Growing roses from seed...

Have you ever looked at the newly released roses in a mail order catalogue and wondered where the new releases come from and how they are made? There seems to be no end to what kind and colour of rose they can make! Large-scale breeders, like the UK’s David Austin, will painstakingly choose thousands of varieties with which to perform crosses. From those more than 350,000 seeds will be sown each year and after 8-12 years of exhaustive trials they manage to cull this number down to maybe 5 roses to release each year.

You and I are never going to perform this many crosses or plant this many seeds so our chances of breeding the next big thing in roses is... slim. However, the truth of the matter is that ANYONE can grow roses from seeds and it isn't as hard as it sounds. You will need to be patient but it will all be worth it when you see the first flower open and think WOW! I made that! For various reasons that I will discuss briefly later in the article a lot of rose seedlings will fail, however, there will always be a few in every batch that will want to keep. As they grow you will be rewarded by many beautiful displays for many years and loads of free plants that no one else in the world will have because every rose seedling grown is unique.

Rose seeds develop inside fleshy structures called hips. Hips range from a yellow/brown colour when ripe to brilliant glossy reds and even shiny round and black as in Rosa pimpinellifolia. Most hips are ripe and the seeds mature when the hips have coloured up fully. At this point it is important to note that seeds from a rose will NOT breed true. The seedlings will not, unless they are true species roses, look like their parents for reasons I will explain later.

Rose seeds are not really seeds. They are structures called achenes and the seed is found within these structures. For the purpose of this article however, it is sufficient to call the achenes seeds. Many rose species grow naturally in areas that experience very cold winters and this added protection, and other physiological changes, helps them survive these winters. In fact roses have evolved to a point where the seeds actually require a period of intense cold to break the dormancy of the seed before it will germinate. We can mimic these conditions by putting them in the fridge for a period of 1-3 months followed by warmer temperatures that signify to the seed that spring is here and germination can now begin safely. This process is called stratification. There are many ways to stratify seeds but for a small number of seeds I find the following method works well;

1. Mix up a fungicide such as Mancozeb as per the instructions and soak the seeds in the solution for a period of half an hour or so. Your main problem during stratification is fungus though mouldy seeds may still germinate. Some have said a little mould actually improves the chances of a successful germination as it weakens the seed coat enough to allow moisture in to initiate germination. Soaking the seeds in a fungicide will help to reduce the risk of fungus destroying the seeds but don’t be too concerned if a little bit grows. Some will tell you that floating seeds contain air spaces inside instead of a seed germ and so will never germinate. This, however, has been shown to be an unreliable means of telling good seeds from bad and I don't bother looking any more. It doesn't take any more effort to stratify all of the seeds than it does to pick out and discard the floaters.
2. While the seeds are soaking moisten some paper towel with some of the mancozeb solution, squeeze it out and (wear gloves), flatten it out again and fold it over a few times.
3. Strain the seeds and lay them on the moist paper towel, fold the towel over the seeds and place the paper towel into a plastic bag. Seal it with a kitchen tie and write on the bag the date and the parent(s) of the seed.
4. Place them in the fruit crisper of your fridge and leave them for 1-3 months. This period of time varies between varieties.
5. After about a month begin checking the seeds. What you are looking for are seeds that have started to germinate. You'll be able to identify these as they have the beginnings of a root working it's way out through the seed coat.
6. If there are none just fold the paper towel back up and stick it back in the fridge and check it again in a few weeks time. Be aware that every time you touch it and open it you will introduce fungal spores and possibly contaminate your seeds.
7. When the first seeds begins to germinate I remove all the seeds and sow them on ordinary seed raising mix and cover them to about their own depth with fine moist potting mix so they can continue to germinate.


Here is some pictures of what I am referring to from this seasons seedlings:

Germination has begun

[You must be registered and logged in to see this link.]

The seedling emerges and unfolds its cotyledons. This particular one was a little strange in that it had three cotyledons instead of the normal two.

