deutsch
english
francais
italiano
espanol
Photo
KEYCODE BAYER #309

AFRICAN CENTRE FOR BIOSAFETY

OBJECTION TO BAYER CROP SCIENCE'S APPLICATION FOR COMMODITY CLEARANCE OF GENETICALLY MODIFIED COTTON LL 25

http://www.biosafetyafrica.net

JUNE 2007

1 INTRODUCTION

Bayer CropScience has submitted an application to the Executive Council established under the Genetically Modified Organisms Act in South Africa, for commodity clearance of its Liberty Link 25 (LL25) genetically modified (GM) cotton. This is the first ever application for commodity clearance to enable the importation of GM cotton into South Africa.
To date, Australia and the United States have approved LL25 for commercial growing while Canada, Japan, Korea and Mexico have approved it for importation as food and feed Clearly Bayer's application to the South African authorities is part of its global strategy to penetrate the GM cotton market. If the Executive Council grants the application, it will open the doors for the importation into South Africa of GM cotton from the US, where cotton production is heavily subsidized. It is well documented that these subsidies are destroying livelihoods in Africa and other developing regions.
South Africa's cotton production is small, averaging around 20 000 ha for 2006/7 and steadily declining. Although 90% of South Africa's own cotton production consists of GM varieties, this production contributes minimally to South Africa's overall consumption. Indeed, South Africa imports cotton from several countries in the Southern African Development Community (SADC), especially Zimbabwe, Zambia, Mozambique and Botswana. Bayer's application if granted, will severely impact on and have far-reaching ramifications for the livelihood of cotton farmers and rural populations who rely on cotton as a cash crop and farm income, quite apart from the loss of revenue for SADC governments and the concomitant negative socio-economic consequences flowing from this.

2 SUMMARY OF SCIENTIFIC ASSESSMENT

1. We have found that the results of the Southern blots to be
unclear and lacking the sensitivity to detect all the expected
fragments. More importantly, the data furnished by Bayer
only looks at individual plants and does not use several plants for the study;
2. The use of the CaMV promoter generates a
recombination hotspot, which causes increased
rearrangements and gene instability. This transgenic
instability raises many known and unknown risks including
altering plant gene expression and metabolism, and the
possible generation of new viruses;
3. The direct effects on human health and any adverse
effects are related to the consumption of the processed
seed (oil) or cotton linter fiber and cotton cake is used as a
feed for livestock. Of main concern is the higher levels of free
Gossypol in Gossypium barbadense.
4. In the USA the maximum recommended amounts of
gossypol in cattle feed is 0.05-0.01% free Gossypol (Kirk and
Higginbotham 1999). The LL25 contains levels of
approximately 0.5% fresh weight BK02B005.PG 22 TABLE8). If
the cake is used in an amount more than 10% of the animals
feed, health problems can be expected, including male sterility;
5. Of additional concern is that the company data show
that LL25 cotton- seed contains significantly less vitamin E
than the non-transgenic cotton (83.6 IU/kg compared to
187.8 IU/kg - table 7, BK02B005). It is therefore a poorer quality feed;
6. About 40-60% of cottonseed passes through the animal
gut and into the environment. Soil bacteria will be exposed
to the transgenic DNA where is may be taken up and
incorporated into its genomic DNA (horizontal gene transfer, HGT);
7. Resistance to phosphinotricin has been found in bacteria
from five other genera. HGT to soil bacteria from plant leaf
material has been shown to occur. The fact that bacteria
contain DNA sequences similar to the bar gene
(transacetylases) will increase the likelihood of homologous
recombination and HGT. The release of transgenics
containing antibiotic resistance genes will spread antibiotic
resistance genes to pathogenic bacteria in the soil thereby
compromising the ability to treat current and future diseases.

