Overview

Glyphosate herbicide formulations (e.g. Roundup(Herbicide) have been used as broad-spectrum herbicides for 25 years. Many studies have been conducted to assess the environmental safety of the active substance and its formulations. This backgrounder describes information that has been collected relative to beneficial arthropods. Beneficial arthropods refer to terrestrial arthropods, mostly insects, that are important in the biological control of agronomic pests, typically through predation or parasitism. Pollinators such as honey bees are not typically included in this narrow definition, but are included in this summary since they are non-target arthropod species important in agricultural ecosystems.

Glyphosate formulations have been tested in the laboratory and in the field to determine toxicity to terrestrial arthropods. Honey bees exposed to glyphosate formulations directly through ingestion or topically (simulating direct overspray) were not affected. Screening tests have been performed on a number of other beneficial arthropods including beetles, mites, spiders, and wasps. These tests are designed to maximize exposure (maximum use rate, no interception, etc.). Large beneficial arthropods such as ground predators (spiders and beetles) are not affected by glyphosate formulations. Although most foliar dwelling species were not affected in the screening tests, several species were (e.g. parasitic wasp and predatory mite). The effects observed in these screening tests under extreme conditions were reduced or eliminated when more realistic exposure conditions were used. Modification of vegetation in the treated areas can affect beneficial arthropod species as indicated in a number of field studies described below. Use of glyphosate formulations will have no effects on beneficial arthropods occupying areas adjacent to treated areas.

Detailed Summary

Laboratory studies

Standard laboratory tests with honey bees have shown that glyphosate based formulations are practically harmless at levels higher than would be encountered in the field. The oral and contact LD50 values for these compounds in honey bees range from > 25 (g/bee to >100 (g/bee (Huntington 1985; Wildlife International Ltd. 1992), indicating minimal risk to honey bees.

In addition to toxicity tests with honey bees, laboratory toxicity tests have been developed that can be used to evaluate the toxicity of pesticides to beneficial arthropods and other non-target arthropods (SETAC 1994).

Initial laboratory tests are highly artificial tests where the pesticide is applied at the maximum use rate onto artificial substrates such as glass or synthetic soil and the organisms are forced to remain in contact with the pesticide film for several days. Depending on results of these screening tests, subsequent tests such as extended laboratory tests, semi-field and field tests may be needed. These "higher level" tests incorporate more realistic environmental exposure conditions.

The effects of Roundup herbicide have been investigated in the screening level assay with 18 different beneficial predators and parasites (Hassan et al. 1988). Roundup was found to be harmless to thirteen species, slightly harmful to four species and moderately harmful to one species of carabid beetles (see Appendix A). The authors did not believe that sufficient toxicity potential existed to warrant semi-field and field tests that were performed on some of the other compounds tested in the same program.

Moreover, a subsequent semi-field test with a similar glyphosate formulation was conducted on carabid beetles (Mead-Briggs 1991). Results of that test indicated that even if beetles were directly oversprayed at the maximum use rate, no mortality was observed. The reason for the discrepancy between the laboratory and semi-field study results is not known but may be related to the highly artificial nature of the laboratory glass plate assays (e.g. potential stickiness of the formulation on the glass substrate).

Beneficial arthropod tests are currently being conducted on several glyphosate formulations used in the European Union. Ground beetle, ground spider, parasitic wasp, and predatory mite are being tested in screening assays. The formulations at the maximum use rate caused no effects on beetles or spiders. Some effects were observed on parasitic wasps and predatory mites at the maximum use rate in the screening tests. As described earlier for Roundup, the effects were reduced or eliminated when real exposure conditions and substrates were included.

Field studies

A field study was conducted to investigate the effects of aerial application on honey bee hives (Burgett and Fisher 1990). Bee hives and blooming vegetation in the immediate vicinity (1.5 acres) were treated at a rate of about 5% Roundup in 60 gal of water (6 lbs a.e./acre or 6.8 kg a.e./ha). No acute or chronic effects were observed for adult honey bees or for brood production. These findings were further supported by conducting direct feeding trials in the field. No effects were observed as a result of the direct addition of Roundup to the hives.

