I think it was Paul Barden who mentioned a few years ago on the old forum that for early pollen releasers he would collect the anthers then spray the stigmas with a water filled spray bottle to wash the pollen off. When the stigma dried he would do his cross. I thought it was a good idea.
1 Like
You should see me during spring with Jewlers magnifying glasses on,I look like the nutty Professor. I use them to see when pollinating and to see if the stigmaâs have been contaminated.
I do the same thing with the water to the stigmas.
I also purposely try to cross a species with its self in order to know how likely it will cross with its self. So far most of the species I have tried this with have crossed quite happily with its self. I personally feel but donât have the proof yet that in any given species that a few members of that species will probably not cross with themselves. And when one finds these members that are more willing to take foreign pollen over their own it is a valuable trait. However this will also lead into problems when one is compelled to backcross or cross siblings, because this unwillingness can crop up again.
By the way Warren I forgot to tell you I love the Mimas X Maigold seedling.
[quote=Warren]
I am not sure who it was, but there is this chaps therory that all roses evolved from a single artic decaploid rose dropping ploidy numbers as they interbred, and then there is the other therory, that ploidy numbers increased as they interbred. Its all sort of like the Astro Physicists two therorys of the expanding and contracting universe. .[/quote]
Warren, the chap was C. C. Hurst. He found that most of the diploid species of Europe, Asia and North America fall naturally into 5 groups, disguished by sets of characteristics. He reasoned that these are the traces of primordial species (pre-Ice Age) that crossed and doubled chromosomes to build up a few octoploids or a decaploid that served as a âtime capsuleâ of rose genetic potential. After the Ice Age, the higher polyploids lost chromosomes as they migrated into less severe climates.
The modern diploids are not exactly like their ancient ancestors, though large sets of shared characters suggest a strong family resemblance.
Studies of chloroplast DNA indicate that the North American species (except Setigera and Minutifolia) there has been very little change since the Ice Age. Strange to say, the authors assume that Asian diploid species flooded into North America, and only later began to build up higher polyploids adapted to the northern regions ⌠through which all those diploids must have migrated.
Systematic Botany, 32(2): 366-378 (2007)
Phylogenetic Relationships in the Genus Rosa: New Evidence from Chloroplast DNA Sequences and an Appraisal of Current Knowledge
Bruneau, Starr and Jolly
Karl
Karl, IMHO figure 3 (p 362) of the article cited by David here gives the best chart, to date, of phylogenetic relationships of rosa.
http://www.rosebreeders.org/forum/read.php?2,15121,15121#msg-15121
Unfortunately it omits some important species including, especially, R. chinensis so you have to extrapolate from other charts to see where they fit.
Extrapolation gives mixed results with these hierarchical models. Iâm hoping that sooner or later weâll see a network model that gives a more valid representation of hybridization than the hierarchical ones do.
Don,
A phylogenetic tree based on nuclear DNA is likely to give very different results than one based on chloroplast DNA. Both are informative, but in different ways.
For example, we may assume that âBasyeâs Amphidiploidâ inherited the cDNA of its seed parent, but nuclear DNA from both. Therefore it would be located in different positions on the respective trees.
Nuclear DNA will usually confuse the relationships of species when polyploids are compared with diploids. There are very few autopolyploids among roses, and these are typically garden varieties (e.g., Rosa macrophylla var. Korolkowii). The polyploid species are allopolyploids, combining chromosomes of two or more distinct diploid species. If the DNA tests cannot isolate the chromosomes of a polyploid (which they canât, so far as I know), the results will be deceptive: âBasyeâs Amphidiploidâ would represent a form intermediate between R. abyssinica and R. rugosa, indicating that those species must be closely allied.
âScientificâ tests are only as good as the assumptions made by those who use them.
I think it would be amusing to present âBasyeâs Amphidiploidâ to a student as an unidentified species. Then let the student use the latest genetic tools and parsimony analyses to identify the correct phylogenetic position for this âunknownâ. It would be too cruel, of course, to hoax a hapless student. But it might go a long way towards convincing us not to believe every poorly reasoned and virtually incomprehensible scientific paper that gets published.
Karl
The polyploid species are allopolyploids, combining chromosomes of two or more distinct diploid species.
You stopped me cold here, Karl. Can you tell me why you say this?
Don,
Polyploid varieties (or autopolyploids) gain no apparent advantage over the original. They are commonly larger, and maybe later in flowering and ripening fruit, but otherwise they gain little. Furthermore, an autotetraploid from a diploid parent tends to have reduced fertility due to tri- and quadrivalent groupings of chromosomes during meiosis. No obvious advantage to this, either. Autotetraploid Rosa maxrophylla var. Korolkowii is preserved in gardens, but has not overtaken its diploid parent in the wild.
