(prepared from the videotaped discussion*)
Moderator: John Vandenberg, USDA/ARS
Panel Members:
Ray Carrruthers, USDA/ARS National Program Staff
Rebecca Goldburg, Environmental Defense Fund
Davis Pimentel, Department of Entomology, Cornell University
Alan Wood, Boyce Thompson Institute
Vandenberg:
Risk assessment is a formally defined field of study which has applications to the issues of today, but this field is not the approach we are taking today. In this one-hour discussion we will not be able to give comprehensive coverage to the issues, but we hope to contribute to or expand perspectives and knowledge of them and to stimulate thinking and interaction between the audience and the panelists.
We will look at risks in three ways:
Before we begin, the panel and I will introduce ourselves:
John Vandenberg:
I am a Research Entomologist for the USDA Agricultural Research Service for the past 15 years. I've been in Ithaca for the past 3 years. Brian Arroyo, who is on the program and is from USDA, Fish and Wildlife Service Endangered Species program, was unable to come to this conference due to current budget uncertainties at that agency.
David Pimentel:
I am from the Entomology Department at Cornell University. I have been involved in biological control theory and practice for over 30 years.
Ray Carruthers:
I am with the USDA/Agricultural Research Service, currently as
National Program Staff Leader for Biological Control. My current
administrative responsibilities include overseeing biocontrol
programs across the country and at international locations in
France, Argentina, Australia and China. My formal training is in
entomology and population dynamics and I have filled a biocontrol
role in a number of projects from the conservation of
natural enemies to augmentation to classical introductions,
including the use of exotic natural enemies of insect pests.
There are important decisions that need to be made on these
issues, so I'm here as an individual, to represent our agency,
and to learn from others here.
Rebecca Goldburg:
I am with the Environmental Defense Fund in New York. I am an ecologist by training with a background in plant and insect ecology. I've done a lot of work on risks involved with genetic engineering, much of which is on organisms intended for pest control.
Alan Wood:
I am with Boyce Thompson Institute. I'm an insect virologist involved in genetic engineering in insect viruses. I have also been involved in field releases of recombinant viruses here and in Cape Cod.
Vandenberg:
The topic of this discussion is "New associations and risks at both organism and gene levels that result from our efforts in biological control." Definitions will be in the broadest terms - any combination of organisms and/or genes, new and/or alien, and any new associations, broadly defined. We are not assessing risks; we are evaluating them.
The first question is: "What are some realized or potential risks and how might we approach them?"
Wood:
At the morning meeting most people seemed to agree that the risks of biological control are minimal compared to the potential benefits. A point was made that we need to take a proactive rather than a reactive approach to safety issues, but we also need to do things as a society that preserve our technologies. Probably in the 1940's a similar debate took place when people said, "Look, we have all these compounds for pest control that make things much less risky for man and the environment." And, I believe they were right.
In the assessment of biological control agents today, whether they are recombinant or natural organisms, we need to learn from history: we need to take a pro-active approach to assessment of risks rather than a re-active approach. I'll give an example of how I was caught up in the paradigms of biological control in which we all were sitting around telling each other how good this is when it occurred to me what a mistake could be made. This table shows the concerns we had:
| List of Potential Risks Associated with Biological Control | |
|---|---|
| Safety and Health | Humans and other vertebrates Non-target invertebrates |
| Displacement | Natural virus populations Ecological niches |
| Persistence | Mitigation |
| Genetic Stability | Genetic rearrangement Transfer |
The first issue on this table, safety and health concerning vertebrates, is in actuality a non-issue as far as baculoviruses are concerned. They do not affect vertebrates and they have very restricted host ranges that do not include any beneficial insects.
The vertebrate testing that was done was very sophisticated in order to assure us there is no problem with baculoviruses in terms of humans, as shown below:
| Vertebrate Safety Testing |
|---|
| DNA:DNA hybridization |
| Viral-induced protein synthesis |
| Reporter gene expression |
| Chromosomal aberrations |
| Sister chromatid exchanges |
Non-target invertebrates: Host range data was based on exposing invertebrates to the virus and looking for signs and symptoms. It does not address virus replication, which often occurs in the absence of signs and symptoms and is not an issue with non-recombinant viruses. The problem is with recombinant viruses. An apparent infection with a virus expressing some type of toxin, will become very apparent, and it becomes an important issue.
