Home Page Link AgBioWorld Home Page
About AgBioWorld Donations Ag-Biotech News Declaration Supporting Agricultural Biotechnology Ag-biotech Info Experts on Agricultural Biotechnology Contact Links Subscribe to AgBioView Home Page

AgBioView Archives

A daily collection of news and commentaries on

Subscribe AgBioView Subscribe

Search AgBioWorld Search

AgBioView Archives





April 21, 2002


Some Corny Ideas About Gene Flow and Biodiversity


Some Corny Ideas About Gene Flow and Biodiversity

- Allan Felsot, Washington State University, Department of Entomology,
Food & Environmental Quality Lab

(To Be Published in Agrichemical & Environmental News, Issue #193, May
2002; Available On-Line @ http://aenews.wsu.edu after May 2, 2002)

Mother Nature has been taking a beating. Her products are receiving a bum
rap. Carbon dioxide, the gas that plants need to make sugars and that
nearly all organisms respire, has been decried as a pollutant amidst fears
that it is the principal cause of global warming. The latest hit against
Mother Nature's ways came in the USA Today headline "Gene-altered DNA may
be 'polluting' corn' (Manning 2001). Behind the headline was a tale from
the science weekly, Nature, about genetically engineered snippets of DNA
that were found in native varieties of corn that are grown in Mexico. The
DNA was claimed to have flowed into the native corn varieties (or
landraces) via pollen from U.S. corn hybrids that contained a gene from
the insect pathogen, Bacillus thuringiensis (Bt). The gene was inserted
into the genome of the U.S. corn hybrids using the techniques of
biotechnology so that the plants would produce a protein that is
selectively toxic to specific insect pests, namely the European corn borer
and the corn earworm. Such plants can be called biotechnology derived
(i.e., BD plants or crops) to distinguish them from plants bred
conventionally by laborious crossing and selection of desirable traits
over many years.

The DNA in question was called a pollutant because it shouldn?t have been
in the Mexican corn. Bt-corn, as the genetically modified commodity is
called, is not allowed into Mexico. Perhaps some farmers who wanted to
grow more food and make some money innocently made an innocuous mistake.
Apparently not, according to the newspaper-quoted author of the report
that appeared in Nature (Quist and Chapela 2001). The principal
investigator from the University of California-Berkeley (UCB) warned, "The
probability is high that diversity is going to be crowded out by these
genetic bullies." Furthermore, the UCB investigator stated categorically
that plants with the Bt toxin have "been shown to have potentially very
bad effects on insects and the microbes in the soil."

Stimulated by the Nature paper, environmental advocacy groups (EAGs)
issued yet another proclamation for a total ban on all BD crops. No one
wants to see biodiversity destroyed and soil fertility ruined by 'crop
pollution'. A spokesperson for the Union of Concerned Scientists (UCS)
summed up another belief among the EAGs when he said, "We should not be
going forward on an experiment when we have no idea of the parameters"
(Manning 2001).

If carbon dioxide and DNA are considered pollutants, could it be that
Mother Nature is meaner than we think? Are we threatening biodiversity
and soil health by our complete lack of knowledge of what the heck we are
doing? Or are the reports and hand wringing over the UCB investigators?
letter to Nature magazine just one more mischaracterization of what is
really going on? What do we really know about the parameters related to
biodiversity of corn in its native homeland and the possible impact of BD

Gene Sleuths

First, exactly what was reported in the letter to Nature (Quist and
Chapela 2001)? Samples of corn from an isolated region near Oaxaca,
Mexico were collected and analyzed for genetic markers that would indicate
the presence of transgenic DNA. Transgenic DNA in this case would be any
DNA that is not naturally present in the corn genome but comes from other
species (whether they be plants or bacteria). For example, the UCB
scientists were specifically looking for either a Bt toxin gene, i.e., a
whole gene that codes for a protein known as Cry1Ab, or a snippet of DNA
called the cauliflower mosaic virus 35S promoter (CaMV 35S). The CaMV 35S
sequence could also come from Roundup Ready corn, a variety that is
modified to resist the herbicide glyphosate. CaMV 35S DNA does not code
for a protein but rather functions to help BD plants transcribe Bt genes
into messenger RNA for eventual translation into proteins. Obviously,
corn plants don?t normally have genes for Cry1Ab nor DNA for CaMV 35S
unless they are introduced by biotechnological methods.

