There is a new article studying unreduced/2n gametes in roses. Below is the link. I feel the authors did a great job studying this topic. There are some typos, lack of capitalization of rose cultivars, etc. In some of these open access journals there is often less quality control. They cited my 2009 paper and misinterpreted the data I presented on alba roses. "In R. alba hybrids, 2n pollen was produced and
abnormal ploidy was observed [14] ". They are mixing up n and x. n refers to life stage. 2n means the sporophyte stage (typical plant cells in leaves stems and roots). The gametophytic stage is 1n or n. These are the gametes or gametophytic stage. When there is a gamete (like pollen or an egg) that has the same chromosome number as the parent plant and the gamete did not go through the full standard meiosis leading to typically half the chromosome number, we can say it is 2n pollen- pollen with the same chromosome number as the plant (sporophyte) that generated it. Anyways, their photos and data seem reasonable and sound and I really love what they did. I was hoping someone would answer some of these questions in roses and was really glad to see this article.
It is interesting that in the cross they made they found that although the percent of 2n/unreduced pollen in âOld Blushâ is low (1.39% in their samples), after they put the pollen on âOrange Fireâ the stigmas allowed a greater percent of 2n pollen into the style than normal n pollen and the seedlings they obtained then had a higher than 1.39% rate of tetraploids. âOrange Fireâ is tetraploid and âOld Blushâ is diploid. The normal n pollen from 'Old Blush would lead to triploid seedlings and 2n pollen tetraploid seedlings. It would be great if they would have also used the âOld Blushâ pollen on a diploid mom to see if the stigma of the diploid also preferentially let the 2n pollen through. I think from some of the papers by some of the French rose researchers years back, that there were some clues that maybe there was some degree of preference. n pollen tended to fertilize diploid moms to a slightly greater extent and 2n pollen from diploids fertilized tetraploid moms to a slightly greater extent.
I was transplanting polyantha rose seedlings tonight and it is fun to very rarely see some that look like they may be polyploid. The two pictured are open pollinated (likely polyploid on left and diploid on right). Perhaps the one that looks polyploid is a hybrid with a triploid or tetraploid rose, or maybe it is a hybrid with other polyantha and there was some rare 2n pollen grains in that polyantha dad. The typical polyploid characteristics include-wider leaflets that overlap each other more, thicker leaves and stems, and less branching. Iâll check its pollen size and maybe do a root tip squash to double check its ploidy. Itâll be great if it is tetraploid instead of triploid and fertile. Iâd love to cross it with standard tetraploid shrubs and hopefully more readily generate hybrids with a good amount of fertility to move forward with.
A little off-topic, David, but in the case of unreduced pollen from a diploid creating a polyploid, are those polyploids likely to have any more cold-hardiness than their diploid seedlings? I recall reading that polyploid species are more common in colder climates, but donât recall the supposed reasons for such. It might have had more to do with insuring a greater diversity in offspring in a harsher climate than a correlation with hardiness per se.
Do you have any thoughts on cross direction in regards to combining polyanthas with modern tetraploids? Pollinating polys is a painstaking process, and I donât seem to have the knack for germinating them, so Iâm inclined to use their pollen. (also thinking of Darlowâs Enigma, which I donât know if qualifies as a poly but should function the same). Iâve used Candy Oh! as a pollen parent as well as Darlowâs Enigma. In both cases itâs a little hard to verify hybridity of the seedlings, as the seed parentâs characteristics will somewhat dominate. But Iâve gotten seedlings that Iâm pretty sure are successful crosses. Would you speculate that polys would have a similar rate of unreduced pollen as Old Blush did in the study? In that case, using polyantha pollen on modern tetraploids or triploids would have a better chance of creating a tetraploid that is 50% poly.
Those are great questions. It seems like with polyploid series like you mentioned Philip that the higher ploidy is found in the more harsh/severe end of the range. What Iâve heard is that polyploid allows for more alleles to be housed within the plant from which to hopefully accumulate alleles that contribute to adaptation/survival. That is an interesting thought. Polyploids in general (there are exceptions) seem to grow a bit slower. I had a 8x R. acicularis that was really slow and thick in growth. Perhaps they harden off better come winter too and tend not to push as hard into fall? Iâm not sure.
