Docket No. 03-101-1
Regulatory Analysis and Development
USDA APHIS PPD
Station 3C71
4700 River Road Unit 118
Riverdale, MD 20737-1238

To whom it may concern:

The Union of Concerned Scientists (UCS) is grateful for the opportunity to comment on USDA's review of a new genetically engineered product-glyphosate-resistant creeping bentgrass. UCS is a non-profit partnership of scientists and citizens combining rigorous scientific analysis, innovative policy development, and effective citizen advocacy to achieve practical environmental solutions.

BACKGROUND
UCS is responding to a January 5, 2004, Federal Register notice announcing the receipt by USDA of a petition from the Monsanto and Scotts companies requesting deregulation of genetically engineered herbicide-resistant creeping bentgrass (Agrostis stolonifera event ASR368) (1). USDA requests comments on its preliminary assessment of the risks of the product that accompanies the notice.

After receiving comments on the preliminary risk assessment, the department will prepare an environmental document under the National Environmental Policy Act (NEPA) assessing the potential environmental impacts of a decision to deregulate the product. That assessment is to be made available for public comment before USDA decides whether to approve the petition. A decision to deregulate would help clear the way for Monsanto and Scotts to commercialize engineered bentgrass.

The bentgrass variety at issue, also known as Roundup Ready (RR), was engineered to withstand the lethal effects of glyphosate (Roundup) herbicides. According to the Monsanto/Scotts petition, the product is intended for use in commercial grass seed production and golf-course turf management (2).

Creeping bentgrass is unlike any other crop USDA has considered for deregulation. Where previous candidates have been annual crops dependent to varying extents on humans for successful propagation, creeping bentgrass is a perennial crop which establishes without cultivation in a wide variety of habitats, including urban, agricultural, and wild habitats. It reproduces both sexually through seeds and vegetatively by horizontal stems (stolons) that produce roots, and subsequently new bentgrass plants, at various points along the stem (hence its name" creeping").

In addition, creeping bentgrass has numerous compatible relatives in the United States with which it can form natural hybrids. The likelihood of hybridization is enhanced by the fact that bentgrass is wind pollinated.

If deregulated, RR creeping bentgrass could be planted within pollinating distance of many populations of native and naturalized sexually compatible relatives. As a result, for the first time, the deregulation of an engineered crop may lead to the establishment of transgenes in potentially dozens or even hundreds of wild plant populations in the United States.

Under the Plant Pest Act regulations that are the current basis for the USDA regulation of genetically engineered plants (3), the department must determine whether engineered plants present risks as plant pests in the United States. In addition, USDA considers the possibility of a broad set of potential environmental harms under NEPA.

This application from Monsanto and Scotts raises important scientific, policy and legal issues. We congratulate USDA for its thorough, deliberative, and transparent approach to making a decision on this precedent-setting application. In particular, we appreciate the department's acknowledgement of the unique challenges presented by this petition and solicitation of comments from the public on a range of questions relevant to these challenges.

POTENTIAL HARMS OF DEREGULATING GLYPHOSATE-RESISTANT CREEPING BENTGRASS
The USDA's preliminary risk assessment and other information suggest a number of important potential harms that must be thoroughly evaluated in advance of any decision to deregulate glyphosate-resistant bentgrass. These include:

• Exacerbating the harmful effects of existing bentgrass weeds,
• Creating new or worse weeds created through hybridization with wild relatives,
• Creating new or worse weeds that are not related to bentgrass,
• Contaminating the seed supply for nonengineered bentgrass and related grasses,
• Causing harm to people and the environment as a result of increased exposure to highly toxic herbicides, and
• Causing unknown harms as a result of genetic engineering.

Exacerbating the harmful effects of existing bentgrass weeds
Glyphosate-resistant creeping bentgrass may become a more-difficult-to-control weed than its nonengineered counterpart. Creeping bentgrass is a well known weed in areas where glyphosate is likely to be used. For example, spot treatment with glyphosate is currently a common control measure for creeping bentgrass weeds in lawns (4). Glyphosate-resistant bentgrass that makes its way to lawns and other areas such as parks and playgrounds will be more difficult to control because homeowners and managers will not be able to use glyphosate.

Resistance genes may move into bentgrass populations as a result of cross pollination between RR bentgrass and existing populations of creeping bentgrass. Or vegetative propagules of RR bentgrass may find their way from golf courses to lawns.

Creating new or worse weeds through hybridization with wild relatives
As the USDA preliminary risk assessment points out, creeping bentgrass interbreeds with a number of native and naturalized grass species in two genera, Agrostis and Polypogon, creating in some cases weedy hybrids. If RR creeping bentgrass were deregulated, it would only be a matter of time until the RR gene would show up in such hybrids. The transgene may become established in wild populations where it provides a selective advantage, that is, where glyphosate is used often. Where the hybrids are already weeds, a new ability to resist glyphosate may make them more difficult to control. In some cases, glyphosate resistance may convert grasses that are not now weeds into weeds.