In essence, the UCB scientists were testing the hypothesis that pollen
from illegally planted Bt corn had mated with native Mexican landraces. A
landrace is still Zea mays, the specific name for all corn, but it has
been developed in Mexico and adapted to its specific climatic conditions.

The UCB scientists did not elucidate why they suspected that illegal corn
hybrids were brought into the country nor their motivation for choosing
the particular corn samples that were chosen. Nevertheless, the
researchers tackled the corn samples with a technique called PCR
(polymerase chain reaction) that enables detection of very tiny amounts of
DNA by synthesizing many strands from only a single strand (for a lucid
explanation of PCR techniques see
http://www.accessexcellence.org/AB/IE/PCR_Xeroxing_DNA.html).. In other
words, one copy of a gene or DNA sequence in one corn seed out of hundreds
can be amplified into over a billion copies to identity a DNA sequence
that may be derived from genetic engineering.

The UCB scientists concluded they found evidence of CaMV 35S in five of
seven landrace corn samples. The conclusion of BD DNA "contamination" was
solely based on the use of two consecutive PCRs to detect a piece of the
CaMV 35S promoter DNA. In other words, if CaMV 35S was present in the
native landraces, then there was so little of it that two amplification
cycles were required to detect it.

Based on many generations of crossing descendants of the original Bt corn
plants, the gene construct containing CaMV 35S is known to be stable.
However, the UCB scientists used a technique called inverse PCR to
indicate that the CaMV 35S DNA introgressed into the native landrace
genome at multiple regions and also broke into smaller fragments. If this
random insertion of the promoter DNA or its pieces all over the genome did
happen, then it is possible that normal development of the seed could be
disrupted. The UCB scientists also reported that one corn sample tested
positive for the Bt Cry1Ab toxin gene, but they withheld from Nature the
DNA evidence to prove that the Bt gene was actually present yet alone

Considering that small farmers in Mexico select their seed for desirable
traits and then replant it (Louette 1997), the UCB scientists implied that
the presence of BD DNA threatened the integrity and sustainability of the
Mexican corn landraces. Moreover, the UCB scientists stated their concern
for "future genetics of the global food system" in the presence of the
widespread planting of BD crops. Yet, those concerns did not motivate the
UCB team to plant the "rogue" seeds to determine whether the Bt character
or the CaMV 35S were stable introgressions and whether the seed was even
viable. Such experiments seem a necessary first step to even begin
answering the bigger concerns of impacts on biodiversity.

Failure to take the next logical step and redo the tests on the next
generation of plants from the rogue seed before publication brought upon
the UCB scientists more than just the admiring attention of the media and
environmental groups. Molecular biologists from a plethora of academic
and government institutions soon brought their gaze to bear upon the

The Gene is Out of the Bottle

Within days after the release of the UCB report, CIMMYT (International
Maize and Wheat Improvement Center), a public research foundation whose
mission is preservation of maize biodiversity and crop improvement,
released a press release of their foundation?s own results in a search for
biotechnology-derived DNA introgressions (CIMMYT 2001). None of the 43
Oaxacan landraces in CIMMYTs gene bank or a new collection of 42 different
varieties had detectable levels of CaMV 35S promoter.

The editorial board of the journal Transgenic Research issued an essay
critiquing the UCB report (Christou 2002). Furthermore, two critical
letters that were published in Nature also uncovered profound shortcomings
in the methods and interpretations by the UCB researchers (Kaplinsky et
al. 2002; Metz and Futterer 2002). In short, major flaws were found upon
critical examination of the experimental design and techniques. PCR tests
alone are subject to artifacts (i.e., false positives) and must be
confirmed by additional types of molecular tests; in particular a
technique called ?Southern blotting? is usually exercised to confirm PCR
results (for a pictorial explanation of the technique see
http://www.accessexcellence.org/AB/GG/southBlotg.html). The critics
recommended that all claims of introgressed BD DNA should also be
supported by growing out the F1 hybrid (i.e., planting the rogue seeds,
which are actually the progeny, or F1 generation, that grows into the next
generation of plants) and re-doing the molecular tests along with
examining obvious effects on plant morphology. Nature allowed the UCB
researchers to answer their critics in a rebuttal that included additional
data not included in the original report. Nature?s editors encouraged
readers to make up their own mind about the "truth".