Allotetraploids formed from two distinct species (also called amphidiploids) often gain much. They gain considerable adaptive flexibility like ordinary hybrids, but without segregation. Newly formed allotetraploids are not as âhomozygousâ as Hurst and other mendelianists assumed. Insertions, deletions and translocations create considerable genetic variation, and epigenetic changes in gene regulation provide even more. Experiments with Brassica allotetraploids revealed that the changes continued throughout the 5 generations covered in the report.
Allotetraploids are commonly more adaptable than either parent, and are able to spread into regions where the parents cannot. The same applies to allohexaploids, and allo-octoploids.
Allopolyploids may sometimes resemble related diploids, and even be confused with them. as some have confused Rosa laxa Retz. with R. beggeriana. The plant Hurst knew as R. laxa Retz. was a tetraploid that combined the groups of traits he had identified as âSeptet Dâ and âSeptet Eâ. R. beggeriana, on the other hand, was simply âSeptet Dâ.
It is worth noting that common bread wheat, Triticum vulgare, is a hexaploid containing the chromosomes of three distinct species: one Triticum and two Aegilops. Tetraploid species of wheat are similarly auto- rather than allo-. I am not aware of any autopolyploid wheat that has been cultivated.
Karl
So, if one wanted to try chromosome doubling experiments on seedlings it might be wise to use open-pollinated seed from a diploid rose that is self-incompatible and planted near other diploid roses that are not closely related, right?
That way most successes would be allotetraploid rather than autotetraploid. And youâre not wasting effort killing off seedlings from controlled crosses.
Joe I am glad you got your head around this one, I only read with interest.
Very good point Joe, but tis not that hard to make a lot of controlled crosses if the two parents donât suffer incompatiblity with each other, though its even easier if they are self-incompatible. Sort of like male sterile hybrid corn. You just dust pollen from one on the other. No need to emasculate or remove petals. My Carefree Copper is totally self-sterile so all I have to do is smother it in some kind of pollen.
So make the widest cross you can, then double it. For instance seeds of Ralph Mooreâs Pink Clouds might give roof top climbers. I just harvested seeds of a putative cross of R soulieana x Rainbow K.O. but donât know yet if they are diploid triploid or just R.s selfs.
I think Karl may have gotten allo and auto reversed in his next to last sentence. Wheat is a combination of different species of grasses, not a doubling of some diploid, as he says elsewhere in the messageâŚ
Or autotetraploid x different enough autotetraploid = allotetraploid as was done for Lilliput irises.
Every time I mention species roses that are meant to have little or no self-fertility, based on the published work of others, someone mentions one they know of that isâŚ
I also think doubling open-pollinated diploid seedlings is of little value. Even if you do plant it next to a diploid there is no guarantee the successful pollen is from it. The resulting seedlings might well be from another tetraploid around the place and the seedlings might in fact be triploids that would result in hexaploids when doubled (theoretically). Iâd be doing as Larry suggested and smothering the supposedly self-sterile species with known diploid pollen to be sure and then doubling it.
[quote=SimonV]
Every time I mention species roses that are meant to have little or no self-fertility, based on the published work of others, someone mentions one they know of that is⌠[/quote]
Simon,
Thatâs the truth. We really need more information on the natural history of roses, such as preferred soil type. Some species arenât happy or fertile on clay, but seed freely on sand. Others favor clay with not too much humus. The matter gets even more complicated with derivatives of complex hybrids, and some of the older rosarians attributed all the infertility (or reduced fertility) to âgenetic incompatibilityâ.
If we cross two species that like clay, chances are vary good that the offspring will also do well on clay. But if one of the parents is thriving on clay only because it is grafted onto a clay-loving rootstock, the soil preferences of its hybrid offspring will be less certain.
And all of this remains rather mysterious because the information we need has been concealed from us by the wide-spread practice of budding.
Karl
Partially, yes. Climate plays as large a role in fertility in many cases. Mermaid didnât set seed in inland valley heat, but sets freely two blocks from the ocean here where it is up to fifty degrees cooler on the same day and the moisture is constant much of the year. Iceberg does similarly. In the inland valley heat, I virtually never find hips on the hundreds I care for. Put them where itâs foggy, cooler and damper and most blooms form mature fruit. You also have the effects of heat on the actual flower parts. Rosarium Uetersen in high heat, is so fully double, has almost no sexual parts. In the cool, fog belt, it is semi double with a working stigma and full of anthers, stamen and very viable pollen. Iâd often read of it being used in breeding and had always seen it in high heat. I couldnât figure out HOW anyone accomplished that, until I saw plants in the cool, coastal weather.
1 Like
There are many factors that can influence fertility. Years ago, possibly on Chez Vibert, someone commented that Rugosas refuse to fruit in the âfog beltâ of northern California. And Linnaeus observed that many alpine plants become sterile when brought down to sea level.
