GM Free Cymru


A recent paper in Ecological Applications (Relyea 2005a) has demonstrated highly lethal effects of the herbicide Roundup® on amphibians.

A brief description of the Relyea (2005a) study

To determine the effect of Roundup on tadpoles in ponds, Relyea (2005a) added Roundup to pond mesocosms (1000-liter outdoor tanks that contained algae, zooplankton, snails, tadpoles, and several species of insect predators). After two weeks, we went back and determined how many tadpoles survived. The result was widespread death for many of the amphibian species when exposed to Roundup compared to amphibians in the control tanks. Furthermore the death rate was much higher than expected based on previous studies of Roundup.

Roundup (sold under a variety of commercial names worldwide, including Vision®) is the number one herbicide in the world and the manufacturer of Roundup (Monsanto) has expressed a number of concerns about this study on the Monsanto web site.

Below, Dr. Relyea responds to Monsanto's concerns.

Concern #1: Roundup is only intended for terrestrial use, not aquatic use

While it may be intended for terrestrial use, there is overwhelming evidence that Roundup gets into aquatic habitats, typically through inadvertent (or unavoidable) aerial overspray (Newton et al. 1984, Goldsborough and Brown 1989, Feng et al. 1990, Thompson et al. 2004). In addition, in some countries outside of North America, Roundup is sprayed directly on water to control emergent aquatic plants (Giesy et al. 2000).

It is relatively common knowledge that Roundup should not be applied to large ponds and lakes, but it is less commonly appreciated that most amphibians are not produced in large ponds and lakes due to predation by fish. Instead, small temporary wetlands that may appear to be unimportant and only have 6" (15 cm) of water can, in fact, produce thousands of tadpoles including many species that breed only in temporary wetlands. These small, temporary pools are either not avoided or not avoidable by aerial pesticide applications.

Roundup is not only lethal to tadpoles. A new study has discovered that Roundup can be highly lethal to terrestrial frogs and toads as well as well (Relyea 2005c).

Concern #2: The application rate of Roundup was 3 to 7 times too high

The application rate of "6 ounces per 300 square feet" came directly from the label of Monsanto's "Roundup Weed and Grass Killer." Monsanto wrote the label.

The simple issue is that Roundup has a wide range of application rates depending on which weeds one is trying to control. Thus, the recommended application rates for some uses are higher than for other uses. Moreover, some types of weeds require up to four times their recommended application rate to be effective.

Concern #3: The concentration of Roundup used in the experiment was high and unrealistic

Regardless of application rate issues, the real question is whether the concentrations in the water were reasonable.

Before I address this concern, I want to point out the different units used in reporting Roundup concentrations. Roundup can either be reported in mg a.i./L (milligrams of active ingredient per liter) or in mg a.e./L (milligrams of acid equivalents per liter). This subtle point is easy to overlook when reading the Roundup literature. To convert between units, 1 mg a.i/L = 0.75 mg a.e./L.

The concentration used by Relyea (2005a) was 3.8 mg a.i./L = 2.9 mg a.e./L.

A variety of researchers have estimated the "worst-case scenario" for Roundup, which is the maximum concentration of Roundup predicted to be in a wetland that is inadvertently over-sprayed). These estimates include the following:

1.4 mg a.e./L (Canadian government estimates)

2.7 mg a.e./L (Solomon and Thompson 2003)

2.8 mg a.e./L (Giesy et al. 2000, in collaboration with Monsanto)

2.9 mg a.e./L (Perkins et al. 2000)

7.6 mg a.e./L (Mann and Bidwell 1999)

Thus, the Relyea (2005a) experiment was testing the worst-case scenario for tadpoles exposed to Roundup based on Giesy et al. (2000).

Although we have many estimates of the worst-case scenarios for Roundup, we few studies on the actual concentrations of Roundup in wetlands. Observed values include the following:

0.3 to 0.7 mg a.e./L (Wood 2001)

1.1 mg a.e./L (Beck 1985)

0.3 to 1.2 mg a.e./L (Newton et al. 1994)

1.5 mg a.e./L (Couture et al. 1995)

1.7 mg a.e./L (Horner, unpublished data for Monsanto, published in Giesy et al. 2000)

0 to 1.9 mg a.e./L (Thompson et al. (2004)

2.8 mg a.e./L (Legris and Couture 1989)

5.2 mg a.e./L (Edwards et al. 1980)

Importantly, a new study using only one-third as much Roundup as the worst-case scenario (1.0 mg a.e./L) still caused up to 71% amphibian mortality (Relyea et al. 2005). This means that even when Roundup is well within expected concentrations, a substantial fraction of a tadpole population can be killed.