So we told each other that there were no host range problems and that this was very safe, but the basis of the conclusions was flawed.
Pimentel:
I like the session title, "New Associations," because this is what I have focused on for about 30 years. I have investigated better ways of selecting natural enemies for biological control and through my work in ecological genetics, the research suggested that we should use the principle of new associations or interactions between the parasite and/or predator and their hosts. Two examples:
To summarize: of all the successful instances of biological control in the world, about 40% are due to new associations. This is true even though new associations are the second line of defense after classical biological control has been tried.
Relative to the topic of risk and based on new association invertebrates introduced for biological control worldwide, there has not been a single environmental problem. This is true even if the natural enemy was introduced from another entirely different host or prey onto a related host. The only environmental problem that I know of relating to invertebrates was in classical biological control when an insect was introduced for weed control in East Africa. The specificity relating to the introductions of new associations has been equally effective and specific to classical biological control. Thus, the risks, if done properly, are minimal in the new association relationship. The technique is being adopted and implemented around the world.
Carruthers:
The USDA is interested in developing and implementing
environmentally sound, sensible pest management strategies as
best we can. We look at all types of controls, from genetic
engineering of plants to classical biological control, and we've
shifted drastically away from the wide spread and continual use
of chemical insecticides. However, we still feel that chemicals
are very important weapons to have in our arsenal to protect
against pests.
We view new associations as a very important tool, that has been used for a number of years, that can help us to manage many pest populations. They have been used not only on exotic pests, but on native pests. Using new associations has been done for nearly the entire span of history of biological control. It's been done with weed agents, with insect agents, and with plant pathogen agents. For example, new association approaches were used against grasshoppers in this country around 1907 or 1908.
As Dave pointed out, as long as this has been going on with insects and microbes (about 1000 new associations documented), there have been no extensive problems identified. So, I don't think new associations is the problem. I think that host specificity is the issue for all biological control programs.
Goldburg:
I want to talk about genetic engineering and the new directions for genetic engineering that are likely in the future. Today, most of the genetically engineered organisms that have been developed for agriculture and plants, and although there have been a number of concerns expressed about genetically engineered plants, many individuals minimize the concerns, arguing that most crops have been so domesticated by selective breeding that even if you add a new gene to them, you're unlikely to turn them into a pest, and that, at least in the US, the majority of our crops don't have wild relatives with which they can hybridize and so they can't transfer their acquired genes via pollination. I, for the most part, agree with this argument.
But in the future we can look to a much more broad variety of organisms that are going to be routinely genetically engineered. The two areas that I particularly think of are aquaculture and forestry.
The USDA reports that aquaculture is the fastest growing segment of agriculture. It is growing at the rate of 10% per year for a variety of reasons, one of which is the collapse of some of the world's fisheries; another is the strong market for seafood. These are important financial incentives for aquaculture. However, fish, for the most part, have not been changed by selective breeding. There are some catfish and rainbow trout that have been domesticated, but for the most part, fish and shellfish are pretty much wild organisms and there's a great deal of interest in changing them to make them faster-growing, more productive, disease-resistant, etc. I think the risks of genetic engineering are going to be a great deal different between fish and aquaculture vs. already-domesticated plants and agriculture because we will be genetically engineering organisms that are essentially wild to begin with and often have to be grown right where the wild relatives are. So I think the odds of transferring traits that result in ecological problems are much higher.
I think that many of the same things could be said about forestry. There's now increasing interest in genetic engineering. For example, a number of timber companies are funding forestry research at Oregon State University's forestry research center, and genetically engineered Populus which is glycosate-tolerant, is now being field-tested. Clearly, a genetically engineered forest tree will in most cases be a wild organism, not a cultivated one. It has conspecifics nearby to which to transfer genetic material, so that another round of risk assessment will be necessary as we start working on this new class of organisms.