The controversy over "DNA pollution" rose to feverish pitches as a
conglomeration of environmental advocacy groups banded together to issue a
joint statement denouncing industry-paid, biotechnology apologists in
academia and government for personally attacking the integrity of the UCB
scientists. In response, a statement was cobbled together with signatures
from scientists all over the world stating that critique of research
methods are not ad hominem attacks (The Joint Statement 2002).
Furthermore, science could only progress by constant skeptical inquiry and
correction. (Ok, I admit as a skeptic that I signed the latter

Such controversy is the stuff that movies are made of. Well, at least
newspapers are still having a field day months after the story broke.
Witness the March 20, 2002 headline and leader in the Christian Science
Monitor: "Calling Poirot: bizarre case of cross-border 'super corn'.
Scientists claim genetically modified grain from US invades Mexico,
threatening purity of birthplace of corn" (Belsie 2002).

The Genes Flow In and the Genes Flow Out

The leader sentence of the Christian Science Monitor headline encapsulates
a common misconception about plants, namely that their genome is somehow
fixed (i.e., pure). Plants, unlike animals, are immobile and must rely on
dispersal of pollen through physical (e.g., wind) and biological (e.g.,
bees) processes. Without the gene flow that occurs from dispersed pollen,
plant populations are likely to go through a genetic bottleneck from too
much inbreeding and consequently suffer reduced genetic diversity and
possible fitness (Mayr 1970). Indeed, farmers in Mexico have noted
reduced productivity after growing local varieties (i.e., landraces) for
numerous generations in the same field without the benefit of significant
pollination from other varieties (Gonzalez and Goodman 1997). Thus, wind-
and insect-pollinated plants are naturally very promiscuous, and it is for
their own good.

Frankly speaking, the idea that Mexican landraces are "pure" is absurd.
Let's set the record straight. With few exceptions, modern food crops are
not ancient inviolate species. In essence, they are human directed
inventions of genetic manipulation by educated trial and error coupled
with intense selection pressure. Without human intervention our crops
would not be here for our use, whether they are U.S. improved cultivars or
Mexican landraces. It is the very fact that genes could be easily
exchanged between our food crops and their ancestors that has allowed
continuous improvement in agronomic traits. Such an exchange of genes
between unlike populations of the same or related species is called
hybridization, and it?s perfectly natural, especially in plants (Mayr

Somehow the myth was started that introgression of "foreign" genes into
native landraces of corn would reduce biodiversity. Ironically, Mexican
farmers have long been exchanging seeds from local varieties with each
other to improve productivity (and genetic diversity) of their corn
(Louette 1997). The difference between Mexican and U.S. seed corn
production practices boils down to open pollination vs. hybridization. In
the U.S., inbred seed lines (i.e., corn varieties that are allowed to
pollinate only themselves) are crossed each year to produce superior
performing (and more genetically diverse) hybrids. U.S. farmers pay a
premium for hybrid corn bought every year from seed companies. Hybrid
corn has a certified genetic makeup, and it consistently yields well under
the environmental conditions in which it was developed. In Mexico,
farmers grow their own seed from varieties that are open-pollinated. In
other words, they grow varieties that are subject to cross pollination
(i.e., gene flow) from similar varieties or non-local varieties.

Indeed, studies of grower practices in Mexico show that there are many
different distinct varieties of corn grown in fields with close proximity
to one another. In the region of Cuzalapa on the western Pacific coast of
Mexico, 26 distinct varieties were grown in a 24,000-hectare watershed
containing 1000 hectares of corn (Louette 1997). For example, 53% of the
corn in the watershed was produced from an individual farmer?s own seed
planted in previous years. The rest of the corn was produced from seed
exchanged with other farmers in the same watershed (36%) or from seed
outside the region (11%). One of the non-local varieties was identified
as an improved cultivar of hybrid corn from the U.S.

Because Mexican farmers make no attempt to segregate different varieties,
plenty of cross pollination has been occurring (estimated at 38%
probability for outcrossing in the Cuzalapa region) (Louette 1997).
About one-third of local corn varieties may already have introgressed
genes from non-local and improved varieties (Gonzalez and Goodman 1997).
Consequently, a continuum of morphological traits and genetic
characteristics exits among all the major local varieties (Louette 1997).
In other words, within a region abrupt shifts from one morphological trait
to another were absent. For example, seeds were not necessarily all one
color (white, blue, or yellow) in one field, but there was a lot of
mixtures (a.k.a., heterozygosity). Yet, despite the tremendous amount of
gene flow from non-local to locally adapted and selected cultivars, the
varieties survived intact as recognizable entities.

