Over the years I have tried to remove the plain orange and just let the more interesting colours go to seed. Completely unsuccessful. I still end up with mostly orange and a scattering of other colours ranging from pale yellow through scarlet to mahogany. I even get the occasional variegated leaf so someone somewhere must have sown Alaska one year.

teach me up ...

what is the full latin name ...

i always thought.. that 'hybrid' color plants.. had a high propensity ... to start regressing in the genetic pool.. back toward the species ... over seed generations ...

in other words ... if they manipulated the flower color from say a standard red.. out the spectrum to a peach ... adding.. yellow to red to get the orange.. then more yellow and perhaps white ... to get out to peach.. if the peach is still fertile ... and you plant only peach.. eventually.. years later .. left to seed.. you will start getting all the colors.. and eventually a majority will get back to red ... that one being the most vigorous ...

i dont want to get into a full discussion of genetics ... but can we discuss it superficially from a science point ....

my experience is alyssum ... no matter what color i ever planted... a few years of seedlings later ... i usually ended up with almost all white.. with a rogue purple...once in a blue moon ...

ken

Tropaeolum majus is the botanical name and they tend to revert to orange.

Since I'm already doing the hybridizing gig with zinnias (nod to the zenman), I have toyed a bit with other things this year. I, too, have been interested in getting some of the other nasturtium colors to "stick around". Last year I had added Empress of India to the mix and this year, scarlet Alaska. I LOVE the variegated leaves and the mounding habit of the Alaska, but I have to admit I'm a bit overwhelmed with scarlets and orangey-scarlets at the moment. This photo is from July while things still looked fairly neat, but before the Alaskas exploded (!!) into bloom. You can see in the photo that my yellows also seemed to be dominant in the trailing plants from my saved seeds. There is only one lonely salmon in view.

Am thinking of adding something like Milkmaid, if I can still find it, or some other very light color, and maybe trying some hand pollination. I did try a cross between one of the yellows and an Alaska, though I'm not sure if the resulting seed is from my cross or not - we'll see.

As to the original question, it would seem to me, that if you only collected seed from the colors you wanted and made sure not to include any of the oranges or orangey-scarlets, and continued to do this each year, making sure to immediately discard (and I mean ruthlessly dig up and throw on the compost) any unwanted colors before they get the chance to open fully, that theoretically, you should be successful in eventually breeding the desired color or colors. But what do I know about genetics? :)

- Alex

Forgot to add the photo:

Wouldn't the easiest way be to wintersow the desired colors in containers and plant out, while removing all of the self-sown ones? I know that defeats the purpose of having an annual that self-sows easily, but the pains we take for beauty..

Ken is saying it the way I meant it, in that you want to make sure the bees don't get the chance to cross-pollinate with undesired colors. That's why I said to make sure you culled them before they were open. That is, when you can see what color the bud is, but it isn't open for pollination. You could raise them in containers, but why go to that bother since it's possible to make the cull out in the garden. The hard part for me is throwing away a good plant just because it's the wrong color. :( And if it was in a landscape that you'd arranged for show, you'd want to make sure they were a bit crowded, I suppose, to begin with so that when you culled, you wouldn't end up with giant empty spots. The remaining plants should spread to fill in your empty spots.

But, as Ken said, if you really want to go whole hog about it, you'd have a nursery bed specifically for them, isolated a good distance from any others, and then you hand-pollinate the ones you like the best, too, as well as cull.