Vandenberg:
Each of you, please speak to the current level of risk code addressed by current regulations. Or, rephrased, "Are current regulations adequate?"
Carruthers:
This morning in Del's talk, it was pointed out that the
regulatory world regarding biological control is in a state of
confusion and disarray. There are a number of laws set up to
regulate things other than biological control. The Plant Pest Act
is primarily what allows that to happen. FIFRA was set up to
monitor pesticide control and NEPA is a primarily environmental
regulation. My impression of the process is that it's not
well-suited to deal with biological control organisms at all.
Most of the scientists working in biological control, whether federal or state or practitioners in industry, are really at odds with the system as it exists. The system inside APHIS has been in a state of flux for the past 5 years, and I don't think that it is addressing either the risks appropriately in the area of biological control or comparing the risks with associated benefits. So, I believe that we need a total restructuring within the regulatory world to make the system sensitive to both the environmental needs and the benefits associated with using biological control on pests.
Goldburg:
There are 2 problems I see in the regulatory world concerning genetic engineering. One is genetic engineering of fish. There is now no federal agency with any authority to regulate genetic engineering of fish. USDA's CREES has established some voluntary performance standards for research for genetically modified fish and shellfish that are quite excellent, but they are entirely voluntary. This area needs some attention.
In terms of transgenic crops, I am quite disturbed by a rule proposed last fall by APHIS. To give some background, APHIS administers a regulatory program under the Plant Pest Act by which it issues permits for field tests of genetically engineered crop plants. In 1993, APHIS proposed a finalized rule which removed permit requirements and substituted very basic notification requirements for the 6 most frequently field-tested transgenic crop plants: corn, soybeans, cotton, tomatoes, potatoes, and tobacco. At the time that rule was proposed, I very much supported it. These were crops that were heavily domesticated, that don't have wild relatives in the US, and for which it was inappropriate to require a permit for every single field test of them. Unfortunately, APHIS is taking this concept a step, or leap, further and is close to finalizing a rule that will only require permits for field tests of transgenic crops if they are on a noxious weed list or meet one of the few other narrow criteria. So, basically, anything can be field tested without APHIS review, and that includes forest trees, for example, and plants that will clearly hybridize with wild plants. I find this proposal very disturbing.
I think there is a long way to go in rationalizing our regulatory system. There clearly are some areas where we need to ratchet up the regulations as well as perhaps facilitate biocontrol with more suitable regulatory requirements.
Wood:
We get ourselves in trouble when we make new laws concerning the release of recombinant biological control agents There probably should have been a restructuring 20 years ago. I work primarily with EPA, and if you try to read the letter of the law, it's currently a mishmash. But the system works, as the people who have been involved with regulating are a very reasonable group. However, one problem is that APHIS does not deal with environmental issues - only with plant pest issues.
In my opinion, if there is one thing wrong with the regulatory system today, the most glaring issue concerns pest issues. This is why pesticidal plants are now being regulated by the EPA. If APHIS doesn't respond to environmental issues, we're going to have some more rules where the regulation of biological control is going to get pulled into EPA in other ways, and I think that would be very unfortunate. It would create another layer of regulation. But the impetus to change will have to be by the agency.
Carruthers:
One of the things that happened over the past years that has been
problematic is that APHIS tried to make a shift a few years
ago to doing more in-depth environmental assessments, but they
didn't really have the personnel or the expertise in place for
scientific assessments to be made. Decisions were instead based
more on politics or legal challenge. So we got away from
having a reasonable, scientific-based assessment process.
In the past 6 months APHIS has changed tactics again, and now is asking other agencies to handle there own NEPA assessments if the organism is not a plant pest. Clearly, we need something in-between no decision and clearing all organisms regardless of potential threat. We need a system that helps scientists in developing a program, and to assess what the true risks are, because we, as biocontrol practitioners, don't have the full perspective. We need to link the people who have different perspectives. We need to know what are the environmental concerns and the other regulatory issues that have to be addressed.