[You must be registered and logged in to see this link.]

As the seedling grows it will develop its first leaves:

[You must be registered and logged in to see this link.]

And in next to no time they will look like this:

[You must be registered and logged in to see this link.]

In just a few short weeks the roses that will become repeat flowering roses will have developed their first buds. Roses that take up to a year to get their first flower are probably not repeat flowering forms.

[You must be registered and logged in to see this link.]

And then they will open. Here are three of my best from this year.

As a general rule it is difficult tell what the flower will look like on the mature plant from the first flower because the numbe of petals will generally increase until about its second year.

[You must be registered and logged in to see this link.]

The fact that these first flowers look as good as they do is a very good sign. Now They will go onto testing for vigour and disease resistance and next spring will be planted out in a bed to see how they perform.

I call them; 'Kindred Spirit', 'Kindred Gold', and 'Kindred Ice' respectively. 'Kindred Spirit' has the most delicious perfume

I am a contributor on the American Rose Hydridizers Association forum and 'Kindred Ice' caused quite a bit of interest as it appears that this is the first time anyone has ever been able to successfully germinate seed from it's parent 'Green Ice'. It's a strange tiny flower and I am very excited about it.

Now, this brings me to how they are made!

Hips can be made in two different ways; they can be open-pollinated or they can be hand-pollinated.

Open pollinated hips are made naturally by either self-pollination (it is pollinated by its own pollen) or cross-pollination by another means such as a bee from another plant. There is no way practical way of knowing whether it is self-pollinated or cross-pollinated if the flower was open pollinated. It's an exciting way of doing it because you just never know what you will get. Your biggest test will be to refrain from deadheading your roses and allow them to develop hips. Hips will ripen towards the end of autumn to the beginning of winter but take your cue from the colour of the hip and harvest them when they are fully coloured up as some varieties will mature faster than others will. Most people will start like this and it can be very rewarding. The three roses shown above were from open pollinated hips.

The second way, hand pollinating, requires that you remove the pollen from one of the selected varieties (called the pollen parent) and place it on the stigma of the another variety (called the seed parent because it will be the one that develops the hip). You can't just do this with any flower. It has to be a young, almost open flower and you need to pull it apart! This is called emasculation. This involves removing the pollen producing anthers from the seed parent before they have matured enough to release their own pollen avoiding self fertilisation. You must then collect enough pollen, on a small plate or bowl or in a small plastic bag, from the pollen parent that you can 'paint' onto the emasculated flower's stigma to pollinate it with a small paint brush. Repeat this many times to improve the odds of successful hip-set.

Once pollinated you don't want any other pollen to land on the flower so a physical barrier of some kind is needed to prevent this. Many people use a small plastic bag that is slipped over the emasculated flower and lightly tied on so that air can still circulate but insects carrying pollen and rain can be excluded. Add a label with the seed parent’s name, pollen parent's name and the date it was done so that when the hip is mature it can be snipped off inside the bag and stored until the seeds can be removed. Some use aluminium foil wrapped to wrap around the maturing hip but Australian summers make this problematic. The hips would quickly bake in our extreme summers and kill and developing seeds. Careful note taking is necessary when making planned crosses for many reasons. On such reason is that should you be lucky enough to breed a rose of outstanding beauty worthy of being released you can then register the parentage of the rose as well for the benefit of future breeders trying to unlock the secrets of breeding roses. For nerdy science-types like myself examining the pedigree of a rose (or any other plant or animal species for that matter) can reveal the very nature of inheritance to help make rose breeding more predictable.

