re: Re:HYB: Breeding strategies
iris@hort.net
  • Subject: re: Re:HYB: Breeding strategies
  • From: &* W* <i*@telp.com>
  • Date: Wed, 26 Oct 2011 19:29:16 -0600
Hi Linda -

You've got some really good questions here, and I'm delighted that my 
article stimulated your thinking about these things - that's the nicest 
complement a writer can receive! I'll do my best to answer your questions 
and help make a connection between what I wrote (which was very much in 
generalities) and your specific work with rebloomers. It's not always a 
simple matter to make a connection between what we read about breeding and 
the realities of figuring out what to do with a batch of seedlings that 
aren't what we expected.

But first, a serious disclaimer. There are folks on this list who are 
breeding rebloomers successfully, which is something I have no experience 
in to speak of. (Well, there was a washed-out bicolored IB seedling that 
bloomed in my garden in August about 30 years ago, but that's as far as it 
goes. LOL) Advice from anyone actually doing this trumps anything I might 
have to say.

> Tom, does line breeding (as you intend its meaning in your article) 
> using seedlings only descended from one initial cross?  Or does it also 
> include additions of outcrosses to the mix now and then?  You say 
> "closely related" - not sure how close you mean.

I talked about this a bit in the article. In the strictest sense of the 
term, I take line breeding to refer to never going outside the original 
parents, probably just crossing siblings for a number of generations. This 
is not something you see much of in iris pedigrees - usually some outside 
material is brought in to even the most narrowly focused lines. So "line 
breeding" is a relative term, with no sharp boundaries. In my mind, it also 
has something to do with the hybridizer's intentions. If you are not 
deliberately "throwing something new into the mix" but instead just trying 
to get the full potential out of the original parents you chose, you are 
probably line breeding. You might also ask "would this be legal if it were 
people instead of irises?" ;)

Your example of 'Immortality' x 'Matrix' is something I'd consider 
borderline. 'I Do' is parent of one and grandparent of the other, but the 
rest of their background is different.

> 
> But one of my selection 
> rules has been that if a seedling isn't an improvement of some sort 
> (either trait(s) I'm specifically breeding for or just something I like 
> more than in the parents), I haven't used it for breeding.
> 
> In your article, I notice you say that in line breeding, the above might 

> not always be a good selection criterion

Right. For me, it depends on what motivated the original cross. If you were 
trying to combine two irises with very different virtues in the hope of 
getting "the best of both", then I think it makes sense to go on for 
another generation or two, even if the initial seedlings aren't better than 
the parents. This is because some of the traits you are trying to combine 
may be recessive (and so not show up at all in the first generation), or 
governed by multiple genes (so that the trait will be "diluted" by the 
genes from the other parent). On the other hand, if both parents have 
similar virtues, and you are just hoping for something a bit better, your 
criterion makes sense.

In the example you gave, you were hoping for an improved summer rebloomer 
for your climate, so you crossed two good summer rebloomers. This makes 
perfect sense, but you didn't see the improvement you hoped for. It might 
come out in a subsequent generation, or if you repeated the cross and got 
more seedlings to evaluate. The line breeding approach would be to cross 
the best seedling back with the best parent ('Immortality', I think, from 
what you said). This is just kind of rolling the dice again. This may be 
worth doing, but the case is not as strong as if the two parent were chosen 
because they had *different* virtues you wanted to combine (say, rebloom 
and excellent form). If that were the case, going to the second generation 
would be a given for me. In your example, it would be something I might or 
might not do, depending on priorities. One can only raise so many 
seedlings, and it might seem sensible to cut one's losses and try different 
crosses instead.

> Another example (posted earlier): about 21 surviving seedlings from 
> HARVEST OF MEMORIES X RENOWN.  HoM is an early fall cycle bloomer here, 
> I think reported to summer bloom elsewhere, but if it does summer bloom, 

> haven't heard of it being reliable in various climates.  RENOWN rarely 
> reblooms here, but has summer bloomed for me once.  5 seedlings from the 

> cross have rebloomed, only one might be a summer, the others cycle. 
> None has form any better than parents (still waiting for the last one to 

> open. but nothing exciting noted during spring bloom).
> Same questions for this cross as the previous one.

My thoughts would be similar to those about the first cross, with the 
exception that you apparently did get a summer rebloomer from this one, and 
I might be inclined to keep it and cross it with an iris with excellent 
form. But that might just be sentimentality and a desire to not let the 
work "go to waste". A more objective appraisal might say that if I'm going 
to cross a summer rebloomer with a "beauty", I should use the best summer 
rebloomers I've got access to, rather than a seedling that falls a bit 
short.

My summary: if you are trying to combine very different parental traits, it 
makes sense to go for another generation at least, even if the initial 
seedlings fall short. If the parents are similar and you're just hoping for 
improvement, and you get nothing inspiring, it might make sense to let it 
go and try different crosses instead.

Happy irising, Tom


Tom Waters   

Telperion Oasis ~ www.telp.com/irises    

Cuyamungue, New Mexico, USA (zone 5/6)  

---------------------------------------------------------------------
To sign-off this list, send email to majordomo@hort.net with the
message text UNSUBSCRIBE IRIS
Other Mailing lists | Author Index | Date Index | Subject Index | Thread Index