So what is the problem with biodiversity should a gene derived from
biotechnology-based breeding outcross to a local landrace variety? Given
that a plethora of genes are moving among distinct local varieties and
non-local varieties all the time without loss of biodiversity, the answer
seems to be ?nothing is wrong? other then some people seem hung up about
the process of breeding rather than the results. Thus, the ecological
effects of gene flow in the context of the local habitat, not the origin
of the DNA, should be the real focus of concern.

Teosinte: The Great Granddaddy of All Corn

For a head start on answers to the question of ecological effects of gene
flow, we can learn a lot by close examination of the relationship between
modern corn and teosinte, a grass recognized as the feral progenitor
species of corn that grows in Mexico and some Central American countries
(Benz 2001). At one time, teosinte was classified as a separate species,
Zea mexicana, but modern genetic analysis indicates it is more likely a
subspecies, i.e., Zea mays subsp. parviglumis. Thus, if gene flow and
introgression were going to have any ecologically significant effects, it
would have already happened to teosinte in regions where modern corn and
the wild grass are growing near one another.

The stinging critique of the UCB study by the editorial board of
Transgenic Research began its argument by stating a long known
reality?wind pollination would inevitably lead to gene flow between
domesticated crop varieties and their wild ancestors when grown in close
proximity to each other (Christou 2002). Indeed, studies on travel
distances of corn pollen (e.g., Table 1) show that the potential for gene
flow between corn and teosinte is very high whether the plants are growing
together in the same field, the teosinte is growing along the borders, or
alternatively, it is growing in dense patches outside of the corn field.
Recommended distances for separation of hybrid seed-corn fields are 200 m
in both Europe and the United States, and corn pollen can be detected at
distances greater than 800 m from a field (Eastham and Sweet 2002).

Despite the tremendous potential for gene flow between modern corn and
teosinte, the literature about the origin of maize and likelihood of
introgressions with teosinte suggest a lot of uncertainty about whether
introgressions are even occurring in the direction of cultivated corn to
the teosintes (Doebley 1984; Kato Y. 1997), or whether such introgressions
can become fixed without selection pressure (Martinez-Soriano and
Leal-Klevezas 2000). For example, if a trait conferring insect resistance
in a landrace introgressed into teosinte, that trait would not be
important unless the particular pests were also feeding on teosinte and
more importantly, were also major mortality factors limiting spread of the
plant. Similarly, if herbicide tolerance introgressed into a wild
relative, the gene would not be important unless herbicides were used in
the areas where the plants are growing.

Flower Power

The big difference between teosinte and corn is in the flower
(inflorescence) and seed morphology (Wilkes 1997) (Figure 1). Teosinte
has multiple branching inflorescences that only produce two seed rows
after fertilization. Modern corn generally produces one large
inflorescence but has multiple rows of seeds. Teosinte produces a seed
covered in a very hard coat called a glume that is not digestible by
animals. The glume of modern corn has been reduced to that white stuff
sticking in your teeth when you take a bite out of sweet corn. Finally,
teosinte seeds easily break off their inflorescence and thus can disperse
themselves. Corn seeds do not break off the cob and are incapable of self
dispersal. Given the differences in morphology between teosinte and
corn, hybrids should be easy to spot. Indeed, hybrids of corn and
teosinte have been found in the field, as well as produced by artificial
pollination techniques, but the seeds either do not germinate or the F1
generation is not very fit (Kermicle 1997). The inability of corn to
disperse its own seed also limits its ability to escape from fields and
invade teosinte habitat.