Rose genetics is an interesting field of research and more is being learnt all the time about how different features are inherited and expressed. As a hobbyist it is important to understand that the raw products you are dealing with (i.e. the pollen and the ova) possess an enormous amount of natural variation. This is due in part to the complex blend of genes that make up the modern rose and in part to the random processes involved in making the pollen and the ova. It is for this reason that modern roses will NOT breed true and why seedlings from open pollinated roses will look vastly different to their parents. Experienced rose breeders use terms such as diploid, triploid and tetraploid to describe the complement of genes a rose may possess and how it will affect the overall success or failure of a particular cross. We are still a long way off knowing how genes are inherited, what the genes are or what their effects are so for now, at least, rose breeding will retain a certain amount of mystery and unpredictability.

My own breeding goals are simple... I want to breed my own roses that will grow strongly on their own roots and not need grafting as most modern roses are. By carefully selecting the strongest (and most pleasing to the eye) each generation this goal can be achieved. If you are planning crosses, as opposed to open-pollination, you need to set your own goals and practice ruthless selection if you are to achieve these goals and above all be patient because it is a very slow process…

Breeding roses can be a hit and miss kind of thing and this is largely due to the rather mixed up heritage of most modern roses. Modern roses are hybrids in that they are the result if crossbreeding many different species or wild roses and the more crosses that are performed the muddier the gene pool becomes. So when your seeds begin to germinate you will get runty little roses that never make it past a few leaves, to straggly looking briar roses but then you will also get ones with good combinations of genes to grow roses of amazing beauty. So give it a go. You never know. You might just breed the next big thing in roses

Hi Simon, I can see my rose garden growing now!!
Great topic this one. I can't wait to hear the outcome of how well they perfom in the ground.

Simon, I think the big breeders are trying to frighten the hobby breeders away by listing such a poor success rate. They should embrace us and offer to buy the rights to some of our hopefully outstanding seedlings. Roses have such mixed up genotype so think the aim should be to breed good breeders which I hope will come in the later generations of seedlings. I may be wrong but there is so much crossing different types with a low number of similar genes. This I feel is the reason why the success rate is so low. The speces roses are very pure as they are the result of generations of repeated self pollination and possible crosses with another similar species growing close. Nature then has it's way by survival of the fittest so vigour is still excellent despite repeated inbreeding.
What happens then, man takes these roses out of their best environment and thats when all the problems start.

Something for you to ponder over is the rat breeder I was friendly with for many years.
He was in charge of a closed rat breeding station in Sydney funded by the NSW Govt.
The aim was to breed rats without bringing in new blood suitable for scientific experiments.
The rats had to be the same from one generation to the next. Some experiments were 2 or 3 years apart but the rats had to be the same types for vigour, size and from the same genotype. He was also doing 3 to 4 generations per year so after 12 years that I knew him some 40 generations were bred. Thats a larger breeding program than cattle, horse, sheep, poultry or bird breeders could do in a lifetime so his work was so interesting for me as a poultry breeder. The rat breeding complex was backed up with a lab where embrio transplants were done so its a magnificently run operation. After a few generations the rats seemed to suffer inbreeding degeneration but enough desireable ones were reared to meet their needs. By selecting these for breeding , the inferior ones started to dwindle and future generations were, some not as good, some suitable for experiments and some larger with to much vigour so they could not be used. His work fasinated me and when I asked him about numerous times inbreeding degeneration he claimed that you can breed past that with proper selection of breeders. Unfortunatly the govt closed this program down and the staff have retired or sought other employment. Its a very interesting experiment and perhaps we can breed more reliable rose breeding plants with a bit of good selection.

G'Day Bruce,

One thing you need to take into account when looking at roses, compared to other living things such as rats, is that roses don't follow the same 'rules' as do higher animals. When rats breed the first thing that happens is that their entire genome randomly halves giving rise to what's called a haploid gene count and given the symbol 'n'. Before it halves the full compliment is called the diploid gene count and is given the symbol 2n. When fertilisation occurs the egg, containing half the genetic material from the mother, combines with the sperm which contains half the genetioc material from the father and in doing so they make a full set again. This is repeated over and over again for many generations.