In years past, APHIS was concerned only with how programs in insect pathology affected nontarget insects such as honey bees. In our own programs ARS went well beyond those issues and addressed many dimensions of the problem. But the regulatory system did not drive that, and the current system is deficient in many ways as well.
We need a much more rationally-based system that leads us to ask the right questions, do the right research, and when there are conflicts of interest, have a very systematic way of dealing with them instead of having political or legal issues determining the final outcome. A scientific resolution or process is clearly needed and the Strawman proposes this. It may be too far in one direction in some areas, but it's a good model and we can work with it to gain a balanced system. I think that's the direction we ought to move in.
Goldburg:
Briefly, I think one of the problems with APHIS regulations is that the agency is operating under an antiquated statute, the Plant Pest Act of the 1920's, which was designed to do different things than it's doing today. So, in order to satisfy their lawyers, APHIS does things sometimes that seem a little silly. That may seem a little crazy to scientists, but that's what APHIS must do legally.
Pimentel:
Two points quickly: As far as native pest problems, in the United States, 60% of our pests are native pests and so classical biological control will not work. This is true also in South Africa, where 60% of their pests are native, and in Europe where the figure is 80%. Thus, classical biological control will not work with native pests.
Point two regards genetic engineering. We can't be too careful, whether it is in biological control or new association introductions or genetic engineering. We studied all crops that had been intentionally introduced into the United States that have become pests. It turned out there were 128 species of crops that were intentionally introduced that had become pests and weeds! Thus, we can't be too cautious.
Vandenberg:
I would like you to address particular examples of biology of new associations - things that work and things that don't work. For example, a biological control project involving tamarisk, an introduced weed that now provides habitat in southwestern rivers for an endangered fish species. Can you address this particular case as well as the question in general?
Carruthers:
I don't know the details of this project. Tamarisk was introduced
as an ornamental species and also was used for windbreaks in
agricultural settings. It very quickly invaded riparian areas and
now grows along streambeds and washes and has squeezed out
a large number of native plants from many areas and is a distinct
problem over the entire western area of states and into the
northern tier of states as well.
ARS initiated a biological control effort approximately 10 years ago based in our lab in Temple, TX. They have looked at many of the natural enemies of tamarisk and found a number that have quite a potential for success, very specific to the tamarisk, and which won't have effects on any other plant species that we know of. But there are two associated problems. One is the fish problem that John just mentioned, and the other is that tamarisk is a nesting site for a rare and endangered species of bird since tamarisk has replaced the natural vegetation that exists in those areas where this bird once nested. FWS will make a determination as to whether we go forward with these releases or not.
If we were to release these agents, we wouldn't expect them to totally eliminate the tamarisk plant, and if it were to do so, it wouldn't happen immediately. Biological control works gradually, so there is sure to be a transition period.
Another problem is that as you take out the tamarisk, what will replace it? Another weed species or the native species returning?
The question isn't biological control in isolation, but overall land management. As you put biological control practices into action, you put other types of activities into action as well to displace the exotic species with something beneficial. You don't just release the agents and call that the show - it's a larger process. The same occurs in agricultural situations. We're interested in managing the overall ecosystem and not just the pest
Vandenberg:
A friend who is a sedimentary geologist in Utah tells me that another part of the issue is the dams on the rivers which prevent periodic flooding. This has allowed tamarisk to grow where it otherwise wouldn't, so it's not necessarily just a matter of a native species replacing the tamarisk, but the issue of changing of the habitat with the dams.
Pimentel:
The whole issue is decision making. For instance, mesquite became a pest on rangelands because it was thorny and the cattle couldn't eat it. Biocontrol agents were considered to control it, but this was opposed because mesquite made good biomass charcoal. A thornless species of mesquite from South America has been discovered and possibly could be introduced to displace the thorny US native mesquite.
The questions become complicated. In California, poison oak is established on various slopes and biological control was recommended. Poison oak can make people miserable because it causes a severe rash like poison ivy. However, it helps control erosion. No one knows what would replace poison oak if it were controlled. Thus, I recommended that the project be turned down.
