This is second version of the abstract:
http://www.ars.usda.gov:80/research/publications/publications.htm?seq_no_115=226444&pf=1
Link: www.ars.usda.gov:80/research/publications/publications.htm?seq_no_115=226444&pf=1
Don and Fa, thank you for your responses, and I too value your comments and the spirit of the discussion.
Don, it is too bad that some rose breeders prefer not disclosing rose parentages. I think that it is kind of silly that they donāt, given the very large number of possible types of seedlings in any given cross. No one is going to try to copy Kordes work to come up with the same commercial varieties. It would be helpful though to know parentage, particularly when trying to āuncoverā a desirable recessive trait that may or may not be present.
You may find that the ancestral cultivars are recently more diverse given the number of varieties that Mr. Ralph Moore has produced from species or near species, and Bill Radlerās Knockout series, and even Tom Carruthās work has added a more diverse gene pool. Working with any of these roses offers a whole new range of possibilities of combinations and recessive genes.
Though fragrance may have a complex biology and mode of inheritance, I agree that existing cultivars offer a generous range of possibilities. I am not certain that ease of rooting is necessarily a complex biology. It might be simple genetics with a very few number of genes involved. I definitely agree, however, that horizontal disease resistance is very complex, which is why it is so hard to āselectā for (quotation marks used because I think that it is impossible).
As for what the market place is demanding in the way of ādisease free rosesā, you have only to look at the sheer numbers of Knockout roses that are being produced each year. It is phenomenal! This may not be happening with the exhibitors, but they are a very small segment of the rose-buying public.
It is lamentable that many do treat roses as annuals, actually it is almost disgusting to me. Roses really are something like pets. They can and should have a very long lifespan (at least the ādesirableā ones should).
āIt is probably a mistake to regard disease resistance as having simple Mendelian inheritanceā. Yes, I agree that it would be a mistake. The discussion about dominant and recessive genes involved in horizontal disease resistance was with regard to the probably very numerous loci on rose chromosomes that are involved in resistance. Some of the loci involved in resistance may produce relative resistant phenotypes when a dominant allele is present, whereas other loci may require the presence of homozygous recessive alleles. Letās say for example that there are 50 different gene loci involved in resistance in roses. Where there are dominant genes involved, it is likely that these will be selected for and any homozygous recessive resistance genes will be ālostā, particularly if a diploid species resistant rose (having good horizontal resistance), is being introduced to the tetraploid modern roses. After a very few generations, I think that you would probably end up focusing on a very few number of genes, effectively breeding for verticle resistance.
Fa, I agree that it is so much more fun for the amateur breeder to breed for what is simply pleasing to them. It is like painting from the heart, rather than being on assignment. Individual preferences can be focused upon and the results enjoyed.
Regarding the ātonsā of recessive genes - I agree that there have to be many available. One of my goals in working with the hulthemias has been to develop several different lines of repeat blooming hulthemias each going back to āTigrisā. In this way, I hope to combine several of these lines to perhaps uncover an unexpected recessive trait that might be interesting. For example, the single leaf character of H. persica, rather than the compound leaf of roses might be fun to recover in the repeat blooming hybrids (or the absence of auricles). This may only be possible by using several different distinct lines from āTigrisā and crossing them with one another. No doubt there are other recessive genes that may produce an unexpected effect in roses.
Henry thank you for the link. If I am reading the abstract correctly, the introduction of new species roses may not bring as much diversity into the germplasm of roses as we might be expecting. I believe that there are probably very few genes between rose species that are truly ādifferentā. If horses and humans have a large number of their genes in common (and they do), then certainly differing rose species are really very much alike.
Jim Sproul
May someone think that strength and desease resistance or tolerance were intentionally bred out then he could think that they are only to be bred in.
If, as I do, he thinks this was achieved unwilling then looking for what did not work right and for another breding path is legitimate.
Or not?
Hi Jim Sproul:
Let me second your frustration with breeders who donāt disclose parentage (I know I am preaching to the choir). When you see examples like Out of Yesteryear and Star Magic both coming out of the same cross as well as all the wonderful pictures Timo posted on āTimoās seedling topicā showing the diversity of roses coming out of the same cross it is most unlikely that anyone is going to get duplicate results ā and if they did there would be no commercial demand for another of the same. It is like expecting subsequent children to be like your first ā can anyone on this forum say that their second child is the same as their first. I am not sure my daughter and I would be making such a big commitment as beginners to this hybridizing stuff if we hadnāt gotten so much encouragement from all the wonderful, unselfish and GENEROUS (you know who you are) folks on this forum.
