Yeah, I don't do anything special. If one were to be all proper about it, one would put the seed plant in a greenhouse. Then get the stamen from the pollen parent, go inside the greenhouse and pull the petals from a flower that has barely not opened, carefully remove those stamen and apply the pollen to the stigma. Then cover up the stigma and label the flower.
When I started out, I noticed plants didn't really set seed unless the flowers were hand pollinated. But back then I had a few plants. Now, there are bumblebees zooming about all day. And seed does set on some flowers. Azaleas avoid cross pollination by having a stigma that's extended further out from the stamen. But this not true for all cultivar. I noted that 'Shisen' and sports (I have 'Hakusen') have shorter stigma, and they set more seeds without me doing anything. One could be connected to the other. I can't really say how much these bumblebees are interfering with my crosses. But I like bumblebees.
So I look for a stamen where I see white stuff on the anther, and I just pull it out. I learned that you can pull out 2 at a time easily. But if you pull out more, the force to pull 3 of them becomes equal to the force to maybe pull off more
. Pulling off only the anthers is unpractical, because you have no hold to apply the pollen. The pollen is always inside this sticky matrix that comes out in strings. Best is to use a pincer and pull it out, because using my fingers I usually touch the anther and some of the pollen comes off already and goes on my finger. Then I try to find a flower where the stigma seems fresh/not dried out/moist and without any sign of pollen on there (there almost never is a sign of pollen, for some reason). Best to remove the stamen there first, without releasing any pollen and contaminating the stigma. And I rub the pollen on there, like you say. You apply some force, rotate, and look if strings are coming out or not.
Many times, no or much less pollen will come out than you hope. You can actually pinch/flick the anther and make strings of the pollen come out, and if you are lucky strings of pollen will start to hang out from the anther, and they are easy to apply.
Sometimes the pollen becomes really dry and is not that sticky and stringy anymore and trying to apply more pollen will result in the applied stuff already falling off. It is easiest when the stigma is coloured because then it is obvious to see you are applying the white stuff and covering the for example pink stigma with white stuff.
Then I label the flower using nylon string and tape, and I tighten it around the base of the flower, around the stem below the ovule. You may thing that maybe labeling around the style is a good idea, but that will dry out and always snap off or get lost. And you need to tighten the noose of the nylon string quite a bit, already imaging the petals are gone. It has to be tight enough to not come off over the now still very tiny ovule. For example, some hail or wind and you could easily lose your label, because the noose can still move over the ovule very easily.
What I think I notice is that once pollen is applied, the flower will lose petals quicker. After 2 days or so they drop. And then the ovule begins to swell. If the flower is not pollinated, the entire ovule will try out and fall off a few weeks after flowering. I already can see some ovules of my azalea right now swelling up, 3 weeks after pollinating them. It is obvious they are seed pods somewhere in July. And then they are green and become hairy.
But they will only be done around December or so. I never had any problems with collecting too early. In my climate, in late November the seed pods will become more and more brown. There is actually quite some window to collect the seeds. If you do nothing, the seed pod will crack open. Then the crack will widen and the seeds will fall out. It is very gradual. Maybe you have seen seed pods that are old. They will stay on for quite some time and eventually become star-shaped as they open up at a more than 45% angle (where 0 degrees is closed, like a pacman/smiley mouth). I just remove the seed pods in December, and my labeling method works good enough that many of them will still have their label. So I put them in the correct envelope and dry them inside. They will pop open quite quickly once inside. But you can also split then and use a knife or needle to prod all the seeds loose.
Sphagnum moss is the best seeding substrate. Of course it has to be wet. They have to be surface sown as light is needed for them to germinate. And the first 2 to 3 weeks, I put a bag around them for increased humidity. I have had some fungus grow, but I don't think it ever attacked the seeds. They germinate for me in 3 to 4 weeks. Then you see the root emerge from the seed and soon after it will become upright and the leaves emerge as the seed cover drops from the cotyledon. And then you have a long way to go to get them to big enough plants to flower.
It's kind of a low skill hobby, I would say. You just need to be patient and lucky. You can't really control what happens, as it is so complex and unpredictable. You just pick two parents. then invest the patience and some horticulture. And picking 'good' parents, I am not convinced there really is such a thing as a good parent. Pick two plants you like. Best thing to determine is if they are actually fertile. Could be frustrating trying to pollinate a plant that turns out to be seed infertile (for example because of hose-in-hose). Probably parents that are very different from each other are better, unless you have a specific goal. And even more ideal; the four grandparents are all very different. If you cross two azalea that are very homozygous, you get many different seedlings, which should be interesting and surprising and therefore more fun. If you were to cross two species, following very simple Mendelian genetics (which we know are incorrect), all seedlings would be heterozygous and they would all express the dominant phenotype. Meaning all seedlings will be identical.
Among satsuki, there are some parents that have been very popular/successful, like 'Kozan', 'Juko' and 'Suisen'. But sporting flowers in satsuki is such a peculiar and specific thing, any variation in gene expression or transposons can give different flower patterns very easily. Just keep in mind that in general, colour is dominant over whiteness, intense colour is dominant over paleness, purple is dominant over red, and solid coloured flowers are dominant over bicolour/variegated flowers.
If you just want a seedling that is nice enough to keep, 10 seedlings to flowering is fine. If you want to register your cultivar and have nurseries grow it, that's a very different challenge; 100 to 10 000, maybe.
I very likely will be able to share around seed this December. In the past, I mailed people, some in countries where they didn't have many satsuki cultivar, seeds, with varying success. It cannot really replace getting a bunch of cuttings from the nicest Japanese satsuki. But it is a lot better than nothing. You get some azalea, some can be really nice, in their own way. And they are truly yours.
It is best to use your favourate parent as the seed parent. Then you definirely know that one is definitely the parent. But furthermore, the seedling inherits plastome DNA from the seed parent only. Just as we get out mitochondrial DNA from our mothers only. Now, I think there have been some exceptions described to this rule in plants. And not sure if people know what kind of impact tht truly has. But just something to think about. Crossing A x B is not identical to crossing B x A (seed parent first, pollen parent second).