The striking evolutionary divergence in inflorescence morphology of
domesticated maize and teosintes exists to this day, suggesting genetic
isolation after the initial characteristics of consumable corn were fixed
despite the known gene flow between the subspecies. Recent research shows
that one gene, called tb1, largely controls the difference in
inflorescence morphology. The key to understanding why teosinte and corn
remain morphologically distinct and teosinte is able to retain its
diversity may lie in the function of tb1. Like many genes, tb1 has a
region that actually codes for a protein (the transcribed region) and a
region that is not transcribed but acts like a controller over the
transcribed region (the regulatory region). The regulatory region
functions as a switching area to turn on and off transcription of DNA to
messenger RNA. The transcribed regions of tb1 in both cultivated maize
and teosinte have maintained their polymorphic character (i.e., their
genetic variability or diversity still exists). The non-transcribed
regulatory region of tb1 in modern corn, however, has only 3% of the
genetic variation found in teosinte (Wang et al. 1999). Thus, both corn
and teosinte maintain their separate diversity in inflorescence character,
which is coded for on the transcribed region of tb1, and only the control
mechanism of modern corn has been altered over time with loss of its
original genetic diversity.

Given the fact that at minimum several hundred years of artificial
selection were required to fix the changes in the regulatory region of
tb1, it is difficult to support a hypothesis that a transgene coding for a
pest resistance character would all of a sudden change biological
diversity in teosinte or native landraces in the absence of intense
selection pressure. Indeed, after thousands of years of cultivation of
different varieties of corn in the presence of teosinte, teosinte still
retains diverse forms but none of them look (or act) like modern corn.

Biodiversity Redux

One of the arguments about the ?problem? of BD plants grown near wild
relatives is that a transgene could flow to its feral ancestor. The
resulting hybrid would acquire a fitness that could elevate it to the
status of super weed, crowding out its unfortunate ?pure? bred wild
cousins. The problem with this hypothesis is its focus on the derivation
of the gene rather than on the biology of the plants. For example, crop
hybridization with wild relatives has long been known, and in some
isolated cases there have been increased weediness of the hybrids
(Ellstrand 2001). But the highlighted examples of potential problems
involve conventionally bred plants and presumably introgression of
numerous genes. Whether a new trait, for example insect resistance or
herbicide tolerance, due to a known single gene will increase fitness can
be tested. For example, field research from the U.K. has shown over a
period of 10 years, that herbicide resistant corn did not survive well
outside of the agricultural field and never took on the characteristics of
a weed (Crawley et al. 2001). In recognition of possible advantages in
fitness or acquisition of weedy characteristics in introduced plant
species, the U.S. regulatory system requires consideration of such events
for BD crops (NRC 2002).

Concerns about gene flow and biodiversity need to focus onspecific crop
species and ecological situations on a case by case basis. If one is
concerned about loss of teosinte, or even Mexican landraces, whenever BD
corn is introduced into Mexico, then the following question should be
considered. If hybridization between crops with enhanced traits and feral
relatives is so prone to reductions in biodiversity as hypothesized, then
why has teosinte remained distinct with its known diversity of subspecies?
After all, distinct landraces of Zea mays have been grown nearby for at
least many hundreds of years. Such deductive reasoning suggests no effect
on biodiversity for corn-teosinte interactions, but it does not absolve
responsibilities for careful testing under field conditions.

A corollary question to the ecological effects of hybridization between
crops and wild relatives is why crops with superior qualities for insect
(and/or herbicide tolerance) have not themselves become weeds after 10
years of testing to date (Crawley 2001). Part of the answer is lack of
appropriate selection pressure if indeed the hybrids are stable plants.
Also, some crop plants themselves are probably not fit enough to take on
the habits of weeds. For example, corn seeds do not disperse and
therefore are not likely to become invasive.

On the other hand, certain characters, such as ability to survive drought
or salty soils, might impart different selective advantages (Crawley
2001). In that case, those situations should be studied, but the problem
is independent of how the characters were bred into the crop. . The
National Research Council (NRC), the research arm of the National Academy
of Sciences emphasizes that how crops are bred, whether by laborious hand
selection and crossings over many years, or quickly by the techniques of
molecular biology, is irrelevant to assessing ecological risk (NRC 2000,
2002). The characters produced by the techniques should be the focus of
discussion, and they should be assessed in the relevant environments where
the crops will be produced.

If we really care about biodiversity, then we should pay attention to
efficiency of land use and environmental benefits of crop improvement.
Obtaining more yield per acre of land with reduced inputs of pesticides
should make more land available for conservation. This goal seems
attainable in Mexico where research suggests that all the gene flow over
the last half-century between local landraces and non-local varieties,
including improved hybrids, has increased per acre yields (Gonzalez and
Goodman 1997) (Table 2).