Thanks, bob in New Orleans
Hi Pierre,
In re-reading some of my recent posts on this thread, I can see why it looks like I might be suggesting that we shouldnāt use clean, vigorous species in our breeding programs. I believe that we should do whatever we can to improve the cleanliness of roses. What I do believe though is that it will be nearly impossible to fix good horizontal disease resistance into a breeding program. Nevertheless, the more clean genes that are incorporated into breeding lines, the better chance we have of thwarting the diseases - an approach similar to rotating pesticides might work??
In the medical field, we are finding more and more ānewā antibiotics are producing stronger bugs. Interestingly, sometimes the older antibiotics are proving useful when the ānewā, better and stronger ones fail.
Jim Sproul
lol, speaking of which, Im on SMV for life. Talk about an old antibiotic.
Jim, do not worry, I like friendly controverse.
About DNA most has no known utility. Less than 20% is considered effective in humans that are diploids. Even less for tetraploids from tetraploid species. And it is at DNA level that little diversity has been found among modern roses.
About pedigrees things are not as simple as some think. Years ago I had a discussion on that matter with Alain Meilland. We amateur breeders cannot duplicate others achievements but it is not at all out of reach for the pros. Identical duplication is useless and almost impossible but take advantage of others findings is usual. I.e that one can get recessives from some parents will be easily done again as it was for early pelargonidin orange red vars.
So that many published pedigrees are not reliable.
That is far worse than undisclosed.
About strength and desease resistance or tolerance being unintentionally bred out it was my opinion years ago. It is no longer as in my experience very, very few modern roses seedlings are better at these than parents. Usually few are better than the weaker most desease prone parent.
In easier climates cumulating partial resistances from different sources may seem hopefull as is horizontal resistance. Up to now in bad enough seasons all failed. Here in a wet episode Pretty Lady, Baby Love and Knock Out all spoted to defoliated rifht after the first flowering flush.
Finding new never failing and not yet known resistance genes is illusory as all the most promising species have been tried long ago.
That is why I am looking for another breeding path.
Am really enjoying this discussion and finding very little that I can disagree with.
I would point out though, that as much as I like fragrance in roses, Iāve been surprised to find quite a few people who would RATHER have little or no scent. [LOL - I can almost hear the gasps from my fellow fragrance devotees - LOL]
And just a thought that occurred to meā¦ isnāt horizontal resistance just the natural version of āstackedā vertical resistances. If a rose had a single gene that gave resistance to a pathogen, it would be relatively simple for the pathogen to evolve a way around that resistance. But if a species of rose has been dealing with a pathogen over many generations of natural selection, there would likely be an accumulated bunch of many different resistance genes. It would be much more difficult for the pathogen to evolve a way around all of the resistance genes simultaneously. I agree that it would probably be nearly impossible to intentionally select for a single rose plant that had combined all of these many different resistance genes. But we can still move in that direction, by bringing together as many different sources of resistance as we can find. I think that is where the various species and tough old roses have particular value.
I especially agree with the advice to āreinvent the wheelā, and think that there is a lot of merit in going back and repeating successful crosses that involve species and tough old roses. For example, I remember reading somewhere about a dominant gene giving a vertical resistance to blackspot, that had been found in Rosa multiflora. If a single multiflora hybrid were used to introduce that gene into the modern rose gene pool, that would be great, until blackspot evolved a way around the resistance. But there are probably other resistance genes also present in that multiflora, that give it more durable, āhorizontalā resistance. If instead of a single multiflora hybrid, we were to use multiple hybrids, we would be much more likely to bring in some of those other resistance genes.
I guess my point isā¦ that although we might have incorporated some disease resistance from a particular species by using a hybrid or two, there is still so much more to gain by going back to that same species, again and again.
Sometimes I wonder if it might just be easier to breed a relative few traits from the moderns (repeat bloom foremost) into the already disease resistant species.
The concepts of vertical and horizontal resistance are interesting academic topics but really are meaningless for the amateur applied biologist. It is more useful to think of disease resistance as being a spectrum where a few progeny among hundreds (thousands?) from any given cross may possibly stand out as having exceptional resistance, and shoot for creating and finding those few exceptional individuals. The trick is finding it coincident with exceptional blossom and form.
