HYB: aphylla analysis

From: Sharon McAllister <73372.1745@compuserve.com>

If you're not interested in species or hybridizing, hit that DELETE key --
because I don't have time to water this down.....

Mike Lowe wrote:
>

Black Forest (R. Schreiner 44) is Dymia (Shuber 36) X Ethiop Queen 
(Schreiner 38). Bob Schreiner always leaves you guessing, a bit. 
Dymia (Shuber 36)       is Harmony (Dykes 23) X Crusader (M. Foster 
13). Harmony is one half aphylla. The pedigree of Ethiop Queen was 
not given in the registration and the question has always been, is 
there more here? Either Black Forest STRONGLY expresses the Harmony 
aphylla gene pool or there was more lurking behind the other parent. 
Phil and Keith laid that speculation to rest with the pedigree of 
Ethiop Queen which has no recorded trace of aphylla. The bottom line 
is that the percentage of aphylla in the blacks is very small but in 
some cases strongly expressed. If the Craigs' line of aphylla talls 
works out, it truly will be a new direction.
<

This is an excellent subject for a demonstration of pedigree analysis.  

Randolph counted DYMIA as having 46 chromosomes, indicative of an
amphidiploid, which raises the question:  how can that be?  The simplest
explanation, of course, is that it got one set of aphylla  chromosomes
[base 11] and one set of base-12 from each parent.  

Knowing that HARMONY is one-half aphylla tells us that it's either triploid
or amphidiploid, with the non-aphylla chromosomes of unknown origin.  IF
the other parent were base-12, like the vast majority of both diploid and
tetraploid TBs of the time, either would be capable of producing the
required aphylla + base-12 gametes.

CRUSADER was registered as I. cypriana X ? but we know that I. cypriana is
a base-12 tetraploid so depending on the nature of the unknown parent
CRUSADER could be tetraploid, triploid, or amphidiploid.  We also know that
Foster was experimenting with inter-species crosses at the time, so the
unknown parent COULD certainly be aphylla and  IF that were the case then
CRUSADER would also be capable of producing the required aphylla + base-12
gametes.  

Taken together, we have an explanation of DYMIA's reported state.  But it's
a matter of speculation so these are clues -- not proof. 

If authenticated plants of HARMONY and CRUSADER could be observed for
aphylla traits and their chromosomes counted, we could gather evidence that
would support or refute the theory.  But  authentication is the stumbling
block.....

If first-generation crosses between aphylla and base-12 species could be
made and observed, we'd gain more clues.  An interesting experiment, but
not likely to produce anything worthy of introduction so not likely to be
done....

But we CAN extend our conjecture to the next generation, if we digress
slightly:  

Amphidiploids are neither simple diploids nor true tetraploids.  They are
tetraploids in the sense that they possess four sets of chromosomes.  They
are diploids in the sense that they have two sets of paired chromosomes. 
We tend to talk about homologous and non-homologous chromosomes, but it's
really not that cut-and-dried.  It's one thing to line-breed a set of
amphidiploids produced by crossing two known species, but quite another to
use those amphidiploids with autotetraploids or allotetraploids.  
And with such outcrosses, it all depends on how closely the foundation
species are allied.   If they are distinct enough, the offspring tend to be
infertile.  If they have a number of homologous sets, in each generation
the homologous chromosomes have a tendency to function as they would in
true tetraploids while the non-homologous tend to be eliminated to achieve
more complete fertility.   IOW, with each generation the total composition
becomes more like that of the tetraploid line.

Now -- not knowing the parentage or chromosome count of ETHIOP QUEEN, but
understanding that it has no aphylla ancestry, it is reasonable to assume
that BLACK FOREST has one complete set of aphylla chromosomes.  From my own
experience with it in those long-ago days of TB hybridizing, this is
plausible.  I found it less fertile than most TBs I worked with, although
it clearly was fertile enough to become an important breeder.  The question
becomes whether the specific chromosome that carries the branching trait is
homologous, and thus easily retained through artificial selection, or
non-homologous so that it tends to be eliminated through natural selection.


WHAT IF aphylla-type branching is not only carried on a near-homologous
chromosome but is also dominant or of mixed dominance, as opposed to
recessive?  In that case it could be readily injected into the mainstream
tetraploids and preserved through deliberate selection.  

Which means that today's TBs with good branching could easily have the
aphylla genes, even though their aphylla heritage computes to less than 1%.

And Bill Shear responded:

>
Perhaps we are missing the role played by selection.  It might be that
Belle Meade and especially JS might retain none of the genome of I. aphylla
unless certain parts of that genome were selected for.  To say that JS is
1% aphylla is probably statistically correct, but does it in fact retain
any aphylla genes?  That would depend on the selection regime that produced
it.  It could carry no aphylla genes or much more than 1%.
<

Right.  As I used to explain to my trainees, statistics merely PREDICT.  To
complete the analysis you must examine the consequences of that prediction
and assess the real outcome.  To go back to the classic coin toss, while
the coin is in the air the probability is 50/50 -- which means it will come
down either heads or tails, NOT that it will stand on edge.  Once it has
landed, it is either heads or tails and there's no point in debating what
it "should" have been.

If, in a given cross, the odds of transmitting a desired trait are 50/50 --
then it is either going to come through or it isn't.  If it does, and the
hybridizer selects that offspring for use in the next generation, the odds
will once again be 50/50.  Judicious selection can certainly eliminate OR
preserve a trait down through the generations, long after the ancestor from
which it came appears to be statistically insignficant.  

Bill continued:
>
By putting an emphasis on pedigree and crosses (certainly very, very
important) we might tend to ignore or shortchange the selection process.
From a cross that may yield 500 seedlings, a breeder selects one or two to
continue to work with.  After several generations of this, all the genetic
information of one or the other parent may have been selected out, and to
expect characteristics of that parent to crop up in future generations is
wishful thinking.
<snip>
So there is a question in my mind about the utility  of compiling long
pedigrees for irises....
<

IMO, both pedigree analysis and the ability to select seedlings are
extremely important  -- they go hand-in-hand but have very different roles.
  Analyzing pedigrees helps us plan crosses to maximize our chances of
getting what we want, while an eye for selection enables us to make the
best of whatever we get.  

Trying to decide which is more important is like comparing apples &
oranges.  Together they have taken me from an introductory rate of 1 in
5000 to about 1 in 50 and I can't imagine doing one without the other.

Sharon McAllister
73372.1745@compuserve.com   

--------------------------- ONElist Sponsor ----------------------------

FREE ADVICE FROM REAL PEOPLE!  Xpertsite has thousands of experts who
are willing to answer your questions for FREE.  Go to Xpertsite today 
and put your mind to rest. 
<a href=" http://clickme.onelist.com/ad/XpersiteCPC ">Click Here</a>

------------------------------------------------------------------------
Other Mailing lists | Author Index | Date Index | Subject Index | Thread Index