Wood:
How much do we know about the ecological relationships of organisms such as the tamarisk and its biological control agents before we bring them in? In talking about new associations and risk assessments, it's important to note that we're not just assessing the risk to the targetthe native or exotic host involved. The risk is linked to the other non- targets in that ecosystem. By definition, new associations are not species-specific. They didn't co-evolve with that species and there can be a wider range in host specificity. So I think the risk is the same whether you're talking about new associations or old associations. It's that non-target pool of host species that you're most concerned about.
Vandenberg:
My question to you, then, would be, "Are there ways to evaluate host range outside of the laboratory?" I assume we all agree that laboratory evaluations of host range provide one set of data, but they may be limited. So, if the only way to evaluate it is to introduce it, then what? Or, are there other ways?
Pimentel:
Caution! Great caution! You never really know initially. When we make these introductions it's forever, and we better be cautious! This applies to genetic engineering as well. To recover all organisms after they've been introduced is impossible.
Goldburg:
I think you can't say too much for having people follow a step-by-step process. First you start by testing an organism in the lab. When you go to the field, you start on a small scale, so that there is some control should things go wrong, and so on. Even though some folks might get impatient, that is the way we have to go.
Vandenberg:
As Del said this morning, you acknowledge that there are risks, you identify potential risks, you accept that what you're doing has risks and at some point you accept that you can't guarantee success - that you can't guarantee that you won't have failure or disaster. But you put in place the mechanisms to mitigate in the future.
To prepare for questions, I want to provoke some thought. None of the panel members represent industry or any aspect of it. With genetic engineering and their products becoming an important part of biological control, an arena is opening for commercial development and exploitation for a profit. Ray spoke of regulatory systems that involved the participation of scientists. Should there also be regulatory systems that involve the part of industries that would profit from these regulatory processes? Also, what about the role of industry in forging changes to these regulations?
A common perception is that there are industries with deep pockets represented here - pockets deep enough that Congress might be influenced in ways academic scientists and government bureaucrats cannot do. I think that is worth thinking about and throwing into this pool about regulations, risk evaluation, and risk acceptance.
Dick Root, Department of Ecology and Systematics:
I'm disturbed that more of a distinction was not made between native targets and introduced targets because native species are very special. They are the product of a long evolution in which they have become integrated into the ecosystems and as a result, their interactions are very subtle, making it extremely difficult to identify in advance the cascade of implications that might occur should exotic species be introduced. It does not seem to be a good reason to continue with introductions that we've already been making introductions for a long time or that many of our pests are native. This seems to be a dangerous practice, to me.
Pimentel:
Again, 60% of our pests are native: the potato beetle, apple maggot, cabbage looper, and many more. However, it's a tradeoff, Dick. We could use pesticides to reduce these pests, but that has serious risks also, not only for the target pest but it has a broader spectrum impact on the environment than using a biological control agent. I see fewer environmental problems associated with using biological control agents to control native pests than with controlling native pests with pesticides.
Carruthers:
I don't think anyone involved in biological control
is looking to disturb natural ecosystems more than they have to.
Pests are a problem - they produce major economic losses.
The biggest controversy that I have been involved in is the grasshopper project which was aimed at a native pest. Insecticides were being applied on approximately 10 million acres per year over many, many years in a row. This has been a continuing problem that has not been adequately addressed by the continued application of those chemical pesticides. This included areas that had many rare and endangered species, many of which were insects. The development of biological control that was more specific to grasshoppers was, in my eyes, a very environmentally pro-active process. Clearly, disruption will occur and risks exist. This evaluation is important: What are the risks? What are the potential benefits?
Some people would say we don't need to control grasshoppers at all, and many of us might agree, at least partially, with that statement. But, there are many who don't agree, and it then becomes a political issue. Grasshoppers are presently deemed a major pest that the USDA is required by law to address.