This does not happen with garden roses. It does with species roses. Species roses are mostly diploid. That is they have half their compliment of genes from one parent and the other half from the other parent. For species roses the 2n number if 14. That means that in each somatic cell (body cell) in the rose you will find 14 chromosomes. The haploid number is therefore 7 (so n = 7 and 2n = 14). When you start making hybrids there can be errors introduced in how the genes segregate and divide and instead of dividing normally whole sets can go into sex cells instead of dividing in half. The resulting seedlings can be triploid (3n = 21) as in most polyanthas, tetraploid (4n = 28 ) as in most garden roses, can be anuploid which means that instead of whole copies of the genome being duplicated only certain parts of it are, pentaploid (5n = 35) and I have heard of hexaploids (6n = 42). It creates very difficult set of circumstances when it comes to trying to predict what results to expect when certain crosses are made. Normally, in a 2n individual, you will have dominant genes and recessive genes. You might have other cases such as incomplete dominance or co-dominance, or epistasis etc but overall predicting the outcome is somewhat easier because you only have two versions of the gene at any one time. To illustrate what I mean I also breed show rabbits. One of my goals a number of years ago now was to develop a line of albino rabbits that contained nothing but recessive colour genes so that when I mated any other rabbit to it I could see the hidden genes because the recessive genes in the recessive parent were not capable of masking the other genes. To complicate things there can be multiple versions of the same gene. In humans an example is eye colour. There is only blue and brown genes (every other colour is a shade of either blue or brown), but in rabbits, there can be up to 6 versions of the same gene which have an order of dominance at each locus. Now I know this stuff inside out and can breed rabbits and know exactly what I'm going to get and part of this predictability is because they are diploid. But in a tetraploid rose (4n = 28) at each locus (gene location on the chromosome) there is four versions of the same gene and each one may be different! This means that modern roses are an incredibly diverse group of living things. It's almost like they collect genes!!! Now to give you an idea of the breeding capacity of a rose let us consider humans first. Humans have a diploid number of 46 which means our haploid number is 23. When sex cells are made they undergo two divisions and each time they divide they do so randomly. Taking all this into account statisticians have calculated that a single a male can produce sperm with about 8 million different gene combinations. That means a single female will also be capable of producing about 8 million different gene combinations. It follows then that the combined probability of one man and one woman producing two genetically identical children is in the ball-park of 1 in 64 trillion. That's a huge number of possibilities. Now let's look at a rose. A diploid rose has a haploid number of 7. That means it can produce something like 3.3 million different combinations of pollen or ovum. This means that any two plants could be cross pollinated more than 9 trillion times before getting a genetically identical offspring. So when you start adding triploidy, tetraploidy, pentaploidy and hexaploidy into the equation the odds become enormous.

So this little jaunt into statistical nirvana ( ) shows a number of things... first we cannot farely compare rose inheretance to animal inheretance. Second, roses have an immensely complex gene pool that has enormous diversity. One could spend their entire breeding career self pollinating just one rose! If that rose was tetraploid as most garden roses are, then you will have to breed them more than 100 trillion times before you have exhausted every genetic combination. You will reach a point of diminishing returns when the differences between the progeny will become more and more subtle and eventually only be different at the biochemical level, but that's how disease resistance can be introduced. If you do the same cross enough times you are likely to come across one sooner or later that has the particular gene combinations for disease resistance... so long as those genes we present to start with! You can increase your odds by starting with known examples.

With rats, and any other animal, you can almost elliminate the effect of a inbreeding with careful selection because degradation only occurs when deleterious genes are expressed and if these are selected against (i.e. removed) and the population becomes more homozygous (the same at the gene level) then, unless mutations occur, the only genes left will be good ones. This becomes harder with roses because of polyloidy... they say that a recessive gene can lie dormant in poultry for more than 11 generations and poultry are normal diploid animals. In polyploid roses this means that recessive genes can remain hidden for much longer. Consider this hypothetical example (based on fictional genes). Lets say that at the fictional flower colour locus a tetraploid rose has four genes and they are all different. They are red, pink, white, and yellow. Now lets say the the most dominant of these is pink followed by red, then white, then yellow (keeping in mind that I am making this up as I go LOL). If we mate this to another tetraploid rose how will we know which genes have been removed and which ones have been preserved???? It's a nightmare I tell you LOL.