Hibiscus breeders struggled with the mysterious appearance of sterility among the progeny of complex hybrids, until someone sorted out the problem: trace minerals. Some species are native to coral atolls, some to volcanic islands, and others to fragments of the continental shelf. Each type of island has its own chemical composition, the the native species must adapt or die. Some of the micronutrients are key to pollen or ovum production, but an atoll species may require a different trace element than a related species growing on a decaying volcano. The nutrient needs segregate in later generations, leaving breeders puzzled ⌠until they learn the origin of the problem.
The famous Cuzco Corn, the one with kernels an inch wide, grows too tall in most of the U.S., and many stalks are barren. But in its native home, high in the Andes, the plants are fertile, productive, and no taller than most of the corn grown in the U.S. today. Another case of temperature affecting fertility.
Karl
1 Like
[quote=Don]
The polyploid species are allopolyploids, combining chromosomes of two or more distinct diploid species.
You stopped me cold here, Karl. Can you tell me why you say this?[/quote]
Hereâs another, perhaps more persuasive argument from Ake Gustafsson, Differential Polyploidy within the Blackberries (Hereditas 25: 33-47 (1939)), found that tetraploid blackberries occurred only where two or more diploid species coexisted; never among an isolated population of a single diploid species.
He also observed that some crosses (diploid x diploid) are more likely to result in polyploid formation than other such crosses. Again, the reason is differentiation of chromosomes. This also happens with roses. Diploid teas have been crossed with various Synstyles (multiflora, moschata, wichuraiana, setigera) with no rush to double up their chromosomes. On the other hand, rugosa crossed with Synstyles more often results in tetraploids, sooner rather than later. E.g., âBasyeâs Amphidiploidâ, and a true amphidiploid from âMax Graffâ.
Chromosomes tend to be attracted to their own kind. The Synstylae and Indicae species have generally similar chromosomes (with similar genes in most of the same places). This allows comfortable pairing of chromosomes, and a high degree of fertility. Rugosa, however, apparently has enough differences in its chromosomes (relative to the Synstylae/Indicae) to weaken pairing, and encourage the formation of unreduced gametes. These lead to triploids and tetraploids.
Teas/wichurana may line up well but I have done wichurana x âMonsieur Tillierâ and âImmenseeâ x âMonsieur Tillierâ extensively and every seedling has always been weak, not survivng more than a few weeks after germination.
I have some plants of âPaul Noelâ and Iâve just received âJersey Beautyâ. I think it might be fun to try and double âPaul Noelâ as it seems to have lousy fertility.
I recently came across some further information on alloploidy in roses.
Annals of Botany 85: 557-561, 2000
Nuclear DNA Amounts in Roses
Yokoya, et al.
âIn the Pimpinellifoliae, DNA amounts of tetraploids were disproportionately larger than those of diploids which suggests that they originated as hybrids with species of sections with larger DNA amounts.â
I looked up Hurstâs âseptet formulaeâ for the roses listed, and noted that two other species, R. bella (BBEE) and R. fedtschenkoana (BBDD) have 2C DNA amounts agreeing with the tetraploid Pimpinellifoliae species.
Hurstâs analysis of numerous traits (around 100 in the first studies) associated R. gymnocarpa and R. willmottiae with the diploid Pimpinellifoliae species (e.g. R. sericea, R. xanthina).
It is interesting to note that R. persica and R. stellata have nuclear DNA in the same range as Hurstâs BB diploids. It is also notable that Harkness (1977) wrote:
http://bulbnrose.x10.mx/Roses/breeding/Persica/PERSICA.HTML
"In 1970, we had 33 hybrids germinate, of which 27 were H. persica x Canary Bird and 6 were > H. persica > x Ballerina.
All the Canary Bird crosses survived, providing a most interesting range of variations in this cross. All had 5 petals, all had shrubby growth from 3-6 ft. tall, all flowered in May, and all had leaves and stems showing Canary Bird influence. We observed interesting variations in the red centres and in flower sizes."
This does not prove that Persica is more closely allied to the diploid Pimpinellifoliae. It is at least possible that the low nuclear DNA amount is an adaptation to local conditions. And that the success of the Persica x Canary Bird hybrids had more to do with shared physiological adaptations than taxonomic affinity. Still, it is interesting.
The practical implications of differential chromosomes and allopolyploidy is that crosses between species with chromosomes that pair regularly will allow freer reassortment of traits that tend to hang together in crosses where the chromosomes donât pair properly,
For example, hybrids of R. foetida and R. fedtschenkoana (both BBDD in Hurstâs analysis) should allow recombinations, and allow breeders to avoid linkages that come when Foetida is crossed directly with HPs, HTs and so on. One might get yellow that is not stained with red.
Thank you, Karl! Very interesting!
âFor example, hybrids of R. foetida and R. fedtschenkoana (both BBDD in Hurstâs analysis) should allow recombinations, and allow breeders to avoid linkages that come when Foetida is crossed directly with HPs, HTs and so on. One might get yellow that is not stained with red.â
If youâre lucky. My experience with what Iâve paired with Fedtschenkoana, you are lucky not to get blush pink to white. It seems to bleach out many petal pigments much of the time.