Finding new never failing and not yet known resistance genes is illusory as all the most promising species have been tried long agoā¦That is why I am looking for another breeding path.
The best new path might be right in your back yard. The local wild population of R. multiflora here in north central Connecticut seems impervious to disease, for instance - not entirely immune, but shrugging it off with impunity.
No native multiflora here but polyanthas and hybrid musks are high in multiflora genes.
Sempervirens a local species I am considering is an underused species. Dark green in alcaline soils and always very healthy.
Our local multifloraās had been VERY healthy (to the dismay of many farmers); but now Rose Rosette Disease is killing them in great numbers.
I have a plant of Rosa helenae that is still untouched (by any disease) though surrounded by the dying multifloras. I think this would be another good species to use.
A nearby Rosa wichuraiana seems to be holding on too, although Iām not sure if it is resistant or just keeps regenerating from rooted pieces that have separated.
Do the polyanthas and hybrid musks have better than average health in your climate, Pierre? And I would certainly try sempervirens if it were local for me :0)
Hi, folks,
Iād like to add my āHear, hear!ā to the general kvetching about people not disclosing a roseās parentage (if they know it). I agree with Jim; it -is- silly, even if just mathematically. The chances of being able to get the same result from the same cross more than once are genuinely astronomical; and getting -similar- results could teach us all so very much. Wonāt go into the math, but itās interesting if youāve had too much chocolate or caffeine. 
(That said, all my 2007 seedlings are open-pollinated, alas; but at least I think I know what went wrong last year.)
Hi,
me too Iād like to add my āHear, hear!ā
I am keeping track on what Kordes, Tantau and Noack are dealing with in breeding for resistance.
There are some traces on projects between 1970 and 2000.
I found lots of reports on websites, documents in public libraries and in special interest books and I had personal communication with some of the involved people.
There is a trace #1 beginning with a triploid bastard of Goldfinch x R. multiflora.
The primary bastard was backcrossed to R. multiflora.
There was at least one seedling of the backcross which had deregulated (substituted) to diploidy.
In 1988, after some inbreeding generations they got an homozygous (RR) genotype for resistance against all blackspot-races known at their trial fields.
It has some additional desirable properties like
-
resistance against some isolates of powdery and/or downy mildew
-
dwarf growth habit
-
repeat blooming
-
stems without prickles
Tetraploid clones were made by in vitro propagation and colchizine treatment, resulting in a quadruplex type (RRRR).
These clones were crossed to some of the best tetraploid garden roses e.g. Sexy Rexy.
After backcrossing the new hybrids to the garden roses they got final products, e.g. Eliza.
In 1996 they went back to the diploid 1988 breeding line and crossed to R. wichuraiana.
The F1 was crossed and later backcrossed to a diploid hybrid of trace #2, which again is a diploid bastard from R. multiflora * garden rose.
The trace is lost but in 1998 there were some other species discussed as parents for further resistance pyramidizing.
Now take the facts that Liz posted, add 9 years to 1998 and have a look on the varieties Thomas Proll listed.
I would like to add some names that I think they are crosses from those projects: Felicitas, Maxi Vita, Juanita and Unicef-Rose.
There is no chance and no need to do crosses with the objective of getting similar results.
This is a chance to hook on the professional breederās work and go on with own ideas - simply buy their roses.
Ren
I disagree here. Making as many as possible different crosses inspired by others achievements is a necessity if we want to preserve genetical diversity.
I have Eliza and its BS resistance is inferior to Elinaās one here. Elina has the best resistance/tolerance for an HT I know. I grow Felicitas that if an uncivilized larger plant with long stems and reduced fertility (triploid?) is healthy. Maxi Vita is not healthy here.
I grew a lot of Polyanthas and Musks including some of Lens and bred them. Those I know are not very resistant. Only very tolerant that is if not immune they are not debilitated and never entirely defoliate. Progenies when crossed with HT were strong but not healthy enough.
Were it available there were a lot of possibilities for a multiflora resistance to BS gene to be introgressed in actual HTs.
Sp multiflora is known as having a very wide area and is said to be very varied. So it is possible there are some more BS resistant.
Another possibility is that this resistance was lab tested on young leaves and is not so strong in my rose field.
Pierre, do you think it would be better to go back to Wiesse max graf(Felicitas parent). I wonder how its disease resistance is. It is probably diploid.