In practice we can, however, use the HMF website to give us an idea of what makes a good breeding line. Let's take the miniature 'Rise n Shine' for example. This is one reliable breeder in terms of fertility both as a pollen parent and a seed parent. If you look on HMF and look at the descendants list there are more than 700 direct descendants. That pretty impressive. There are 140 in the first generation alone. These numbers a dwarfed by varieties such as peace. So you can get some idea of where to start to make breeding lines by analysing the HMF lineage data. In fact I think this is the single best tool available to rose hybridisers to allow them to select good parents from the classes of rose they choose to work in. The seocnd best tool is forums like this where breeders can ask each other how different roses perform for them in different locations.

It's also good to look at these results for more than one generation to see if these fertility traits are passed on.

WOW... this is getting too huge... I should stop now... I think it should be pretty clear now though that rose breeding is a lot more involved than animal breeding and it comes back to being a numbers game and how to try and play the numbers to our advantage.

Wow Simon, that was really interesting, and amazing too. I have a greater respect for rose breeders now, it's such a hit and miss hobby or interest, call it what you will. The successful rose breeders have achieved great things, and each rose on the market has a long history of trial and error. Thanks to your input i now have a clearer picture of the whole subject - not that I'm pretending to understand it all, not by a long chalk! Thanks again.
Val

After seeing Bemo's seedlings from Bonica and how different they can be from the parent plant I have refrained from deadheading my roses so they can form hips. Today I removed two hips from "Marijke Koopman"and extracted 28 large seeds and a few small ones which probably are to small to be viable. Can anyone tell me why put them in the 'fridge ? Although we have put some rugosa seeds in the 'fridge some in paper towel and some in vermiculite - both in plasic bags - just to try it. I cannot understand why they would germinate in the cold. I have a heated propergation bench with auto' misting and a range of propergating materials e.g. perlite, vermiculite, sand and fine potting mix. Any advice would be appreciated as before we just put the seeds into potting mix and hoped for the best. Carole.

Hi Carole... going to move this to its own thread if that's ok

That's fine with me

Thanks again big time, Simon. I will buy that stuff you mentioned above to soak the hips in as soon as I can get it from Mudgee.
Might even collect some open pollinated seeds to do too.

In nature does not the hip fall to the ground or a bird takes a seed and the process then starts. More on this tonight, have got to go and earn a quid.

Facinating subject, Simon. Congrat's on your success with Green Ice.

Betty

Hi
I was wondering if anyone can help me.
Maybe i just need to be patient too.
I bought some rose seeds
[You must be registered and logged in to see this link.]
I contacted the lady about the seeds on the 7th October 2009 to find out how she did her seeds, and this was her reply.

Hi, each rose seedling is a new, never-before-seen variety so out photo is only for display.

Roses are not difficult to grow from seed, if you do it right. If you just put a lot of seeds in a pot of soil, water it, and wait, you'll likely be disappointed. But by using the procedure explained here, you should be able to grow them.
You can not soak seeds, just rinse and wrap them in a moist paper towel. Place that packet into a zip-loc bag, and place it in the refrigerator (not the freezer). After 4 or 5 weeks, start checking on the seeds every week or two. At some point, you'll notice little root tips poking out of some of the seeds. Carefully transfer these to pots of soil, cover them with about 1/4 inch of soil, water them in, and keep them at room temperature. They should emerge in a few days. Be careful in transferring them since they are quite fragile, and they tend to stick to the paper towel. It's very easy to break the root off of the seed. Put the unsprouted seeds back in the refrigerator for another week or two, and check them again. Depending on the variety, seeds may continue to sprout for several months. So you'll likely make several transfers to warm soil, over that period.