Wood:
With the natural pests that we have, we should be very sensitive to how we try to control them, but we must take into account that agriculture presents a very unnatural ecosystem in which they exist. If we restrict them to that, it is a less severe problem.
Root:
But the grasshoppers would be an exception.
Carruthers:
I disagree. A large percentage of plants in the rangeland
ecosystem are not native plants. The large numbers of vertebrate
herbivores active in the rangeland ecosystem are not native
vertebrate herbivores. We've changed the overall dynamics of the
system and we can't pretend that system is in balance in the same
way as when these grasshopper complexes evolved.
The rangeland systems aren't totally independent systems, as well. They border major agricultural areas. The grasshoppers develop into very high levels on the rangelands, and they move into agricultural crops and have to be sprayed very much more intensively than the rangeland. It's not a simple problem and there's no simple solution.
Vandenberg:
And we prevent fire, which is a natural part of that ecosystem which would periodically kill grasses and probably grasshoppers, too.
Goldburg:
Before we use biological control or pesticides to control native pests, we should look to see what other options we have. Pest control isn't just something we should automatically do when a problem erupts. We are speaking, at this conference, of bringing pest control decisions together. If there are fundamental questions, we need to go back and look at the basic issues, such as, should we control grasshoppers? If we're willing to ask these basic questions, which may be politically quite difficult, we may not be forced into such hard decisions.
Carol Glenister, IPM Laboratories:
Would Ray Carruthers please summarize how easily the regulatory process can be disrupted at APHIS? I am referring to a recent situation involving fungi and grasshoppers?
Carruthers:
One of the major problems with the regulatory system is there is
no existing conflict resolution process. When a challenge was
made to a proposed action, APHIS didn't know what to do and so
the overall process was stopped and nothing was done for about
3-4 years. No permits were granted and the process only restarted
when there was a major agricultural problem. The process had
stopped not only for the grasshopper project, but for all
biological control.
Glenister:
This had a major impact on private industry because APHIS, having looked at the regulations again, decided that the permit process for interstate movement of all beneficial insects had to be reevaluated and revamped. As a result, we faced all kinds of difficulties that we'd never seen before.
Bruce Black, Cyanamid:
Mr. Carruthers, you've been charged to change national agricultural practices to 75% IPM by the year 2000 - only 4 years away. Many of the examples we have celebrated today are wonderful but are marginalized. The role of biologicals is only about 1-2% of the world's insecticidal use, and the big markets are in low cost treatment of cotton, corn, beans, etc. with herbicides, insecticides, and fungicides. You have a long way to go. So, I'd like to know what your plans and strategies are and how you will achieve this in just 4 years.
Carruthers:
First, Cornell University and New York State, through the College
of Agriculture, have a very deep involvement in this entire
process. This is a mandate that came down through a memorandum
from USDA, FDA, and EPA. The number was set after
being rounded off politically, and although people are striving
to reach the goal, in reality, we do view it as a nebulous goal.
Whether we reach it probably depends more on how we define IPM
than on what programs we put into action.
Therefore, we are currently exploring the questions, "What is the proper way to define IPM?", and "What programs do we consider to be IPM?" Clearly, it's not as fully implemented as we would like. The definition is clearly more than using biocontrol in programs or we wouldn't reach that goal or even be close to it. And, although a large percentage of acreage already is under IPM there remains a lot of work to be done.
The federal government is allocating large sums of money to the process. There is something like $15 million in the budget that is to be made available to not only federal researchers and implementation agencies, but also to universities. The idea is that we have a goal, it's an admirable goal in my eyes, but how we get to it is yet to be seen. And Congress must agree and provide the money.
Vandenberg:
Do you know if USDA or any other branch of government has it as part of their initiative to change regulations that might have to be changed to enable us to reach that goal, such as the number of insect parts allowed in tomato ketchup, or the number of blemishes in a batch of apples to reach highest marketable standards?
Carruthers:
There are a number of things going on. I don't know what they all
are. We deal with these kinds of issues in a national
committee on which I sit. Many different agencies are involved in
this process. For example, one thing that is being worked on
right now is taking crop insurance dollars and instead of using
them for such things as frost and environmental disasters, using
them to bridge the gap between chemical control and IPM control -
taking a bit of the risk out of jumping into the IPM arena.