Creating new or worse weeds that are not related to bentgrass
Widespread adoption of RR bentgrass is likely to lead to a major expansion in the use of glyphosate. Just as the increased use of glyphosate herbicides in RR field crops has selected for glyphosate-resistant weeds, greater use of the herbicide on golf greens and other places will select for populations of resistant weeds.

Since the introduction of RR crops in the United States in mid-1990's, enormous quantities of Roundup herbicides have been applied to millions of acres of RR soybean, corn, cotton, and canola fields. Consequently, more and more farmers are reporting problems with Roundup-resistant weeds, which have evolved, not from gene flow from RR crops, but from greater selection pressure resulting from increased use of glyphosate. According to Syngenta, a competitor of Monsanto, three states reported glyphosate-resistant weeds in 2000 and by 2003 2.3 million acres of U.S. cropland were infested with Roundup-resistant weeds (5).

Deregulation of RR creeping bentgrass will contribute to the selection pressure for glyphosate-resistant weeds on golf courses and other areas where the new product is used.

Contaminating the seed supply for nonengineered bentgrass and related grasses
If traditional seeds for creeping bentgrass and other grasses are contaminated with the glyphosate-resistant bentgrass, RR bentgrass will be perpetuated by plant breeders and could emerge in unexpected places at unexpected times far into the future. UCS has just published a study, Gone to Seed (6), which shows pervasive, low-level engineered contaminants in traditional seeds of corn, soybeans, and canola in the wake of widespread adoption of engineered versions of those crops. Although the Monsanto/Scotts petition indicates steps the two companies will take to prevent contamination of traditional grass seeds by RR bentgrass seeds, the measures may not be enough, given the wind blown nature of bentgrass pollen, the difficulty of controlling such small seeds, and the likelihood of human error. Geographic isolation of all phases of RR bentgrass seed production many miles away from all other grass seed production (not just bentgrass seed production) sites and facilities may be needed to prevent contamination.

Causing harm to people and the environment as a result of increased exposure to highly toxic herbicides
If users of RR bentgrass follow the same path used by growers of other RR crops, the evolution of glyphosate-resistant weeds will mean increased use of highly toxic herbicides. To control glyphosate-resistant weeds emerging in RR-crop fields, agronomists and others are recommending the use of herbicides like paraquat and 2,4-D (7), which are associated with serious impacts on humans and non-target organisms (8). To the extent that the users of RR bentgrass follow the lead of other RR-crop growers in their response to glyphosate-resistant weeds, golf courses, parks, and other areas, including people and other organisms in those areas, could be exposed to increased amounts of harmful herbicides.

Causing unknown harms as a result of genetic engineering
Genetic engineering is a powerful new technology that has been in use for only about two decades in the field of agriculture. Engineered crops have been used in the environment in large numbers for less than ten years. Few ecologists have had the resources to study the impacts of engineered organisms in the environment. As far as we know, no systematic long-term ecological studies of engineered organisms are underway in the United States. But, we do know from a handful of studies that genetic engineering can cause unexpected effects even in the short term, for example, increasing outcrossing in herbicide-resistant Arabidopsis (9) and gene-silencing in herbicide-resistant oilseed rape (10). Little work, either long or short term, has been done on the impacts of transgene flow into wild plants.

As a result, we have little information on which to predict the potential impacts that the deregulation of engineered bentgrass would have among the many wild populations of plants to which the transgene for glyphosate-resistance would likely flow. Deregulation would launch an unprecedented experiment on wild populations-an experiment whose effects, some of which may be harmful, may not be known for years.

LEGAL CONSIDERATIONS IN DEREGULATING GLYPHOSATE-RESISTANT CREEPING BENTGRASS
The bentgrass petition presents an opportunity to revisit some of the legal questions that have long surrounded the use of the Part 340 regulations (11) that continue in force implementing the Plant Pest Act (now repealed).

The legal essence of the decision to deregulate a plant under Part 340 is the determination that it is not a plant pest, and therefore no longer subject to the Plant Pest Act regulations. One of the peculiarities of the Part 340 program is that most of the crops, grasses, and trees regulated under it are not traditional plant pests-indeed they are the economically important plants that pests attack. Since most crops are not plant pests to begin with, the determination that they are not plant pests often makes little scientific or legal sense.

In practice, Part 340 has been used to conduct reviews of environmental impacts ranging far beyond the arena of plant pests. In the past USDA has approved most applications for deregulation, so jurisdictional issues and their implications have received little attention.