The problem with biodiversity does not lie with how crops are bred.
Rather it lies with land management. Perhaps a statement from the
executive summary of a meeting concerning the impacts of modern corn on
prospects for survival of teosinte sums up well our misplaced concerns
about transgenic corn cultivars. "Changes in land use ? especially
increased grazing and urbanization ? are the principal threats to
teosinte. In recent decades there has been a drastic reduction in
teosinte populations and the danger of extinction is real. In fact,
transgenic maize may be considered a marginal threat, compared with the
effects of urban growth" (Serratos et al. 1997). Ironically, if teosinte
did adapt with a few more weedy characteristics, it would probably fare
better as its habitat is reduced in scope. Based on past experience, it
doesn?t seem likely to happen.


Benz, B. F. 2001. Archaeological evidence of teosinte domestication from
Guila Naquitz, Oaxaca. Proceedings National Academy of Sciences

Belsie, L. 2002. Calling Poirot: bizarre case of cross-border 'super
corn'. The Christian Science Monitor, March 20, 2002
>:http://www.csmonitor.com/2002-0320/p05s01-ussc.html (accessed March 25,

Christou, P. 2002. No credible scientific evidence is presented to
support claims that transgenic DNA was introgresssed into traditional
maize landraces in Oaxaca, Mexico. Transgenic Research 11:iii-v.

CIMMYT (International Maize and Wheat Improvement Center). 2001. Further
tests at CIMMYT find no presence of promoter associated with transgenes in
Mexican landraces in gene bank or from recent field collections. Press
Release, December 14, 2001

Crawley, M. J., S. L. Brown, R. S. Hails, D. D. Kohn, and M. Rees. 2001.
Transgenic crops in natural habitats. Nature 409:682-683.

Doebley, J. F. 1984. Maize introgression into teosinte--a reappraisal.
Annals of the Missouri Botanical Gardens 71:1100-1113.

Eastham, K. and J. Sweet. 2002. Genetically modified organisms (GMOs):
the significance of gene flow through pollen transfer. European
Environmental Agency, Copenhagen, Denmark

Ellstrand, N. C. 2001. When transgenes wander, should we worry? Plant
Physiology 125:1543-1545.

Gonzalez, F. C. and M. M. Goodman. 1997. Research on gene flow between
improved maize and landraces. In Gene Flow Among Maize Landraces,
Improved Maize Varieties, and Teosinte: Implications for Transgenic
Maize, CIMMYT, Mexico, D. F. pp. 67-72 (available at

Joint Statement in Support of Scientific Discourse in Mexican GM Maize
Scandal. 2002. http://www.agbioworld.org/jointstatement.html (accessed
March 18, 2002)

Kaplinsky, N., D. Braun, D. Lisch, A. Hay, S. Hake, and M. Freeling.
2002. Maize transgene results in Mexico are artefacts. Nature 416:601.

Kato Y., T. A. 1997. Review of introgression between maize and teosinte.
In Gene Flow Among Maize Landraces, Improved Maize Varieties, and
Teosinte: Implications for Transgenic Maize. Serratos, J.A., M.C.
Willcox, and F. Castillo-Gonzalez (eds.). CIMMYT, Mexico, D.F. pp. 44-53
(available at

Kermicle, J. 1997. Cross compatibility within the genus Zea. In Gene
Flow Among Maize Landraces, Improved Maize Varieties, and Teosinte:
Implications for Transgenic Maize, CIMMYT, Mexico, D. F. pp. 40-43
(available at

Louette, D. 1997. Seed exchange among farmers and gene flow among maize
varieties in traditional agricultural systems. In Gene Flow Among Maize
Landraces, Improved Maize Varieties, and Teosinte: Implications for
Transgenic Maize, CIMMYT, Mexico, D. F. pp. 56-66 (available at

Manning, A. 2001. Gene-altered DNA may be 'polluting' corn. USA Today,
November 29, 2001 :p. 15D.

Martinez-Soriano, J. P. R. and D. S. Leal-Klevezas. 2000. Transgenic
maize in Mexico: no need for concern. Science 287(5457):1399.

Mayr, E. 1970. The breakdown of isolating mechanisms (hybridization).
Populations, Species, and Evolution. Harvard University Press, Cambridge,
MA :Chapter 6, pp. 69-81.