That's one of the things being discussed in Washington.
Pimentel:
Having two Drosophila eggs in your ketchup has no public health impact, but putting more pesticide in your ketchup does. Since 1976 when my graduate students and I pointed this out in a paper and sent to the head of the FDA, there have been no further reductions in tolerance levels for insects in foods. So, one paper can have an impact.
Returning to the question of having 75% IPM, which also relates to what Becky was saying, we really could achieve a significant reduction in pesticide use in the US without any change in cosmetic standards and/or yields in crops. This is illustrated by the following USDA data: Since 1945, the quantity of insecticide use in the nation on our crops has increased tenfold and crop losses to insects have increased from 7%-13%. Why? Changes have occurred in agricultural technology. One example is corn, which is the crop that receives the most insecticide in the US Field corn in 1945 received 0 insecticides and it receives the most today; in 50 years insecticide use has risen over 1000-fold while crop losses to insects on corn have gone from 3.5% to 12%, according to the USDA.
We now have a great opportunity to examine these issues and problems and make adjustments. According to a paper we've written, we could reduce pesticide use by 50% in 5 years and save the growers and the public money!
Carlos Perez, Entomology, Cornell University:
How are you defining IPM, then?
Carruthers:
The Economic Research Service, the economic branch of the USDA,
is involved in assessing grower activities, practices, and
the way IPM is implemented. They are using many of the
grower-defined methodologies in defining IPM, and they're trying
to be very specific in their definition. It's going to include
the integrated use of insecticides with biological control
agents, cultural practices, and, probably, host plant resistance
for control of pests.
That's a broad definition; I don't know how you're going to get it any finer than that because many of these tactics are used in a whole number of crops to deal with pest problems.
Overall, the goal is truly to reduce the amount of pesticide, to shift our current methods of pest control from a chemical orientation to more of a biological orientation. That's the goal; that's what everyone has in mind. It's a nebulous goal and just by changing definitions we can change the numbers, but the purpose is what's important.
Perez:
What keeps the legal system in the US from completely banning pesticides?
Carruthers:
Pesticides are seen as a very important tool to manage pests. We
wouldn't want to ban pesticides in the very near future
because without them we'd be in a disastrous state. We need to go
through an evolution away from their use to finding other
methods of pest control. Biological control, host plant
resistance, and a range of things can be used, but certainly we
wouldn't want to rid our arsenal today of chemical pesticides or
we would all be hungry tomorrow.
Goldburg:
There is tremendous political pressure in Washington to continue the use of chemical pesticides. It is very hard to ban the use of a chemical pesticide even when there is a lot of demonstrable harm caused. I think of alachlor for which cancellation was considered about 5 years ago. It is one of the most widely used herbicides in the midwest. It is genotoxic. There is no question that it's carcinogenic, and it also leaches very readily into ground water and is found in surface water. Most environmentalists would consider alachlor along with atrazine the top 2 herbicides that cause environmental problems in this country. When EPA tried to cancel alachlor, there was a tremendous amount of pressure put on the agency from the chemical company that makes alachlor and also from farmers, and the agency did not cancel the registration.
So, I think that once you get into the political world, you find that making changes in pesticide law is a very, very difficult. I often see the environmental community playing a defensive role in pesticide regulation rather than achieving changes to law that we feel would be beneficial.
Wood:
Even a government regulation that calls for a reduction, rather than a ban that's mandated by law, doesn't seem to be politically acceptable in the US This type of re-education has been done successfully in Europe, achieving good results without any loss. It's pure economics as far as I can see.
Pimentel:
Some day I bet it will come, but it's a long way off compared to Sweden, Denmark, the Netherlands, and other nations.
