for the USDA. Probably the most inter-
esting aspect of this position, in addition
to the excellent scientists that nurtured
me, was that the position had been
vacant for 4 years and most of the germ-
plasm was transferred or gone. Hence,
we needed to rebuild the program from
scratch. In winter wheat and barley
breeding, it takes 12 years to release a
new cultivar and usually at least 8
years to release good germplasm. It
was quite clear that time was working
against us, so we began a doubled
haploid program in hopes we could
rapidly inbreed lines and shorten the
time to release. Though I have never
had sufficient funds to use doubled hap-
loids except for very special genetic
studies, this approach is now very
common in well funded commercial
breeding programs. Working on germ-
plasm improvement also showed me
that despite the massive genetic
resources available to wheat and barley,
germplasm can be limiting so transform-
ation studies are very important in crop
improvement.
After working with the USDA and a
short period with Monsanto, I became
the small grains (winter wheat, barley,
and triticale) breeder at the University
of Nebraska. The collaborative USDA-
University of Nebraska wheat breeding
effort under the stewardship of John
Schmidt, Virgil Johnson, Rosalind
Morris, and Paul Mattern had been
one of the most successful breeding pro-
grams in the United States. At one time
96% of the wheat grown in Nebraska,
40% of the hard winter wheat grown
in the U.S., and 20% of the wheat
grown in the U.S. came from their
program. Here I learned that breeding
can have an impact. I also learned that
each crop has special tools that can be
used to approach specific scientific
questions. While maize had excellent
molecular markers, wheat initially had
few. However, wheat had chromosome
substitution lines (developed at
Nebraska by Rosalind Morris) where
we could study single chromosome
effects across the diverse environments
of the Great Plains. In this work, we
found that chromosome 3A would
increase or decrease grain yield by
15% in the two backgrounds that
Rosalind Morris developed. We then
used cytological tools to break up
these chromosomes by recombination
and coupled them with molecular
markers to study this chromosome in
great detail. In this way, we developed
the populations and the phenotypic
data while waiting for the molecular
marker technology to catch up. It took
Rosalind most of her professional
career to develop the substitution lines,
and after 20 years we are still studying
various aspects of this chromosome
because grain yield is still the most
important trait in plant breeding. These
studies involve huge numbers of lines
and the randomized complete block
designs were inadequate with the
highly variable conditions under which
wheat is grown. Working with statis-
ticians we implemented various statisti-
cal methods (nearest neighbor,
incomplete block designs) to remove
spatial variation in the fields, and to
improve our phenotypic estimates.
Large experiments require these statisti-
cal approaches wherever fields lack uni-
formity. If a breeder must be
knowledgeable in a number of scientific
disciplines, and if breeding is built upon
the work of previous breeders, perhaps
my program has benefited as much or
more than most breeding efforts.
However, I hope that curiosity and con-
stant questioning of how to measure and
understand the traits that breeders work
with has been my contribution. That,
my cultivars, and my students will be
my legacy.LIFE BOX 3.2. P. STEPHEN BAENZIGER 81