In addition, field studies have investigated the effects of glyphosate formulations on beneficial arthropods other than honey bees. No substance- or dose-related effects on mites or springtails were observed in a sandy soil in an Argentine semi-arid region up to 96 days after application of Roundup herbicide at rates up to 2.8 kg glyphosate/ha (Gomez & Sagardoy 1985). No effects were noted for the number of nematodes, mites, or springtails in the top 3 cm of soil 180 days after treatment with Roundup at 2 kg a.e./ha (Preston & Trofymow 1989).

Reduced populations of herbivorous insects and ground invertebrates were observed in a 4-5 year old clear-cut planted with spruce (Picea sp.) up to 3 years after treatment with Roundup herbicide (Santillo et al. 1989a; 1989b). During this 3 year study, the vegetation failed to recover completely, suggesting that the majority of effects on invertebrates were mainly due to habitat change. On the other hand, no changes in populations of predatory insects were observed in a clearcut located in Maine (United States) that was sprayed with 1.7 kg glyphosate/ha (Whitehouse and Brown 1993). The authors of the report concluded that insect communities would not be effected by the use of glyphosate applied to the base of trees in farm forestry.

The effects of glyphosate on beneficial insects has also been investigated by studying indigenous carabid beetle populations after field application (Brust 1990). Both relative densities and movement of carabid beetles in and out of treated areas was monitored. No direct toxicity was observed in the field nor was there any repellency noted. Shifts in population densities were observed in the weeks following treatment, but these changes were more closely associated with changes in the vegetation of the plots rather than direct toxicity following the glyphosate treatment. The absence of direct toxic effects was confirmed in the laboratory. Other researchers concluded that the primary factors influencing the changes in carabid beetle and spider populations were deprivation of a particular species of suitable food and change in habitat (Asteraki et al. 1992). Several studies have shown that the application of glyphosate can increase populations of beneficial insects. In laboratory experiments to simulate treatment of cotton fields, numbers of the western bigeyed bug, Geocoris pallens, increased (Yokoyama and Pritchard 1984). However, these authors did not measure behavioral effects and cautioned that responses might vary under field conditions. No effects on the number of common butterfly species were observed when glyphosate was used to control trees, shrubs and blackberry in wire zones; but numbers of individuals did increase (Bramble et al. 1997).

Conclusion

Honey bees are not affected by glyphosate or Roundup formulations based on data from laboratory and field studies. Screening tests have been performed on a number of other beneficial arthropods including beetles, mites, spiders, and wasps. These tests are designed to maximize exposure (maximum use rate, no interception, etc.). Large beneficial arthropods such as ground predators (spiders and beetles) are not at risk from glyphosate formulations. Several foliar dwelling species (e.g. parasitic wasp, predatory mite) for several glyphosate formulations are potentially affected based on laboratory screening tests. However, under realistic exposure regimes, testing showed that it is unlikely that effects will be observed. Within treated areas, alteration of the vegetation following glyphosate treatment can result in substantial change in habitats over the short term, and, consequently, in some cases,insect populations.

Michael J. McKee, Ph.D.
Ecotoxicology Specialist
Monsanto Agricultural Group

Literature cited

Asteraki, E.J., Hanks, C.B., and Clements, R.O. 1992. The impact of the chemical removal of the hedge-base flora on the community structure of carabid beetles (Col., Carabidae) and spiders (Araneae) of the field and hedge bottom. J. Appl. Ent. 113: 398-406.

Bramble, W.C., Yahner, R.H., and Byrnes, W.R. 1997. Effect of herbicides on butterfly populations of an electric transmission right-of-way. J. Arboriculture 23(5): 196-206.

Brust, G.E. 1990. Direct and indirect effects of four herbicides on the activity of Carabid beetles (Coleoptera: Carabidae). Pestic. Sci. 30: 309-320.

Burgett, M. and Fisher, G. 1990. A review of the Belizean honey bee industry: Final report prepared at the request of The Belize Honey Producers Federation. Department of Entomology, Oregon State University, Corvallis, Oregon.