3 PROFILE OF BAYER

3.1 BAYER CROPSCIENCE
Bayer is a Germany-based transnational corporation with 350
offices on five continents. It is best known worldwide for its
aspirin. The cornerstones to the company are in Europe, North
America and the Far East with a growing presence in China.
Bayer AG, the main Bayer corporate name, is now a
management holding company with several subsidiaries. Bayer
AG is a massive Germany based chemicals and
pharmaceuticals manufacturer, and it is a key player in the
development, commercialization and sale of GM crops.
The Bayer subsidiary dealing with crops is Bayer CropScience
AG. Bayer Cropscience was founded in 2002 after buying out
Aventis Cropscience, which is known for its contamination
scandal in the UK, which involved 'Starlink' maize. The
company's Crop Protection unit makes fungicides, herbicides,
and insecticides. Bayer CropScience also has two other
divisions, Environmental Science and BioScience, that focus on
non-crop chemicals (such as consumer lawn care products)
and genetically engineered seeds, respectively.

3.2 Bayer In South Africa
Bayer Cropscience has been forced out of the UK, withdrew its
plans to commercialise GM canola in Australia, and have
abandoned its research in India. Bayer Cropscience, has
dumped toxic waste in the South Durban Basin, which led to
water being contaminated with chrome VI, a carcinogenic
toxic substance. Bayer has bankrolled GM sugarcane research
in the hope that it could corner the GM sugar to ethanol
biofuels market and has also applied to the South African
authorities for a commodity clearance of its risky GM rice, LL62.

3.3 Bayer and GM Rice contamination
2006 will go down in history as the year Bayer's Liberty Link (LL
rice 601) unapproved GM rice contaminated commercial
trade around the world and food aid in Africa (Sierra Leone and Ghana).

4 SOCIO-ECONOMIC REVIEW

4.1 Cotton Production In Africa
(Sections 4.1 and 4.2 has been taken from research conducted
for the African Centre for Biosafety by Stephen Greenberg)
Cotton and handicraft cotton textile production were part of
African economies, in particular in West Africa, long before the
arrival of colonists. Cotton production for export took off from
the early 1960s in West Africa. Production of seed cotton rose
from 30 000 tons in 1960 to 1.9m tons in 2002 with 95 per cent of
lint produced in the region being exported. Sub-Saharan Africa
as a whole recorded a 175 per cent increase in cotton
production between 1993 and 1998.
A fairly broad estimate is that in 2002 Africa contributed around
8-10 per cent of global production and around 15-18 per cent
of global exports. Thirty African countries produce cotton, the
most significant being Egypt in North Africa, and Mali, Cote d'
Ivoire, Benin and Burkina Faso in West Africa.
Between them, these five countries produce slightly less than
two-thirds of Africa's total production. Egypt uses most of its
own production domestically, while the West African countries
produce mainly for export. In the 1980s, cotton exports from
Africa rose dramatically, although from a low base. The
overwhelming majority of African cotton is exported as lint, that
is after the first step of separating seed and fibre.
Cotton exports are very important for the economy of West
Africa accounting for approximately 40 per cent of all
merchandise export earnings in Benin and Burkina Faso, and 30
per cent in Chad and Mali. On estimate more than 10 million
livelihoods depend on the cotton industry in West and central
Africa, with cotton a typical, and often dominant, smallholder
cash crop. Although they are amongst the cheapest cotton
producers in the world subsidies have skewed the market to the
extent that West and central African producers receive only 60
per cent of their costs and prices have dropped 31 per cent
despite a 14 per cent increase in yields in recent times.

4.2 Map Of Cotton Production In Africa
Egypt in North Africa is the largest single cotton producer on
the continent. However, domestic consumption accounts for
more than half its production and the country does not export
much raw cotton. Egypt has a strong spinning and weaving
industry, and produces high quality fabrics from long staple
extra fine cotton fibre. Other North African countries producing
cotton are Morocco and Tunisia.
In East Africa cotton was an important crop before the 1970s.
However, since then cotton's share of production has declined.
Uganda, Ethiopia, Kenya, Somalia and Sudan all produce
some cotton. Since liberalisation, cotton production in the
region has shifted to a low-input basis and export strategies
target the 'market window' when there is a limited supply of
new cotton coming onto the market from other parts of the world.
Zimbabwe and, following some way behind, Tanzania are the
largest cotton producers in Southern Africa. Zimbabwe is the
sixth largest cotton producer in Africa exporting a significant
share of the crop. Cotton production in Zambia has grown
quite significantly since 1995, while South Africa's cotton
production has dropped since the mid-1990s and particularly in
the new millennium following low prices and drought.
Mozambique, Angola and Malawi also produce small amounts of cotton.

