The development of double flowers, for example, of Gallica, Rugosa and Spinosissima roses was likely by self-pollination, so this isn’t a new phenomena in roses . Yet there doesn’t seem to have been much work done using this breeding technique to develop roses having double flowers. It appears there is very little scientific literature on this subject, although I haven’t done a thorough search.
An interesting thing happened to me this spring. I have two selfed seedlings of a selfed ‘High Voltage’ seedling, the latter having flowers similar in size and appearance to its parent, although they are not as yellow and so fade quickly to white within a day. These F2 seedlings of ‘High Voltage’ are less yellow than their parent and have extremely double flowers. One in fact appears to have no pistils. The growth of these seedlings isn’t vigorous.
I also have a very double seedling of ‘Sevillana’ from a hip that wasn’t from a controlled cross, and so it was likely selfed. The flower colour is similar in appearance to its parent.
I’m sure some breeders on this forum have had similar experiences to mine, but if so it appears they haven’t talked much about it.
Anyway, it’s something I’m going to work on more in my breeding programs. I have often looked, for example, for sports of ‘Morden Sunrise’ having more double flowers, but I’ve never encountered any. Perhaps by selfing for a couple of generations like I’ve done with ‘High Voltage’ I’ll get something I want in this respect.
By the way, when the stamens are replaced by petals in seedlings or in a sport, this is known as a homeotic mutation. Thought you would like to know this if you don’t.
The double-flowered Rugosas, of the type formerly called “Rosa Regaliana”, were probably derived from Rugosa hybrids that Maximowicz found in a Japanese garden.
But it won’t always happen so easily.
William Paul: The Rose Garden (1848) div. II. p. 124
“Whence arose the Gloire de Rosoméne, the type of this group, it is difficult even to conjecture. It was raised by M. Vibert, but I believe he knows not from what source. The brilliancy of the flowers caused some stir when it was introduced, and cultivators have long been striving to obtain full Roses to vie with it in colour. I have raised several seedlings, some more brilliant, but none more double.”
Karl K delved into the subject of selfing a plant species to get certain genetic expressions.
The article he cited was by Luther Burbank breeding of the Beach Plum, (prunus Maritima).
Maybe Karl can republish some of his research.
Burbank’s Beach plum involved OP seedlings from a population of wild P. maritima that were selected for their superiority each generation. Some were sweeter than average, some larger, and some juicier. After a few generations he had a domesticated Beach plum that was superior in various ways to the original type. This specimen was crossed with a superior Japanese plum. In the second generation Burbank got the Giant Maritima plum, the largest plum Burbank had ever seen. And this was a man who really knew his plums. http://bulbnrose.x10.mx/Heredity/Burbank/BurbankMaritima1914.html
Rose species can be improved in the same way, such as the Double Scotch roses I mentioned in my previous note. It makes more sense to me to breed from these improved types rather than going back to the wild type.