Patrick
Patrick, not Pierre, Iām new at this, but I thought maybe my input could help just a tiny bit from observation?
I got a cutting of Max Graf last summer and itās done rather well considering it being a very wet spring, hasnāt gotten any BS so far here in Maryland that I can tell. We have high BS pressure, I just finished stripping the leaves off of a supermarket mini, so the pressure is there, maybe even worst in the country. However, my plant is new so Iāll judge it harder next year.
It does however have a very pleasing leaf shape, and quite shiny and matte considering the rugosa parent. Iām excited to what it will look like as a shrub.
Also, on HMF it says it is indeed diploid.
Pierre,
I am glad about your critical feedback.
It took a long time for me to find people I can discuss with on these issues.
Again
We agree in this if your target is new genetic diversity, in this case you don
Iāve been tossing this topic around in my head for a while now. The two big decisions facing breeders in any species regardless of scale of operation is (1) their germplasm base - what should be included as parents, and (2) what and how to do selection. This thread sort of encompasses both of these decisions. I decided when I ventured into roses that I would probably steer away from making species crosses, but instead use species crosses of others. The reasons really are less important than the qualification that I put upon this decision - that being that I would try to accumulate different samplings of the same species. For example R. kordesii, I have Rosarium Uetersen which comes thru Karlsruhe, Scarlet Moss which comes thru Dortmund, Champlain which comes thru L02, PALs Niagara and Prairie Celebration which both come thru L83, and John Cabot, WIlliam Baffin, and Henry Kelsey which are directly from R. kordesii. Each one of these groups or samplings of R. kordesii probably have some genes in common but also have some genes that are unique. So overall, I have a fairly diverse collection of R. kordesii genes in my backyard. Certainly in my small breeding program R. kordesii is the best example of different samplings of a species. All of the R. laxa x R. spinosissima in my breeding program traces back to Suzanne, all of the R. eglanteria traces back to Clare Grammerstorf, there are two samplings of R. wichuraiana one thru Simon Robinson and one thru Immensse, R. davidii all goes thru Baby Love, I have only found one R. virginiana species cross - Jim Lounsberry, and I have not been able to completely get a handle on the R. soulieana out of Julia Child/Ebb Tide/Outta the Blue.
That brings us to selection, it is a double edged sword, meaning that you get what you select for, but you also risk loosing genetic variation. There is a selection concept called index selection where you essentially weight the importance of traits and select on the index value. There are other selection concepts that set minimum cut-offs for traits. The consequence is that the ābestā phenotype for an individual trait may not make it into the chosen few, when selection is based on the index or the minimum cut-off. The upside is that you are probably maintaining genetic variation for many of the traits of interest.
Iām still searching for roses that widen my species diversity by representing novel samplings of the species and I am still struggling with tossing out seedlings that do not make the minimum cut-off with it comes to form and disease resistance, likewise I am tempted to toss very healthy seedlings simply because I find the bloom boring.
For what it is worth, those are my hopefully somewhat coherent thoughts, Liz
Rene,
Thanks for the information on the German roses. Although I understand the point Pierre is making, I also use some of the newer Kordes roses that have very good disease resistance in my garden as parents. Interesting to read some more information about what species have been used.
Liz, your way of thinking about what species comes through what rose is interesting also. You point out the āfunnelsā in variation where a breakthrough species-hybrid is used as a starting point for a whole fan of descendent-hybrids. Although each of those descendents have low variation because of the funnel-effect, it is still interesting to cross them together and hope to stack different interesting species-traits that are still present in each of the descendents. Because you would expect an interesting species-trait to be selected for in the descendent, so it should still be present even af few generations later.
Rob
This is confirmed for flower characters such as yellow or orange (and will be for hultemia spot) but up to now it is not for plant characters such as desease resistance or hability to grow ownroot.
As Liz I think that strong selection pressure (including for a desirable feature) leads to genetical impoverishment.
It is a necessity to promote different plant and flower types and to organise diversification of our germplasm.
Using others achievement is something I do and do not warn against. Only there are other things to do than all at the same time focusing on the latest best parent.
Noak and Kordess introduced Immensee seedlings many many years after the first successful intros from this parent. There are no reasons to forgive it.
Breeding another high potential parent comparable to Immensee that is quite close to species is contemplatable. I.e. shruby with large flowers.