4.3 South Africa's cotton trade with Africa
In South Africa, local cotton production is protected in that
cotton mills are compelled to buy South African fibre first and
can only import cotton once local supplies have been
exhausted. However, currently South African cotton production
is at an all time low due to low global cotton prices. As a result,
South Africa imports between 80 and 90% of its cotton
consumption from its neighbours in the sub-region, the Southern
African Development Community (SADC), especially from
Zimbabwe, Zambia and Mozambique. According to free trade
agreement between SADC countries (in force since 2000 called
the Southern African Customs Union Agreement 'SACU') there
has been no duty on cotton imports from these countries since
1st January 2004. Other countries in SADC exporting cotton to
SA include Botswana, Namibia and Mozambique, where whole
cotton- seeds are imported into South Africa where the ginning
also takes places.
According to the National Department of Agriculture's website,
during the 2003/4 marketing season, 84 % of South Africa's
cotton lint needs came from SADC with Zimbabwe and Zambia
contributing 30% and 39% respectively. Non-SADC countries
contribute 16% of South Africa's cotton needs with the US only
accounting for 1% of lint imports. The table below is instructive
of South Africa's trade with Zambia and Zimbabwe as well as
other countries in SADC, for the last ten years.

4.4 Killing the SADC cotton market: US subsidized cotton
The subsidization of US cotton has been thoroughly
documented, and its concomitant effect on global cotton
prices and cotton growing countries in Sub-Saharan Africa.
The scale of the US government support to its 25,000 cotton
farmers is staggering. According to Oxfam, every acre of
cotton farmland in the US attracts a subsidy of $230, or around
five times the transfer for cereals. In 2001/02 farmers reaped a
bumper harvest of subsidies amounting to $3.9bn - double the
level in 1992. To put this figure in perspective, America's cotton farmers receive:

• more in subsidies than the entire GDP of Burkina Faso - a
country in which more than two million people depend on
cotton production. Over half of these farmers live below the
poverty line. Poverty levels among recipients of cotton subsidies
in the US are zero.
• three times more in subsidies than the entire USAID budget
for Africa's 500 million people.

Using data from an International Cotton Advisory Committee
model, Oxfam has attempted to capture the cost to Africa of
American cotton subsidies in 2001/02. For the region as a
whole, the losses amounted to $301m. Eight cotton-producing
countries in West Africa accounted for approximately twothirds
($191m) of overall losses.
The small size of the countries concerned and their high level of
dependence on cotton magnify the effect of US policies. For
individual countries, US cotton subsidies led to economic shocks
of the following magnitude:

• Burkina Faso lost 1 per cent of GDP and 12 per cent of export earnings.
• Mali lost 1.7 per cent of GDP and 8 per cent of export earnings.
• Benin lost 1.4 per cent of GDP and 9 per cent of export earnings.

5 SCIENTIFIC SAFETY ASSESSMENT

In response to the African Centre for Biosafety's (ACB)
application for access to information to Bayer's application
and risk assessment data, the ACB was furnished with more
than 1500 pages of data. We have thoroughly canvassed this
data and present our safety assessment findings as set out below.

5.1 The Transgenic construct
LL25 was constructed first by creating a transgenic Gossypium
hirsutum cotton containing the bar gene under the control of
the cauliflower mosaic promoter, CaMv. The transgenic
Gossypium hirsutum cotton was crossed with the closely related
species, Gossypium barbadense to produce LL25. LL25
therefore contains the antibiotic résistance gene, bar, from the
bacteria Streptomyces hygroscopius that confers resistance to
phosphinotricin, an herbicide. It also contains genetic
characteristics inherited from the cross with Gossypium barbadense- an extra long staple fibre that makes it preferable for high quality cotton fibre.