The seedlings can flower in as little as 8 or 10 weeks after planting. They'll almost certainly flower in the first season. However, some seedlings may not flower in the first year,



Now I have done all the above ever since i bought the seeds except i had the seeds in the fridge in the meat keeper, i have now moved them to the crisper, i dont know what difference it makes but i have done it as word correct.
I still have not seen any shoots starting,the seeds are still moist.
Should i still keep them in the crisper for longer as i have been reading it can take years for the rose seeds to sprout.
Or should i sow them in soil.
I am kinda lost as what i should be doing here.
Any assistance would be appreciated.
Thank you

Hi HOH,

Gordon (a man ) is actually a member of this forum and his advice is pretty good. If your seeds have only been in the fridge 4-5 weeks, however, it s probably too soon to have any germonations at all. Some species seeds will germonate this quicky but in general I leave my seeds in the fridge for pretty much the whole of winter (so three months) and even then some don't germonate straight away (this is why I don't throw away the empty seed beds... I just move them to the back of the rack and forget them until next season). Many of the larger modern rose seeds take much longer again to germinate... so just be a little more patient... Plan ahead though because after 2-3 months in the fridge the weather is going to be starting to cool down a bit and your seedlings will not be very large by the time winter comes around. This won't be such a big issue up there in Sydney as it is down here, but you'll need to provide a protected space for them until spring.

Simon, what does Gordon Russell go under on the forum if this is not a rude question.

I'll have to check... It's been a while since I spoke with him...

Hmmmm... seems he's not a member here anymore.. he was a member before it became rose talk.. must have been one of the members who opted out when we went roses only

Hi Simon
I said the lady because in the reply i assume it was from a lady by the reply see below.
>>Kind Regards Elena.
But thank you for mentioning it.
I had no idea they are members of this site.
I am so looking forward to the seeds shooting and then growing into beautiful roses like in the pictures.
It will almost be 3 months on the 7th January, but i will be patient and keep watching.

What did you mean by this please >> you'll need to provide a protected space for them until spring.
Thanks again for helping me
Lu

Don't be too surprised if they don't look like the ones in the pictures. Even when you know what the parents are you can get a very diverse range of seedlings none of them might be worth keeping... that's just one of the things we need to live with when growing roses from seeds.

If you germinate your seedlings now... you will need to protect them from the heat... if you germinate them later in Autumn they will not be very big or grow very much before winter and so they will need to be protected from frosts and cold winds.

Have been reading through all this .
I have a question about sowing depth Simon . I see that you say that you sow them at their own depth , sow do I , was just using the old rule of thumb for sowing any seed.
But my seeds continually migrate to the surface , I'm for ever tucking them under or sifting a little more propagating mix over the top but in a few days time they are back up on top .
Rose seeds don't need light to germinate do they ?

cheers. Rosallie

nope.... mine germinate in the fridge in the darkness of the fruit crisper that's just settling of sediment. You know how if you drop a rock into a bucket of sand and shake it the rocks comes to the surface.. .same concept.

I thought it might be something like that so I have been careful not to handle the pots too much . [only look in for signs of hatching 99 times a day now instead of 100 ]
Currently have some seed from Regensburg and Mabella hatching and keep having to sprinkle some mix over the seeds .
Have has some more Soaring Spirits seedlings emerge only to fall prey to something during the night.
Earwigs , slaters , slugs , dunno. The tops are taken clean of the little stem which remains standing and the leaves are on the ground . Whatever it is , is very about which part of the tiny plant it eats .

cheers. Rosalie

Sounds like slugs... my mortal enemy... get some copper wire and wrap them around the pot. When the slugs and snails crawl over it they create a potential difference... gives them an electric shock doesn't stop 100% but works most of the time... needs to be exposed copper... not insulated.