> 1. Let the results be expressed as NET economic returns
I think we should explore all possible ways to express the results, but
Chucks statement and my challenge had to do with yield.
> let's make his hybrid a GE hybrid
Let's let Pioneer sales agronomists pick the exact hybrid, but to keep this
simple, let's use one hybrid, and one OP.
> 2. Let each crop (hybrid and OP) be grown under optimal conditions -
> for IT - which may not be the same - and let the input costs vary
> accordingly.
Trouble is, that would confound the results, and make it much harder to lay
out the plots in a reasonable experimental design. As a compromise, I would
propose a 2 x 2 factorial experiment conducted as a split plot. Management
would be the main plot factor, either Best-management-practices (BMP) for
the OP or BMP for the hybrid. Variety (OP vs hybrid) would be the sub-plot
factor. There would then be four treatments: 1) OP under BMP(OP), 2) Hybrid
under BMP(OP), 3) OP under BMP(hyb), and 4) Hybrid under BMP(hyb).
What this would mean in practice, is that each location would, in a sense,
conduct two trials, one using standard hybrid corn practices and one using
practices optimal for the OP. We will need to work on the exact definition
of the BMP factor and the important details of experimental design and
layout.
> 3. Let each crop compete under a range of growing environments - not
> "low input" per se but on healthy (disease suppressive) and unhealthy
> soils (as per Chuck's message; there is an increasingly alarming
> range of journal lit on this very interesting topic), in dry and wet
> conditions, etc. etc. This will let us test the robustness of the
> genetics of the two corns.
Absolutely! IMO, replication across environments is much more important
than replication within environments. The more sites we can get the better.
If some cooperators can only offer one plot or strip for each of the four
treatments, fine. We can accomodate that statistically. If some
collaborators could only perform one kind of BMP instead of both, we can
probably accomodate that too.
> I can find a grower who has selected his own OP corn, here close to
> Guelph, who might be able to contribute some seed. He came in and
> spoke to our corn breeder recently.
Variety selection is crucial to maximize the value of this experiment.
Let's decide after we see where the sites are located. If we end up with a
wide enough range of regions, we might want to use different pairs of
varieties for the different regions.
> I have been told by a knowledgeable and well respected plant breeder,
> that had as much effort been expended in breeding OP corn as has gone
> into hybrid corn, the yields would be as good.
Three reasons why hybrids might be better in an absolute sense include:
1. Line development efficiently unmasks deleterious genes.
2. Heterosis may involve non-additive and non-Mendelian gene action.
3. Open pollination precludes (because of pollen competition) selection
against apical dominance, and resulting increased sink strength of ear
shoots found in hybrid corn.
Dale
To Unsubscribe: Email majordomo@ces.ncsu.edu with "unsubscribe sanet-mg".
To Subscribe to Digest: Email majordomo@ces.ncsu.edu with the command
"subscribe sanet-mg-digest".