5.2 Genetic Stability
To determine what genetic changes were introduced into LL25,
data is presented on the copy number and site of insertion of
the transgenic cassette into the cotton genomic DNA.
However, we have found that the results of the Southern blots
to be unclear and lacking the sensitivity to detect all the
expected fragments (Appendix 11, pg18 fig 6 and 7 of
application). More importantly, the data furnished by Bayer
only looks at individual plants and does not use several plants
for the study.
Is the insertion point the same for several (>20) plants grown in
the field? This important question addresses genetic stability.
From other evidence it would appear that use of the CaMV
promoter generates a recombination hotspot to cause
increased rearrangements and gene instability (Kohli et al
1999). PCR of genomic DNA from several plants (population of
20+ individual plants) with primers flanking the genetic elements
of the transgenic cassette and DNA sequencing of selected
PCR products should be carried out. Experiments using
comparative genomics are required to fully establish genome
stability of transgenic lines.
Techniques such as repPCR, RAPD and comparative genome
hybridization (CGH) have been shown to be effective in
establishing genome similarity (Bao et al. 1993, Pinkel and
Albertson 2005). This is required since fragmenting and
scattering of the transgenic cassette in the genome
(transpositions with rearrangements and deletions) may result in
loss of the primer binding sites or a large distance (>10kBp)
between genetic elements of the cassette, giving in false
negative results by when detection is carried out by standard PCR.
It is therefore not certain if there are rearrangements of the
transgenic cassettes and genetic instability. There are,
however, known problems with the genetic instability of
transgenic constructs containing the CaMv viral promoter (Kohli
et al 1999). This transgenic instability raises many known and
unknown risks including altering plant gene expression and
metabolism, and the possible generation of new viruses.

5.3 Effects On Human And Animal Health
In South Africa, approximately 30% of the imported cotton is as
a fibre 30% as seed and 30% as seed cake. A small proportion
<0.3% of cotton linter fiber is used as a thickener in baked
goods, dressings, snacks. Therefore the direct effects on
human health and any adverse effects are related to the
consumption of the processed seed (oil) or cotton linter fiber.
The cotton cake is used as a feed for livestock (cattle, poultry,
swine, catfish) (OGTR 2002).
Of main concern are the higher levels of free Gossypol in
Gossypium barbadense (Pima varieties) (Sullivan et al. 1993).
The toxicity of gossypol is well documented causing heart and
liver damage (Lindler 1990) with poisoned cattle displaying
symptoms of difficulty in breathing, weakness, diarrhea and
death (Kirk and Higginbotham 1999). In the USA the maximum
recommended amounts of gossypol in cattle feed is 0.05-0.01%
free Gossypol (Kirk and Higginbotham 1999). The LL25 contains
levels of approximately 0.5% fresh weight. If the cake is used in an amount more than 10% of the animals feed, health problems can be expected. Gossypol
also causes male sterility (Brocas et al).
There is surely no mechanism whereby this information has
been passed on to the farmer to prevent animal toxicity. This is
particularly true if farmers are used to supplies of seed cake
from low or Gossypol free varieties such as Gossypium hirsutum.
Of additional concern is that the company data show that LL25
cotton seed contained significantly less vitamin E than the nontransgenic
cotton (83.6 IU/kg compared to 187.8 IU/kg - table 7,
BK02B005). It is therefore a poorer quality feed.
A study on chickens compared LL25 treated with herbicide,
LL25 not treated with herbicide and the parental non-GMO
cotton not treated with herbicide (Appendix 16, Study
#13798.4100). The authors incorrectly concluded that there
was no effect of GM cotton on the health (weight gain) of
developing chicks. While there was no difference between
chicks fed LL25 treated with herbicide and the parental non-
GMO cotton not treated with herbicide there was a significant
difference between LL25 not treated with herbicide and the
parental non-GMO cotton not treated with herbicide. The
authors report this "a significant difference between Group A
(commercial variety) and Group C (LL25, not treated with
herbicide)...the mean breast weight for these two groups were
169.4g and 154.1g respectively". However, they dismiss this
finding since they cannot explain it. (pg28 of Appendix 16,
Study #13798.4100). This misinterpretation is highly relevant
since it suggests that LL25 cotttonseed meal may be an inferior
livestock feed.

