Inbreeding Species and Modern Roses.

Would someone with a Science Degree in Genetics please explain why rose breeders don't like inbreeding when self pollination is happening all the time with little, if any, sign of inbreeding degeneration. Mendel grew beans or peas showing combination of genes then segregation in definate numbers. These F1/F2 from what I read were viable healthy plants?

I have inbred poultry to a point where they should have suffered inbreeding degeneration but did not due to breeding larger numbers and selecting the most viable. Rats being bred for science have not shown inbreeding degeneration after numerous generations of close breeding. This promped me to look at rose breeding in a similar way. Some of the good roses on HMF show a common seed and pollin parent.

Please excuse my backyard genetics etc.

The reality is, Bruce, that most of the bigger rose breeders, who say they don't like inbreeding, have been doing it for years and anyone who is hell-bent on breeding more hybrid teas and floribunda are doing it too.

If you trace most modern roses of the hybrid tea and floribunda type back along their family tree you can trace them back to only a handful of stud roses (like 'Fashion'). Breeding these together over the years is just like stirring the same old pot of genes to make a slightly different permutation of the same old thing.

The reduced health and fitness is probably a clear indication that modern roses are showing signs of inbreeding depression due to breeding from within this same pool of genes for hundreds of years by the millions. The more you breed from within the gene pool the more deleterious recessive genes can become homozygous in individuals and the more the fitness can be reduced. With careful selection, this can be avoided, to some extent, because those deleterious recessive genes (not all recessive genes are deleterious) can be selected against and removed from the population. This is complicated in roses by polyploidy meaning these deleterios recessive genes can take a really long time to resurface.

Plants show a greater tendency to inbreeding depression than chooks do because they can self-pollinate and, you are correct in saying that self-pollintaing occurs all the time but many plants have got prefferential strageties in place to prevent it. In roses, many species do not self easily and they achieve this by ripening the various parts at different times. The anthers ripen first so the pollen is released and then once the pollinators have been in and removed the pollen the stigmas ripen and become receptive. These selfing barriers break down as hybridity increases (but then so does fertility in many cases). Your chooks can't 'self' so there is always cross-fertilisation invovled.

Using species is not inbreeding. There is a certain amount of inbreeding that can take place but it might not always be in your best interest to do so. You want to maintain the hybrid vigour and the effect of the species genes (in theory). I have done a lot of crosses with Rosa wichurana this seson. Repeat flowering, in general, is a recessive trait. In wild populations this might be seen to be a deleterious trait because to repeat flower requires more resources and so the plant would be less able to compete with once-flowering roses in their native habitats because resources are often a limiting factor... only us gardeners think it's pretty good... though there are examples of repeat flowering species roses. All of my crosses will be once flowering (though I might fluke a repeater if I'm really lucky). I will need to either inbreed these individuals to bring the repeat back out, cross the seedlings of this generation together to bring repeat back out or out-cross to a repeater. This is why ploidy is so important in selecting breeders but that is a topic for different discussion. In the last case, if I out-cross to a repeater then I am diluting the species contribution. Of the first two options, the second is the better one because the variation in the offspring will be greater. Every time I out-cross to a species I will have to do follow this pattern for remontancy. Back-crossing species is not inbreeding.

So, Bruce, to answer your question, once you get to a point in a population where you have removed the deleterious genes, inbreeding depression doesn't become an issue. This is, however, next to impossible in roses (more so than chooks) when you take polyploidy into account.

An interesting example of a deleterious recessive gene showing up through in-breeding (cousin marriages are bad news!) has happened a couple of times with crosses between Oklahoma and/or Papa Meilland and/or President Lincoln, which all have the same parents. Some seedlings produced a blue flower and promptly died. Something to do with an inability to use a nutrient but I've forgotten the details.

Simon, recently you brought up a breeding experiment, crossing some Tea's with HT's in order to breed roses that look like Tea's but have blooms with more intense colour. That looks to be a very worthwhile breeding experiment. The way you were going to do this was to keep coming back to the Tea with the first cross that showed best colour.

I would have used Mendels same as the pea experiment to make my Tea's with deeper or more intense colours. I base this on having used this method before with chooks to breed a Mediteranium breed colour into an Asiatic. I have never been able to achieve a good result with back crossing. Polyploidy is new to me though I have tried to inwardly digest information posted on this subject.

Please excuse my reference to chooks which I have been breeding for 70 years though I do want to get on top of rose breeding genetics.
Thanks for your lengthy post and information above.

Margaret, what a wonderful marker, all the blues within that family of roses are duds. In some of the small villages (I won't mention where)
everyone is related and when a family gathering like a wedding is happening everyone goes. There are cousins marrying cousins with very few signs of inbreeding. The rule is no two persons with the same name are allowed to marry. One would think they are closely related anyway but there is a reason why the maleis most important in keeping them healthy.

Sorry, back to rose genetics which fasinates me though may be of no interest to others.

That sounds like the "power of the pollen" myth in rose-breeding ... Which possibly reflects that most rose-breeders have been male!
Unfortunately even if harmful recessive genes could be "bred out", and I don't think they can (we have an average of 20 each, from a vast array of possible ones), new mutations would keep happening.