Thatâs a great question Joe :0). I agree it would be really fun to get more hybrids with polys and a range of polyploid modern roses. It has been very hard, at least for me, to bridge polys and typical fertile modern roses like Carefree BeautyTM and others. Iâve tried both directions and havenât gotten much success. There is one hybrid I still have in a pot from after the move and need to get propagated that is a poly x âChamplainâ. It is a nice deep pink with a very nice plant habit. It is from 1999. There were some hybrids with it I liked years ago, but lost them in the moves. I wish it was healthier. What seems to really work is polys- diploid or tetraploid- with fertile tetraploid minis. âHannah Rubyâ is a triploid hybrid with a diploid poly and âHoneybeeâ is a tetraploid hybrid with an induced tetraploid poly. I suspect what is allowing those hybrids to work so freely is that the Moore minis Iâve used in the past ('Rise âN Shineâ, etc.) have some polyantha-like parents in the background and bring in a good dose of Systylae genetics (R. multiflora, R. wichurana, etc.) to match or work well with the same species backgrounds in the polys. I havenât kept up on the new minis. Maybe there are some healthier/hardier ones that could be a good bridge now. I like what you mentioned about using some triploid modern roses too. âGolden Angelâ is triploid and seems to be a nice bridge between ploidy levels. It is a parent of the diploid âTopaz Jewelâ (other parent a rugosa). Knowing it was the parent of a yellow diploid hybrid is what got me interested in wanting to use it to cross with polys to bring in yellow and stay at the diploid level. Years back I got a plant of GA and put poly pollen on it and got several hybrids. I counted some of them and those I counted were diploid. They were pastel and some had a bit more modern form to the blooms. Unfortunately, they werenât fertile and I lost them over time with harsh winters. Last year I bought some more GAâs from Burling hoping to try again with some of the current polys I have around. I think itâs possible to find some good polyploid modern roses with the right background to accept poly pollen and make the crossing easier. Itâd be great to learn others suggestions for which roses may be good to try.
Simple answer, Not necessarily.
Fagerlind (1958)
k. Rosa multifloraâtetraploids from diploidsâthree individuals, badly frozen each year. Earlier, one of them flowered rather richly, but not at all in recent years. All these specimens are now languishing considerably. http://bulbnrose.x10.mx/Roses/breeding/Fagerlind.html
Fagerlind lived and worked in Sweden, near the northern limit of this species. Doubling the chromosomes just pushed them over the edge.
On the other hand, Iâve read that âHansaâ grows happily and flowers freely in Alaska.
People who still believe that Rosa acicularis is a âpureâ polyploid series including diploid, tetraploid, hexaploid and octoploid varieties are beyond saving. Those who are capable of considering more than a handful of traits at a time (e.g., Hurst) can see that the polyploids combine traits of more than a single ancestral species. It is the combination of traits that allow some polyploids to survive in the far north, but not because they are thrown together.
Hurst (1925)
In the most complex case studied, in the octoploid species BBCCDDEE (R. acicularis Lindl.), the four double septets seem to work more or less in relays in different parts of the plant at different times and seasons, resulting in a periodic predominance of one septet over another in certain parts of the plant, the general result being more or less a mosaic of the four septets of characters arranged end to end or side by side.
Naturally with four double septets working equally and independently in an octoploid species, only about one-fourth of the characters of each septet can be represented at one time. An analysis shows that in a plant of R. acicularis Lindl. carrying four yearsâ growth of surculi, stems, branches and branchlets, about one-half of the characters of each of the four septets B, C, D and E were represented (fig. 174 e and f). http://bulbnrose.x10.mx/Roses/Hurst/HURST2.HTM
I have a question about the ploidy of pollen from triploid parents like âNew Dawnâ: do they produce both n & 2n pollen? Can it be used on diploid and tetraploid seedparents? It seems like that would be the simplest option, but nature is rarely simple.
In orchids, triploids are largely sterile dead-ends.
Iâm not sure if David has seen your question yet, but the answer is generally yes, both types of gametes may be produced by fertile triploid roses. However, for a specific cultivar, the ratio might vary and would be difficult to know except through direct observation. Trial and error can at least give you some idea whether a rose might be usefully fertile in a particular direction. With 'New Dawnâs long breeding record, it seems that it might either favor the production of 2n gametes or else breeders were primarily interested in crossing it with tetraploidsâwhich is probably true regardless. The good news is that its fertile pollen should work well with most fertile diploid seed parents, giving you the opportunity to select some excellent triploid offspring of your own (maybe even fertile ones). The reverse cross can also probably work, but maybe with a reduced success rate, at least in my experience.
Here is a previous thread where David wrote about this very subject:
Great points Stefan. In table 9 I counted seedlings of a limited number of 4x / 3x crosses to look at the distribution of ploidy from a handful of triploid males.
Leen Leus in Belgium studied ploidy in roses as well as disease resistance and still is working on roses doing great work. She overwhelmingly found 4x/3x crosses giving 4x offspring. In discussions we learned that she likely puts a lot more pollen on the stigmas than I do. In other crops it seems like higher ploidy pollen grows faster and gets to the ovules faster and preferentially participates in fertilization. With me being more stingy in pollen application and much of the pollen being aborted anyways, finding so many triploids in table 9 may just reflect that by pollinating more lightly I may have gotten a more direct reflection of the ploidy of the viable pollen grains and avoided pollen competition to some extent that seems to favor the 2x pollen grains. Iâve also gotten some diploids out of âGolden Angelâ (3x) x diploid polyanthas. Some folks over the years have said that triploids, although more challenging for fertility, can be especially nice to work with because of the wider range of variation one gets with their offspring likely due to the variable number of sets of chromosomes they can provide and how those dosage differences can lead to more variability in phenotypes of seedlings. I think there is something to it.