5.4 Environmental effects

5.4.1 HGT to soil microbiota: effects on soil health
The digestibility of whole seeds by animals such as cattle is only
5%. Therefore, the cotton-seed is usually cracked and this
results in a digestibility of 40-60% (Sullivan et al 1993). This means
that 40-60% passes through the animal gut and into the
environment. The effects on soil the biodiversity and
functioning of soil microbiota has not been considered. Soil
bacteria will be exposed to the transgenic DNA where is may
be taken up and incorporated into its genomic DNA (horizontal
gene transfer, HGT).
Phosphinotricin originates from the bacteria Streptomyces
viridochromogenes Streptomyces hygroscopicus and several
other Streptomyces species. It acts as an inhibitor of glutamine
synthetase and therefore has herbicide and antibiotic
activities- it is active against Gram-positive and Gram-negative
bacteria as well as against the fungus Botrytis cinerea (Bayer et
al., Helv. Chim. Acta 55 (1972) 224).
Resistance to phosphinotricin has been found in bacteria from
five other genera (Bartsch 1989), suggesting that they contain
homologs to the bar/PAT gene. HGT to soil bacteria from plant
leaf material has been shown to occur (despite the large
excess of plant DNA) and is most efficient where sequence
homology is present (de Vries and Wackernagel 1998). The fact
that bacteria contain DNA sequences similar to the bar gene
(transacetylases) will increase the likelihood of homologous
recombination and HGT.
Furthermore, the probability HGT is increases because this
genetic construct LL25 contains the CaMV promoter.
The biosafety risks of the viral CaMV promoter include
increased recombination (rearrangements, deletions,
insertions). There is evidence from the laboratory (Kohli et al.
1999) and field studies (Quist and Chapela 2001, Collonier et
al., Ho et al. 2000) that the CaMV is a recombination 'hotspot'.
The CaMV results in very high expression levels that may result in
unintended (pleiotropic) effects from the expressed transgenes.
Increased recombination with other viral elements and the
creation of new viruses (Wintermantel et al. 1996, Vaden and
Melcher 1990, Greene et al. 1994).
This new genetic material acquired by HGT will only be retained
if it has a selective advantage. The regular application of
phosphinotricin herbicides will ensure a selective advantage.
There may also be advantages in the absence of applied
herbicide. Many antibiotics are produced by Actinomycete
bacteria to kill competing bacteria in the soil; therefore
acquiring phosphinotricin antibiotic resistance may acquire a
selective advantage per se. Selective pressures may also
include several stresses such as soil tilling or application of
agrochemicals since current evidence suggests that a stress
response facilitates the HGT and spread of antibiotic resistance
genes. For example the SOS response-induction of specific
genes in response to DNA damage-alleviates the repression of
genes necessary for the horizontal gene transfer of the mobile
genetic element conferring resistance to the antibiotics
chloramphenicol, trimethoprim, streptomycin, and
methoxazole. (Beaber et al., 2003).
Mobile genetic elements have played a key role in spreading
antibiotic resistant genes amongst bacterial populations and
contribute to multiple antibiotic resistance of bacterial
pathogens (Nikolich, et al. 1994; and Witte, 1997). Therefore,
there are risks associated with the spread of antibiotic
resistance genes amongst soil bacteria, even when there is no
selection for the transgenic construct per se. The effects from
these changes in soil biodiversity and soil ecosystem functioning
have not been considered.
The release of transgenics containing antibiotic resistance
genes will spread antibiotic resistance genes to pathogenic
bacteria in the soil thereby compromising the ability to treat
current and future diseases. Even though phosphinotricin-
based antibiotics are not currently being used to treat human
diseases, they represent an arsenal for development of new
antibiotics. Health experts worldwide are concerned about the
spread of antibiotic-resistant microbial infections and the
shrinking arsenal of compounds that can effectively treat them
(see MRSA/drug resistance news 29 April 2007).