K.V. Raman, International Service for the Acquisition of Agri-Biotech Applications at Cornell University:
I have two questions:
Goldburg:
I would like to respond to the first question regarding the notion that companies spend vast amounts of money on data. In some cases, such as Bt, this is true. Monsanto, in particular, has produced a lot of data on food safety for Bt crops because they have to demonstrate to EPA that they are putting new forms of Bt into the crops and that they will be safe for human consumption. However, in most cases the data that are being developed are very slim.
There is a paper in Trends in Ecology and Evolution, August of 1995, in which Joy Bergelson and one of her graduate students looked at all of the petitions received by APHIS up to that date for the commercialization of genetically engineered crops. They looked at which ones actually had data in them - not at the quality of the data itself (some of which was from experiments with no controls) - and for only 2 of the 14 areas that they looked at did the majority of petitions actually obtain data. It has been my experience that in most cases when it comes to convincing APHIS to allow the commercialization of crops, companies depend largely on armchair arguments. And I will say that, if anything, we need more data and not less.
When it comes to food safety, again, with EPA, when you're putting a pesticide, a compound, a toxic into a crop, yes, you're going to have to generate a lot of data, just like when you put pesticides and other chemicals on food. But, if your product is going through the FDA, really, the system is very lax. There are only a few exceptional circumstances in which you have to even come to FDA to get permission to commercialize a genetically engineered crop.
There was a lot of hoopla around Calgene's commercialization of tomato, but Calgene came to FDA voluntarily for permission. That is not happening with other crops. For the most part, FDA is merely asking that companies voluntarily consult the agency before going on the market. That is the extent of the supervision.
Tony Shelton, Entomology, Cornell University:
Recently there was a report from Europe about a transformed plant that had been crossed with a wild weed species. What do you think is the biological and ecological significance of that? What will be the political and policy-making fallout?
Goldburg: I think you're referring to a study published in Nature in March in which Danish scientists looked at herbicide-tolerant rapeseed that hybridized with a closely related wild plant. They found that in the first backcross they got transgenic offspring which were phenotypically weeds.
One of the arguments I alluded to earlier in the discussion about the safety of transgenic crops was that if you transfer via pollination a trait from a cultivated plant to a weed, the hybrid and subsequent backcrosses will be debilitated because they carry a lot of genetic material from the cultivated crops which is not favored in the wild. The Danish researchers showed that intergression occurred very quickly and you can get a plant within 2 generations that is genotypically wildtype and that carries the genetically engineered trait.
As for fallout in the policy-world arena, I have not seen much. A study with more potential for fallout appeared a week later in the New England Journal of Medicine. Scientists at the University of Nebraska found that soybeans engineered to produce a Brazil nut protein were allergenic for people who had Brazil nut allergies. This showed something that had previously been hypothetical: one can accidentally transfer, via genetic engineering, allergens from one food to another. That study has received more publicity, probably because of our society's focus on food safety.
Pimentel:
I don't think we can be too cautious about introductions for classical biological control, new associations, and genetic engineering. All we have to do is make one mistake and we lose our credibility with the public.
Carruthers:
The US Department of Agriculture is interested in solving pest problems as best we can. There are always controversial issues, but I hope that in evaluating the risks we'll try to determine both the risks and the benefits and make the best choices we can with the given set of information. I'd hate to see the system in gridlock because we're afraid to make a decision.
Goldburg:
Risk evaluation is a critical part of doing biological control and of biotechnology strategies towards pest control. I hope that members of the biocontrol community will adopt generally supportive attitudes towards environmental regulation: some of it is very, very important. There are some problems in the system, but we also need to do things as a society that keep our technologies safe.
Wood:
We all want biological control. As David said, biological control can be forever - positive or negative. We don't want chemicals, but remember: if you spray a chemical and find you don't like it, you don't have to use it again, and that's the end of it. That's not the case with biological control. As Ralph Hardy said this morning, we should take a multi-disciplinary view of what we're about to do in these introductions. Whether they're common or natural or alien, biological control agents can be forever. We don't want to shoot ourselves in the foot.
©All material is protected by Section 107 of the 1976 copyright law.
Copyright is held by Cornell University.
If you intend to use this material, please acknowledge the author and the source of the information.