To use a slightly different example, Bruce, to illustrate my thinking of backcrossing a number of years ago I developed a breeding program for rabbtits (normal diploid organisms like you and me) to develop a breed I was calling the Mini Plush Lop.

Rex is a coat type that reduces the length of the guard hairs down to the same length as the soft undercoat. The effect is a coat that has the look and feel of velvet. It is controlled by a standard, Mendelian, recessive gene.

At the time I was breeding and exhibiting a miniature type of rabbit called a Mini Lop. They are small rabbits with big round heads and their ears hang down instead of standing up. Their coat is a normal rabbit coat of the type we call roll-back which refers to the manner in which the hair returns to its orginal position when you draw your hand up against the direction of the fur. The guard hairs are twice as long as the undercoat (at least).

What I wanted to do at the time was to pick out that single rex gene and put it into the Mini Lop type. Others had tried to do this before and failed badly because they were trying to do the crosses too quickly and because they didn't get the results they wanted in two or three generations they gave up.

The Rex rabbit is a large rabbit who lays down when standing because it has a longish body. The Mini Lop stands up and has a shorter more compact body with a large round head, a wide crown, and ears that flop (or lop).

Getting the rex gene into them was easy enough but getting back to Mini Lop type where the ears lopped etc was the hard part. So what I developed was a plan to concentrate the Mini Lop type genes whilst retaining the rex gene because really, the only thing I wanted from the rex was that gene. I figured that if I kept going back to the pure Mini Lop then I could dilute the Rex component and concentrate the Mini Lop component. I was able to get almost all the way back to Mini Lop type again in just 8 generations by using the following theoretical gains stats:

1st cross: 100% rex(R) x 100% ML 50% rex:50% ML (normal coat)

2nd cross: inbreed so no theoretical change in proportions of R:ML because you aren't introducing any new genes but it will bring the rex back out. Progeny will still be 50%R:50% ML. Of course, these figures are only theoretical and better than expected gains can be made in the 2nd cross by selecting the best exmaples to breed with.

3rd cross: 50%R:50% ML x 100% ML. During formation of gametes the material theoretically divides in half. The 50% R: 50%ML would form gametes that are 25%R:25%ML and the 100% ML would form gametes that would be 50% of their total completment. Combined in the next generaion you would get progeny that are 25%R:75%ML.

This pattern is repeated each generation and every two generations you get a theoretical 50% reduction in the rex component and a theoretical 50% increase in the ML component and by generation 8 the progeny were theoretically 93.75% Mini Lop and only 6.25% rex and are pretty much indentical to the mini lop type. By this stage they were true breeding.

In the end I abandoned the project because I felt the shorter hair took away from the stocky fluffy features of the Mini Lop, but I succeeded with this program in a short amount of time by just using a dilution principle of continued back-crossing.

It was this rationale that I was applying to the Tea roses x moderns. We know that in most areas in Australia, Tea roses are far better performers than the modern roses. There are features in modern roses that are good. Take the cluster flowered trait if floribundas for example. If I wanted to pick out this feature of the floribunda and then incorporate this into the Tea without losing all the positive features of the Tea, then I would go Tea x floribunda and then keep going back to the Tea and each time select the progeny with the best cluster flowered traits and the best Tea traits until what I am left with is essentially a straight Tea with the cluster flowered trait. Ploidy confuses things and so it isn't that straight forward (you just don't know what is lying in the recessive gene pool of roses which is why roses seedlings show so many hideous variations even from crosses of the best roses) but I think the principle, combined with clever selection, is the key to the success and if in one generation you don't get what you want to be able to proceed, then you keep going back and repeating the cross until you get what you want.

Another complicating factor in roses is that a lot of the dominance patterns of their genes are not known and many of the genes are not straight forward Mendelian ones. You have partial expressions and other gene interactions that make it very hard to plan. I believe all you can do is make a plan and then try it and apply the selection process ruthlessly and keep going until you get what you want. If it was as easy to do as with the rabbits, everyone would be doing it

On a chooky thing... I have managed to breed a cool baby this year in my flock of muts. It's a red and black barred baby. Like a plymouth rock butthe white has been replaced by red. It's only expressed on the wings and saddle so far with lovely red hackles (it's a pullet I think) like a campine sort of, but this is significant because barring and white are on the same chromosome which means that crossing over must have occurred to replace the white with red... very pretty IMO.

Simon, thanks for your time to explain. Must mention I didn't think of useing Floribunda's to the Tea's. Thought the T's would stamp this trait into it's offspring but useing Floribunda's this should keep the Tea floribunda trait. However nothing's certain and suggest to all to have a go at breeding roses. It's 50% of the pleasure with having roses.

One of the problems with Back Yard Genetics is not being able to express breeding plans and happenings clearly. Hope this didn't put readers off.

I should also mention that the Tea x floribunda was a hypothetical example that I jjust plucked out of the blue and is not something I am necessarily thinking of doing.