5.4.2 Weediness and Hybridization
As mentioned, approx 30% of imported cotton is seed (in 2004
114 490 tonnes of cottonseed imported, Table1). The dossier
states that it is "not expected that cottonseed, once imported
into South Africa, will be transported to cotton growing areas".
However, experience over the last 10 years with transgenic
crops indicates that containment in the wider environment is
very difficult. This can be due to human error or illegal seed
sales and planting. For example, transgenic DNA found in
traditional maize landraces in Oaxaca, Mexico (Quist D and
Chapela IH. Nature 2001, CONABIO) confirmed these findings-
Science 1 March 2002). This contamination of landraces
occurred despite GM maize never being approved in Mexico.
Certified non-GM canola seedlots grown in western Canada
contained transgenic herbicide-tolerance traits after only 6-7
years of commercial production of GM canola in Canada.
Between 59%-97% of the seedlots contained more than 0.01%
transgenic DNA. This level of contamination in pedigreed seed
is noteworthy and disturbing because it shows that even
stringent segregation systems were not sufficient to deliver pure
non-GM canola seed to farmers in western Canada (Friesen et
al. 2003 Downey and Beckie 2002).
Therefore, assurance that uninformed recipients will not receive
and plant the seed represent an ineffective measure for
containment of LL25. Furthermore, some cottonseed fed to
livestock may be undigested, reach the soil and germinate.
This is a distinct possibility since cottonseed is known to be
poorly digested by cattle (digestibility 5-60% depending upon
whether the seed was mechanically cracked) (Sullivan et al 1993).
While Gossypium species are primarily self-pollinated (Wendel,
1995), outcrossing does occur and inter-specific gene flow has
been documented in a number of cases (Brubaker et al., 1993;
Wendel et al., 1989; Wendel and Percy, 1990). Therefore
unintended seed planting will result in outcrossing and
hybridization of LL25 with other Gossypium species. The risk of
outcrossing and hybridization is increased since Africa has 14
species of Gossypium and there is evidence of Gossypium
barbadense becoming weedy (Fryxell 1979). The company
dossier also states that crossing in the field is unlikely since
cotton is mainly self pollinating with cross pollination in the field
minimal due to low levels of insects a consequence of high
insecticide use. These assumptions of agronomic practices do
not represent effective measures for containment. The cross
pollination of cotton occurs by bees so distances of more than
1000m may be required to limit plant mediated gene flow to
<0.025% (van Deinze 2005).
No data has been presented on the weedinesss potential of
LL25 (studies on fitness, dormancy).
It is unclear from the notification if any environmental
monitoring or assessment will take place, as required under the
National Environmental Management Act of 1998 (NEMA) and
the Biosafety Bill (Bill number 1576), and alignment with the
Cartagena Protocol.
In conclusions, due to the lack of Biosafety measures for
containment of imported seed and the negative affects on
livestock health, it is recommended that LL25 be rejected.

CONCLUSION

We are vehemently opposed to Bayer's application on socioeconomic
grounds. Approval will mean the dumping of cheap
subsidised GM cotton on the South African market and in so
doing, substitute the SADC countries with the US, as South
Africa's main cotton trading partner. This will destroy the
livelihoods of millions of Africans in the sub-region.
Ably assisted by Bayer, which already makes a killing selling
agrochemicals in Africa, the entry of cheap, subsidized US GM
cotton on the South African market will inevitably find its way to
other markets in Africa, with devastating consequences for
rural livelihoods of the region as a whole.
Our independent risk evaluation of Bayer's application has
revealed that Bayer's GM cotton poses unacceptable risks to
human and animal health and the environment. We strongly
recommend that Bayer's application be summarily rejected.