Gomez, M.A. and M.A. Sagardoy. 1985. Influence of glyphosate herbicide on the microflora and mesofauna of a sandy soil in a semi-arid region. Rev. Latin Amer. Microbiol. 27:351-357.

Hassan, S.A., F. Bigler, H. Bogenschutz, E. Boller, J. Brun, P. Chiverton, P. Edwards, F. Mansour, E. Naton, P.A. Oomen, W.P.J. Overmeer, L. Polgar, W. Rieckman, L. Samsoe-Petersen, A. Staubli, G. Sterk, K. Tavares, J.J. Tuset, G. Viggiani, and A.G. Vivas. 1988. Results of the fourth joint pesticide testing programme carried out by the IOBC/WPRS-Working Group

"Pesticides and Beneficial Organisms". J. Appl. Ent. 105:321-329. Huntington Research Center. 1985.

The acute contact and oral toxicities of CP 67573 and MON 2139 to worker honey bees. Monsanto study no. HU-85-094.

Mead-Briggs, M. 1991. An evaluation of the toxicity of Roundup to the carabid deetle Bembidion lampros under semi-field conditions. Agrochemical Evaluation Unit, Dept. of Biology, The University, Southhampton, UK.

Monsanto unpublished report XX-90-603.

Preston, C.M. and J.A. Trofymow. 1989. Effects of glyphosate (Roundup) on biological activity of forest soils. In: Proceedings of Carnation Creek Workshop, ed. P. Reynolds. Namaimo 7-10 December 1987. Forest Canada/British Columbia ministry of forests, 122-140.

Santillo, D.J., P.W. Brown, and D.M. Leslie. 1989a. Response of songbirds to glyphosate-induced habitat changes on clearcuts. J. Wildlife Mngmt. 53: 64-71.

Santillo, D.J., D.M. Leslie, and P.W. Brown. 1989b. Response of small mammals to glyphosate application on clearcuts. J. Wildlife Mngmt. 53: 164-172.

SETAC. 1994. Guidance document on regulatory testing procedures for pesticides with non-target arthropods. Eds. K.L. Barret, N. Grandy, E.G. Harrison, S. Hassan, and P. Oomen. Society of Environmental Toxicology and Chemistry - Europe.

Whitehouse, D.M. and Brown, V.K. 1993. Herbicides in farm forestry: effects on non-target insects. Brighton Crop Protection Conference: Weeds. November 22-25 1993. Wildlife International Ltd. 1992. MON 52276: An acute contact toxicity study with the honey bee. Monsanto study no. WL-92-233.

Yokoyama, V.Y. and Pritchard, J. 1984. Effect of pesticides on mortality, fecundity, and egg viability of Geocoris pallens (Hemiptera: Lygaeidae). J Economic Entomology 77(4): 876-879.

APPENDIX A

Terrestrial Invertebrate Ecotoxicology
Arthropoda: Insecta: Hymenoptera (wasps) Toxicity rating
Trichogramma cacoeciae Slightly harmful
Encarsia formosa Harmless
Aphidius matricariae Harmless
Leptomastix dactylopii Harmless
Cales noacki Harmless
Phygadenon trichops Harmless
Coccygomimus turionellae Harmless
Arthropoda: Insecta: Coleoptera (beetles)
Harmonia axyridis (Coccinellidae) Harmless
Semiadalia 11-notata (Coccinellidae) Slightly harmful
Bembidion lampros (Carabidae) Moderately harmful
Pterostichus cupreus (Carabidae) Harmless
Arthropoda: Insecta: Neuroptera
Chrysoperla carnea (green lacewing) Slightly harmful
Arthropoda: Insecta: Heteroptera
Anthocoris nemoralis (minute pirate bugs) Harmless
Arthopoda: Arachnida (spiders & mites)
Phytoseiulus persimilus (mite) Harmless
Amblyseius potentillae (mite) Harmless
Typhlodromus pyri (mite) Slightly harmful
Typhlodromus p. R. (mite) Harmless
Chiracanthium mildei (spider) Harmless

** NOTICE: In accordance with Title 17 U.S.C. Section 107, this material is distributed for research and educational purposes only. **