Bentgrass, however, may turn out to be a case where deregulation and widespread use of the product would cause environmental impacts which are not related to plant pestness. In that case, the application will raise the question of USDA's authority to deny a petition to deregulate under the Part 340 regulations. If the department should deny the bentgrass petition, the petitioners may sue on the grounds that bentgrass is not a plant pest-and win.

RECOMMENDATIONS
UCS urges USDA to convene a panel of experts to consider potential environmental harms that may be associated with the deregulation of RR creeping bentgrass. The department should delay a decision on the new product pending the submission and evaluation of any new data that the panel determines is necessary to evaluate those harms.

As noted above, USDA is confronting unique challenges with the Monsanto/Scotts petition. As a result, the department should seek outside expert advice in a systematic way that ensures a thorough evaluation of the data needed to assess the risks of this new product. While we applaud USDA's attempt through the January 5, 2004, Federal Register notice to solicit information on a variety of topics, such a process does not ensure that the department will receive all the information it needs.

UCS urges USDA to convene a panel of scientists expert in the disciplines relevant to the RR creeping bentgrass petition to aid the department in determining what harms RR bentgrass may cause and what data are needed and how those data should be developed. The panel of experts should be assembled to assure a balance among biases about genetic engineering. The department should make the panel's deliberations open to the public.

Because of the precedent-setting nature of the bentgrass petition, particularly the impact that deregulation would have on transgene establishment in wild plant populations, UCS urges USDA to delay a decision on RR bentgrass until after the department evaluates any new information submitted in response to the panel's recommendations.

UCS urges USDA to delay a decision on the bentgrass petition pending establishment of new regulations under the Plant Protection Act.

USDA is currently undertaking a process to repromulgate regulations governing genetically engineered plants under the broad authority of the Plant Protection Act of 2000. We recommend that the department delay any action on glyphosate-resistant bentgrass until it has in place a program under which it can confidently regulate genetically engineered plants, including barring them from use in the environment, if necessary.

Sincerely,

Margaret Mellon, Ph.D., J.D., Director
Food and Environment Program

Jane Rissler, Ph.D.
Senior Staff Scientist
Food and Environment Program

Notes:
(1) USDA Animal and Plant Health Inspection Service. 2004. Monsanto Co. and The Scotts Co.; Availability of petition for determination of nonregulated status for genetically engineered glyphosate-tolerant creeping bentgrass; Request for information and comment. Federal Register 69:315-17, January 5.
(2) The Scotts Company and Monsanto Company. 2003. Petition for determination of nonregulated status: Roundup Ready creeping bentgrass (Agrostis stolonifera L.) event ASR368, April 11.
(3) 7 CFR part 340.
(4) Koski, T. 2002. Identification and management of perennial weedy grasses. On the Colorado State University website at http://csuturf.colostate.edu/pdffiles/Perennial%20Weedy%20Grasses%20ID%20and%20MGT.pdf, accessed March 5, 2004.
(5) Syngenta. No date. "Resist resistance: Immunity to glyphosate is a rising threat." Special Agronomic Series, www.agweb.com/images/pubs/sm4001.pdf, accessed March 5, 2004.
(6) Mellon, M. and J. Rissler. 2004. Gone to Seed: Transgenic Contaminants in the Traditional Seed Supply. Cambridge, MA: Union of Concerned Scientists. On the UCS website at www.ucsusa.org/GonetoSeed.html, accessed March 5, 2004.
(7) Syngenta. No date. Gramoxone. Resistance management, On the Syngenta website at www.syngentacropprotection.com/prod/herbicide/gramoxonemax/index.asp?nav=resistmgmt. Gramoxone is a trade name for paraquat.; Forbes, B. 2004. Purdue expert advises on strategy for subduing superweeds. Purdue News, January 30, on the Purdue University website at news.uns.purdue.edu/uns/html3month/2004/040130.johnson.super.html, accessed March 4, 2004.
(8) Briggs, S. 1992. Basic Guide to Pesticides: Their Characteristics and Hazards. London: Taylor and Francis, Ltd., pp. 124 and 172; Klaasen, C., ed. 1996. Casarett and Doull's Toxicology: the Basic Science of Poisons, Fifth edition New York: McGraw-Hill, pp. 671-76.
(9) Bergelson, J. 1998. Promiscuity in transgenic plants. Nature 395:25, September 3.
(10) Al-Kaff, N. et al. 2000. Plants rendered herbicide-susceptible by cauliflower mosaic virus-elicited suppression of a 35S promoter-regulated transgene. Nature Biotechnology 18: 995-99, September.
(11) 7 CFR part 340.