References

Scientific Assessment
Bayer, E., Gugel, K.H., Haegele, K., Hagenmaier, H., Jessipow, S., Koenig,
W.A., Zaehner, H. (1972) Phosphinothricin und phosphinothricyl-alanylalanin.
Helv. Chim. Acta 55, 224-239.
Bao, P.H.; Castiglione, S.; Giordani, C.; Li, W.; Wang, G.; Datta, S.K.; Datta,
K.; Potrykus, I.; Sala, F. (1993). State of the foreign gene and of the
genome in transgenic rice (Orvza sativa L.). Cytotechnology 11: S123125.
Brocas, C, R.M. Rivera, F.F. Paula -Lopes, L.R. McDowell, M.C. Calhoun, C.R.
Staples, N.S. Wilkinson, A.J. Boning, P.J. Chenoweth, and P.J. Hansen.
(1997). Deleterious actions of gossypol on bovine spermatozoa, oocytes,
and embryos. Biol. Reprod. 57:901-907.
Brubaker, C. L. and J. F. Wendel. 1993. On the specific status of Gossypium
lanceolatum Todaro. Genetic Resources and Crop Evolution 40:165-170.
Collonier C, Berthier G, Boyer F, Duplan M-N, Fernandez S, Kebdani N,
Kobilinsky A, Romanuk M, Bertheau Y. Characterization of commercial GMO
inserts: a source of useful material to study genome fluidity. Poster
courtesy of Pr. Gilles-Eric Seralini, Président du Conseil Scientifique du
CRII-GEN, www.crii-gen.org
de Vries J, Wackernagel W (1998) Detection of nptII (kanamycin
resistance) genes in genomes of transgenic plants by marker-rescue
transformation. Molecular and General Genetics 257:606-613
Fryxell, P. A. 1979. The Natural History of the Cotton Tribe (Malvaceae,
Tribe Gossypieae). Texas A&M University Press. College Station and London.
245 pp.
Greene, A.E. and Allison, R.F. (1994). Recombination between viral RNA and
transgenic plant transcripts. Science 263, 1423-5.
Ho MW, Ryan A and Cummins J (2000). CaMV35S promoter fragmentation
hotspot confirmed and it is active in animals. Microbial Ecology in Health
and Disease, 12, 189.
Kohli A., Griffiths S, Palacios N, Twyman R, Vain P, Laurie D and Christou P.
(1999) Molecular characterization of transforming plasmid rearrangements
in transgenic rice reveals a recombination hot spot in the CaMV 35S
promoter and confirms the predominance of microhomology mediated
recombination" Plant.J. 17,591-601
Kirk, JH and Higginbotham GE. Pima Cotton, Gossypol and Dairy Cattle - Is it
a Bad Combination. The Western Dairyman 80(8):32-33, 1999
Kowalchuk G.A., Bruinsma, M., and van Veen, J.A. (2003) Assessing
responses of soil microorganisms to GM plants. Trends in Ecology and
Evolution Vol.18 No.8 .
Nikolich, M P Hong, G Shoemaker, N B and Salyers A A (1994) Evidence for
natural horizontal transfer of tetQ between bacteria that normally colonize
humans and bacteria that normally colonize livestock. Appl Environ
Microbiol. 60(9): 3255-3260.
Pinkel D, and Albertson D.G. (2005). Comparative genome hybridization.
Annual Review of Genomics and Human Genetics, Vol. 6, Pages 331-354
Quist D and Chapela IH. (2001) Transgenic DNA introgressed into
traditional maize landraces in Oaxaca, Mexico. Nature, 414, 541-3
Sullivan, J.L., J.T. Huber, and J.M. Harper. 1993a. Performance of dairy cows
fed short staple, Pima, and cracked Pima cottonseed and feed
characteristics. J. Dairy Sci. 76:3555-3561.
Vaden V.S. and Melcher, U. (1990). Recombination sites in cauliflower
mosaic virus DNAs: implications for mechanisms of recombination. Virology
177, 717-26
Van Deynze, A. E., Sundstrom F. J.and Bradford K. J. (2005). Pollen-
Mediated Gene Flow in California Cotton Depends on Pollinator Activity
Published in Crop Sci 45:1565-1570
Wendel JF. 2000. Genome evolution in polyploids. Plant Molecular Biology
42: 225-249.
Wendel JF, Cronn RC. 2003. Polyploidy and the evolutionary history of
cotton. Advances in Agronomy 78: 139-186.
Wendel, J.F. 1989. New World tetraploid cottons contain Old World
cytoplasm. Proceedings of the National Academy of Science USA 86: 4132-4136.
Witte W: (1998) Medical consequences of antibiotic use in agriculture.
Science, 279:996-997. Wintermantel, W.M. and Schoelz, J.E. (1996).
Isolation of recombinant viruses between cauliflower mosaic virus and a
viral gene in transgenic plants under conditions of moderate selection
pressure. Virology 223, 156-64.

Socio-Economic Assessment
Agbios (www.agbios.com)
Stephen Greenberg The Venoms of Scorpions and Spiders Global Agriculture
and Genetically Modified Cotton in Africa, 2004, African Centre for Biosafety,
www.biosafetyafrica.net
Oxfam Briefing paper 30 Cultivating Poverty: The Impact of US Cotton
Subsidies on Africa.
Statistics from Cotton South Africa www.cottonsa.org.za