Metz, M., and J. Futterer. 2002. Suspect evidence of transgenic
contamination. Nature 416:600-601.

National Research Council (NRC). 2000. Genetically Modified
Pest-Protected Plants: Science and Regulation. National Academy Press,
Washington, D.C.

National Research Council (NRC). 2002. Environmental Effects on
Transgenic Plants. The Scope and Adequacy of Regulation. National
Academy Press, Washington, D.C. 320 pp.

Quist, D. and H. I. Chapela. 2001. Transgenic DNA introgressed into
traditional maize landraces in Oaxaca, Mexico. Nature 414(29

Pleasants, J. M., R. L. Hellmich, G. P. Dively, M. K. Sears, D. E.
Stanley-Horn, H. R. Mattila, J. E. Foster, T. L. Clark, and G. D. Jones.
2001. Corn pollen deposition on milkweeds in and near cornfields. Proc.
National Academy Sciences 98:11919-11924.

Serratos, J. A., M. C. Willcox, and F. eds. Castillo-Gonzalez. 1997.
Executive Summary . In Gene Flow Among Maize Landraces, Improved Maize
Varieties, and Teosinte: Implications for Transgenic Maize, CIMMYT,
Mexico, D. F. pp. vii-xi (available at

Wang, R.-L., A. Stec, J. Hey, L. Lukens, and J. Doebley. 1999. The
limits of selection during maize domestication. Nature 398:236-239.

Wilkes, H. G. 1997. Teosinte in Mexico: personal retrospective and
assessment. In Gene Flow Among Maize Landraces, Improved Maize Varieties,
and Teosinte: Implications for Transgenic Maize, CIMMYT, Mexico, D. F.
pp. 10-17 (available at


Table 1. Deposition of pollen (as a percentage falling into each density
category) on milkweeds as a function of distance from the edge of a corn
field (modified from Pleasants et al. 2001). Note that few pollen grains
are deposited beyond 5 m, although a single pollen grain may travel for
over 800 m (Eastham and Sweet 2002). Most of the pollen leaving a field
is blocked by the first several rows in a field Thus, successful
pollination between crop varieties would depend on distance of separation
between fields and density of pollen flow. In corn, one pollen grain
would have to land on the tip of the female flower (called a silk) for
successful fertilization and production of one embryo (seed) on a cob that
would normally have hundreds of seeds.

Pollen Density

Inside Field Distance from Edge of Corn Field (meters)
0 1 2 4-5
0 - 100 52.7 83.3 90.0 97.4 99.6
100 - 200 17.0 9.3 6.2 2.4 0.4
200 - 300 10.1 3.3 2.2 0.0
300 ? 400 7.2 1.7 0.6 0.2
400 - 500 4.1 0.8 0.2
600 - 700 2.1 0.2 0.1
700 - 800 0.9 0.2 0.1
800 ? 900 0.9 0.3 0.1
900 - 1000 0.2 0.1 0.0

Table 2. Comparison of yields of different landraces collected
approximately 30 years apart in Mexico and in Medellin, Columbia. The
data are based on Gonzalez and Goodman (1997) who summarized the research
from two independent studies. Seeds collected years ago were stored in
germplasm seed banks using practices to ensure the viability of the seed.
Periodically the seed is grown out and a new generation of seeds are
obtained. Caution should be used in interpreting the data because
differences in yield could be due to physiological effects from the age of
the seed or it could be due to enhanced vigor associated with
hybridization of local landraces with new varieties. Nevertheless, the
data suggest that gene flow among landraces has the capability of
improving productivity and thus can increase land use efficiency as
measured by yield per hectare.

Geographic Origin (Collection Time) Number of Entries Yield
(tons/hectare) Days to Flowering
Puebla (old) 21 1.8 83
Puebla (new) 110 2.3 81
Mexico (old) 40 1.4 79
Mexico (new) 92 2.1 80
Tlaxcala (old) 4 1.1 76
Tlaxcala (new) 66 2.0 76
Medellin, Columbia (old) 10 2.3 No data
Medellin, Columbia (new) 10 3.16 No data


Figure 1. Evolution in morphology of Zea mays from ancestral teosinte to
modern corn. The middle figure shows possible hybrids of teosinte and
corn landraces. (First and third photograph obtained from
http://bioinformatics.ncsu.edu/buckler/maize/DomestHistory.htm. Middle
photograph obtained from