In your previous post on yield variability (10/28) you wrote:
> I occasionally peruse the results of current yield
> contests and they don't look very different to me...
> I don't know how the contests were supervised a century
> ago...
I don't think the emphasis was on yield, but on esthetics and assumptions.
Corn was hand-harvested back then, and ear size was important. Here is an
amusing excerpt from "The Hybrid Corn Makers" Richard Crabb, 1947, p 143-144
(the book is interesting and relevant to this thread, I believe it has been
reprinted recently).
"On one occasion, Henry A. Wallace, a boy of sixteen, had the chance to see
Professor Holden judge a corn show, selecting the ten most perfect ears and
rating all other entries in order of their excellence. After Holden's
remarks as judge of the show were given, young Wallace went up to talk to
him, asking among other things how he knew whether the blue ribbon sample
would, if planted the next spring, produce a better corn crop than the
sample which had received last place.
Wallace's father joined in the discussion and listened to Professor Holden
give all the reasons why the blue ribbon corn was not only of superior
quality itself but would, if planted, produce a better crop of corn next
year than the less perfect ten-ear samples that failed to place in the show.
The elder Wallace, seeing that his son was not entirely convinced with
Holden's explanation, and in all probablity having some misgivings himself,
suggested to the youth that he take some ears from the lowest ranking
ten-ear samples and plant them side by side the following spring. (snip)
Holden helped young Wallace select fifty ears in all, about twenty-five from
the prize-winning ten-ear samples and a like number from the samples ranked
as the poorest in the corn show. The next spring, in May of 1904, the high
school boy planted by hand in ear-to-row fashion about two-thirds of the
kernels from each ear. (snip)
In the fall he went to his plot, shucked out each row, weighed the corn from
each row separately, and made accurate comparisons for yield. The results
were a revelation to Wallace and his father and a great surprise to
Professor Holden. The highest yielding corn in Wallace's plot came not from
one of the pretty ears in the prize-winning entries but from one of the
tailend samples, an ear that was long, broad-kerneled, and which had only
sixteen rows of kernels and which therefore could not win a ribbon in a corn
show. The yield avergae of corn planted from all the prize-winning ears was
lower than the average established by ears from the lowest ranking ten-ear
samples."
> ... or even how such a contest is supervised today...Perhaps
> someone can comment on this...
Today, every seed company puts out hundreds (thousands in the case of
Pioneer) of strip-trials comparing their varieties with representative
entries from competitors. In some cases these trials are contracted out.
The consensus is that large plots under real farm conditions are more
dependable than small plots (although these have their place in research).
The emphasis is on covering many environments, not on replication within an
environment. A tremendous amount of data is produced, on a scale
unimaginable to typical University folks. From what I have seen (Ciba, NK,
Pioneer) the trials are conducted well and honestly, although the results
are probably manipulated at some companies by selective data presentation
(eg, not comparing your varieties with what is truly the best in a region).
Most of the information is proprietary.
> I don't think anyone would debate that we have embarked
> on this corn hybridization trajectory whole-hog and have
> hardly stopped to look back... the chapter suggests that
> there are other analagous research trajectories that need
> some reevaluation...
The main issue here is whether the research will be done in the public or
the private sector. Undoubtably, hybrid production systems have been
adopted as much to protect the investment in germplasm improvement, as for
improving yield per se. Witness the marketing of cucurbit hybrids. I don't
think it is any secret that cucurbits don't express much heterosis, but it
is still a convenient way to protect genetics.
It would be interesting to examine the decisions CIMMYT has made in the last
few decades. I believe (some of you folks in Mexico, correct me if I'm
wrong) that CIMMYT changed its emphasis from development of synthetic maize
varieties (amendable to seed saving) to pure lines for hybrid seed
production.
> I think that we can all agree that research trajectories
> are not objective... I long for the day when I will have
> the time to investigate more thoroughly the research trajectory
> that has gotten US agriculture to where it is today... this is
> an aspect of agriculture that I have not found to be covered
> seriously in any Agricultural curriculum but that I intend to
> teach someday...
Oh, I think the trajectory was objective from the standpoint of the
entrepreneurs. Whether the net effect was a societal good is a different
question. The same kind of questions can be asked in all sorts of
industries. This subject should be approached from a historical
perspective. Perhaps we can pursue this on SANET.
> My understanding is that rather than yield gains from
> heterosis, the more significant result of the hybridization
> paradigm is the extremely high uniformity of todays corn...
> which results not from hybridization but from the many
> generations of inbreeding prior to hybridization...
Paradigm? I'd just call it a technique. Besides yield and uniformity per
se, don't forget standability and uniformity in maturity, both important for
machine harvest. Also, the ability to tolerate high stand density (also
facilitated by mechanization). IMO mechanization, low input costs, and price
supports, combined with technical innovation to drive the hybrid corn seed
industry to where it is today.
Joel, I will obtain and read the references you sent. I am looking forward
to a slow-burning discussion on the evolution of ag technology using corn as
an example.
Dale Wilson
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".