I would be pleased to hear according informations, if your expectations here might be true.

This is just a terrific theme to learn more!

Think of it as an important tool where it is possible to insert genes by changing the ploidy levels might be in sight more easily for amateurs!

Heavy thing!

Greetings from Germany,


Or maybee…

(R. acicularis X repeat tetraploid) x R. Rugosa

(R. acicularis X repeat tetraploid) should give a hexaploid with repeat bloom offspring potential. This could give a nice basis for mixing diploids like Rugosa later. I looked up Will Alderman and it has R. acicularis blood. So maybee that’s how it got to be a tetraploid



"So a possible course of action could be (R. Moyesii X R. Rugosa) X (R. Moyesii X R. Rugosa)

Giving a tetraploid with 50% Rugosa blood

Has anybody tried this? "

Yes I have tried this, but without success.

What was better, has been this, e.g.:

Rosa alba (hexaploid) x Rosa rugosa (diploid)

But I really still don’t know how a possible canina influence in albas could transform the cell biology in such crosses.

Thats why I am stunning as a child, without any conclusions, if there are first hips then seeds and if there might be … seedlings.

So far none of this crossings germinated, it might be a thing for waiting a few years.

So, i am also curiously asking: has anyone got seedlings like this, yet?



Hi Henry,

From what I’ve read, using diploid pollen on R.glauca does produce a tetraploid. But the chromosomes don’t pair up well and the gamates don’t necessarily have complete sets of chromosomes. Hence a reduced or lack of fertility. It has been speculated that it is the bivalents in Caninae section of roses that are in fact preventing the univalents from forming pairs. When one uses foreign pollen on say R.glauca it will reduce this effect. That is why you will only get some of the univalents to form pairs when using foreign pollen.

The theory I’m speculating on is this. If I used foreign pollen with more chromosomes then this effect would be even further reduced. Using tetraploid pollen on R.glauca would produce a pentaploid. So to me the most logical choice would be to use hexaploid pollen. Also the egg of R.glauca is 3n so using 3n pollen on it just seamed like the best choice.

I know it is an extra step, but it might be worth it if it increases the fertility of the offspring. This is a really long range goal!

Hi Soren,

Pretending that chromosome sets are transferred whole, then each parent would have a 50% chance of producing a gamate with 1 R.moyesii set and 1 R.rugosa set. So there would be a 25% chance of a selfing to produce a offspring with 50% R.rugosa chromosomes. But chromosomes sets aren’t transferred whole because there is a lot of mixing and matching going on. So what the actual percentages are is anyones guess.


A (hexaploid R.accicularis x repeat tetraploid) most likely would produce a pentaploid (3n+2n). I can only assume the fertility of this would be similar to a triploid except the gamates would be 3n or 2n. Crossing these with a 1n diploid could produce either a triploid or a tetraploid offspring.

R.alba has unbalanced mieosis of the Caninae section. Except it is 4+2. Meaning it 4n eggs and 2n pollen. Putting 1n diploid pollen on a 4n egg will produce a pentaploid. This might be why the seeds haven’t germinated yet.

I know Rolf Seivers used tetraploid pollen with great success on Maidens Blush. But I don’t know if anyone else has used diploid pollen on an Alba.

I and others have used R.alba pollen with success. I have a seedling from David and I have several seedlings that just germinated and are under lights.


Hi Paul

Thanx for the help :slight_smile:

I got R.acicularis down as an octoploid?



Modern Roses 11 states that R. acicularis can be 28, 42, or 56. While the diploid form is now called R. acicularis Nipponensis ( which also has a 28 form).

Reference: Modern Roses 11, page 479.

I do not have Modern Roses 12, what does it say?