Concern #4: Past risk assessments have shown that Roundup poses minimal risk to amphibians

The first risk assessment to include any data on tadpoles was conducted by Giesy et al. (2000), in cooperation with Monsanto, and the assessment was based on the available data at that time. For amphibians, data only existed for four species of Australian tadpoles (there are > 5,000 species of amphibians worldwide). For the four Australian species, Mann and Bidwell (1999) estimated the LC50 (the amount of pesticide needed to kill 50% of the animals) to be 2.9 to 11.6 mg a.e./L. A subsequent risk assessment by Solomon and Thompson (2003) also included very few data on amphibians and addressed the risk of Roundup to all aquatic organisms pooled together, without any direct assessment of amphibians in particular.

During the past three years, there has been a surge of interest in conducting LC50 experiments with Roundup on North American amphibians. These studies have demonstrated that North American amphibians can be much more sensitive than the four Australian species. These studies have found the following LC50 values:

2.7 to 11.5 mg a.e./L (Wojtaszek et al. 2004)

0.9 to 3.5 mg a.e./L (Edginton et al. 2004)

2.0 to >8 mg a.e./L (Howe et al. 2004)

0.4 to 1.9 mg a.e./L (Relyea 2005b)

In addition, there are new experiments conducted with the African clawed frog also showing very low LC50 values:

1.1 mg a.e./L (Sutherland et al. 2006)

According to U.S. Fish and Wildlife classifications, this means that Roundup formulations containing the POEA surfactant can no longer be considered slightly to moderately toxic, but rather moderately to highly toxic to North American amphibians.

Concern #5: The results are not consistent with in-situ studies of amphibians in Canada (Thompson et al. 2004)

The studies of Thompson et al. (2004) and Relyea (2005a) differed in several important ways. Thompson et al. (2004) reported mortality rates of green frog and leopard frog tadpoles placed in small cages in ponds after 48 hours of exposure to Roundup. Relyea (2005a) used an almost entirely different set of species (leopard frogs, wood frogs, gray tree frogs, American toads, and spring peepers), the animals were raised in large pond mesocosms, and the exposure time was much longer (2 weeks). In addition, Thompson et al.'s tadpoles experienced an average of 36% mortality in ponds with the highest Roundup concentrations. This high mortality in only 48 hours was not attributed to Roundup (P = 0.12), but it could not be explained by any other natural or anthropogenic factor. When there is high and unexplained mortality, it is difficult to draw strong conclusions from such studies.

Concern #6: The results are not realistic because the pond mesocosms did not contain soil

It is true that the Relyea (2005a) experiment used tanks that did not contain soil. This is an important issue because soil can absorb pesticides and thereby remove them from the water column. However, a subsequent study (Relyea 2005c) has examined the effect of adding soil and found that it made no difference to tadpole survival. In this study, Roundup caused 98% tadpole mortality, regardless of soil additions.

Concern #7: The study attributes the mortality to the surfactant in Roundup but did not test the surfactant alone

Glyphosate-based herbicides are not effective unless a surfactant is added to allow the active ingredient (glyphosate) to penetrate the waxy cuticle of plant leaves. The most widely used formulations of Roundup contain the surfactant POEA (polyethoxylated tallowamine, a chemical derived from animal fat).

Previous work in the Australian species showed that, although the Australian species were less sensitive to Roundup, the death that occurred was completely due to the surfactant (POEA) and not due to the active ingredient (glyphosate). Relyea (2005a) used the commercial form of Roundup Weed and Grass Killer® (25% glyphosate) that also contained the POEA surfactant. Importantly, while most agricultural formulations of glyphosate are 41% glyphosate, the glyphosate:surfactant ratio in the version of Roundup used in Relyea (2005) is identical to that used in agricultural formulations.

The results of Relyea (2005a) only apply to formulations that contain this common surfactant and not necessarily to other forms of glyphosate (e.g., Rodeo). Glyphosate formulations such as Rodeo® do not contain a surfactant (they contain only glyphosate and water), but the consumer must purchase a separate surfactant and combine it with Rodeo® to make it effective in controlling aquatic plants.

Who funded Dr. Relyea's research?

All work was funded by the United States government's National Science Foundation. This research has no anti-pesticide, anti-agriculture, or anti-forestry agenda. We simply asked the question, "What happens to tadpoles if Roundup is present in aquatic habitats?"

Abstracts and electronic reprints of all Relyea articles can be found be clicking on “Publications” at the top of the page.


Beck, A. E. 1985. Glyphosate residues in surface water following initial Manfor Ltd. Field trials, 1985., Water Standards and Studies Report #87-4. Manitoba Environment and Workplace Safety and Health.

Couture, G., J. Legris, and R. Langevin. 1995. Évaluation des impacts du glyphosate utilisé dans le milieu forestier. Ministere des Ressources Naturelles, Direction de l’environment forestier, Service du suivi environnemental. 199 pp.

Edginton AN, Sheridan PM, Stephenson GR, Thompson DG, Boermans HJ (2004) Comparative effects of pH and Vision® herbicide on two life stages of four anuran amphibian species. Environ Toxicol Chem 23:815-822.

Edwards, W. M., G. B. Triplett Jr., and R. M. Kramer. 1980. A watershed study of glyphosate transport in runoff. Journal of Environmental Quality 9:661-665.

Feng JC, Thompson DG, Reynolds PE (1990) Fate of glyphosate in a Canadian forest watershed. 1. Aquatic residues and off-target deposit assessment. J Agric Food Chem 38: 1110-1118.

Giesy JP, Dobson S, Solomon KR (2000) Ecotoxicological risk assessment for Roundup herbicide. Rev Contam Toxicol 167: 35-120.

Goldsborough LG, Brown DJ (1989) Rapid dissipation of glyphosate and aminomethylphosphonic acid in water and sediments of boreal forest ponds. Environ Toxicol Chem 12: 1139-1147.

Howe, C. M., M. Berrill, B. D. Pauli, C. C. Helbring, K. Werry, and N. Veldhoen (2004) Toxicity of glyphosate-cased pesticides to four North American frog species. Environmental Toxicology and Chemistry 23:1928-1938.

Legris, J., and G. Couture. 1989. Résidus de glyphosate dans l’eau et les sediments suite a des pulverisations terrestres en milieu forestier en 1986. Gouvernment du Quebec. Ministere de l’Energie et des Ressources. Direction de la Conservation. Publication #3322. 26 pp.

Leveille, P., J. Legris, and G. Couture. 1993. Results of spot checks in lotic environments after spraying of glyphosate in forests from 1989 to 1991. Quebec, Canada: Ministere des Forets du Quebec. 21 p.

Mann, R. M., and J. R. Bidwell. 1999. The toxicity of glyphosate and several glyphosate formulations to four species of southwestern Australian frogs. Archives of Environmental Contamination and Toxicology 26:193-199.

Newton M, Howard KM, Kelpsas BR, Danhaus R, Lottman CM, Dubelman S (1984) Fate of glyphosate in an Oregon forest ecosystem. J Agric Food Chem 32: 1144-1151.

Newton, M., L. H. Horner, J. E. Cowell, D. E. White, E. C. Cole. 1994. Dissipation of glyphosate and aminomethylphosphonic acid in North American forests. Journal of Agriculture and Food Chemistry 42:1795-1802.

Perkins PJ, Boermans HJ, Stephenson GR (2000) Toxicity of glyphosate and triclopyr using the frog embryo teratogenesis assay-Xenopus. Environ Toxicol Chem 19: 940-945.

Relyea, RA (2005a) The impact of insecticides and herbicides on the biodiversity and productivity of aquatic communities. Ecological Applications 15:618-627.

Relyea, RA (2005b) The lethal impacts of Roundup and predatory stress on six species of North American tadpoles. Archives of Environmental Contamination and Toxicology 48:351-357.

Relyea, RA (2005c) The lethal impact of Roundup® on aquatic and terrestrial amphibians. Ecological Applications 15:1118-1124.

Relyea, RA, NM Schoeppner, and JT Hoverman (2005) Pesticides and amphibians: The importance of community context. Ecological Applications 15:1125-1134.

Solomon, KR, and DG Thompson (2003) Ecological risk assessment for aquatic organisms from over-water uses of glyphosate. Journal of Toxicology and Environmental Health 6:289-324.

Thompson DG, Wojtaszek BF, Staznik B, Chartrand DT, Stephenson GR (2004) Chemical and biomonitoring to assess potential acute effects of Vision® herbicide on native amphibian larvae in forest wetlands. Environ Toxicol Chem 23:843-849

Wojtaszek, B. F., B. Staznik, D. T. Chartrand , G. R. Stephenson, D. G. Thompson. 2004. Effects of Vision® herbicide on mortality, avoidance response, and growth of amphibian larvae in two forest wetlands. Environmental Contamination and Toxicology 23:832-842.

Wood, T. M. 2001. Herbicide use in the management of roadside vegetation, Western Oregon, 1999-2000.: Effects on the water quality of nearby streams. U.S. Department of the interior. U.S. Geological Survey. Water Resources Investigations Report 01-4065. pp. 27