IMPACT OF HERBICIDE ON MOSSY SORREL, RUMEX CONFERTUS WILD., AND PHYTOPHAGOUS GASTROIDEA VIRIDULA DEG. AND GASTROIDEA POLYGONI L. (COLEOPTERA: CHRYSOMELIDAE)

Dariusz Piesik, Robert Lamparski

ABSTRACT

Field and laboratory research was conducted to determine the effect of five herbicides on Rumex confertus Willd. which is the host of Gastroidea viridula Deg. and Gastroidea polygoni L. (Coleoptera: Chrysomelidae). Herbicide treatment effected plant growth, but new leaf rosettes were produced by the end of the summer. Survival, seasonal abundance and development of investigated insects were significantly reduced. Beetles and larvae avoided feeding on herbicide treated plants.

Key words: herbicides, sub-lethal effects, growth, Rumex confertus Willd., biological agent, Gastroidea viridula Deg., Gastroidea polygoni L..

INTRODUCTION

R. confertus, commonly known as mossy sorrel, establishes new stands in Poland every year. As sorrel contains high amounts of oxalic acid, the lethal poisoning of animals can occur when large quantities of the plant are consumed. This weed is a 1.5 to 1.8m high perennial plant yielding from 100 to 40.000 seeds [2]. It usually flowers in July, August, and September, but some individuals can produce seeds twice a year, i.e. in early and late summer [2]. R. confertus occurs commonly in Poland and the world [8, 12, 18]; however, a clear expansion westward can be observed. Mossy sorrel is generally a widely adapted plant, and it is considered to be one of the most dangerous, uncultivated plants in the world due to its very strong expansion which results from an abundant seed production [2].

The chemical control of this weed is difficult because the rich root system permits plants to recover from treatment. After applying chemical plant protection products, the aboveground parts of the plant dry out, but shortly afterward, new leaves emerge [14, 15]. In addition, unjustified chemical treatments can induce the breakdown of plants’ resistance [1, 5, 9].

The pesticides can be directly toxic to herbivorous insects which help to control sorrel populations. This toxic effect is mainly expected for insecticides but can also be observed for herbicides and fungicides [26]. Also, pesticides may deplete the food resources or alter the food suitability for insects [6]. An example of this is herbicides reducing the weed biomass for herbivorous insects and/or changing the nutritional status of the plant [21, 23]. Predatory and parasitic species may be limited by less prey, or they may consume such amounts of contaminated food items that toxic effects occur, even when direct exposure of the predator is insignificant [25]. Finally, pesticides may be repellent to the insects, thus increasing emigration and/or decreasing immigration [3].

An alternative to the pesticide treatments includes biological methods and the use of herbivorous insects, our allies [7, 27]. Gastroidea spp., Apion spp., Pegomya spp., Hypera spp., Mamestra spp. groups are of special interest among the numerous species inhabiting Rumex spp. [24]. G. viridula and G. polygoni ought to be good biological control agents of their host plants, because they have a high reproductive potential [19], a short life cycle, no alternative hosts, and the number of their predator are low.

It is generally known that weeds have a negative impact on crop yield; therefore populations of herbivore species utilizing weeds as food will not benefit from reduced herbicide dosages, as the amount of food and the recruitment to the future plant populations is unchanged. The plants became toxic to Gastroidea spp. when treated with herbicides [6].

The aim of the study was to determine the herbicide influence on feeding, development and survival of G. viridula in the laboratory and field tests.

METHODS AND AREA OF THE STUDIES

The experiments were carried out in 2001 and 2002. The field trials were located in the natural habitat of R. confertus in Bydgoszcz vicinity on the marshy meadow near Vistula river, Northern Poland (53⁰13^'N, 17⁰51^'E). The laboratory study was conducted at Department of Entomology, University of Technology and Agriculture in Bydgoszcz.

Five herbicides were used in the experiment: Aminopielik Super 464 SL (2,4-D = 344 and dikamba = 120), Glean 75 WG (chlorosulfuron = 75%), Harmony 75 WG (tifensulfuron methyl = 75%), Refine 75 WG (tifensulfuron methyl = 75), and Superselectyl 435 SL (mekoprop = 125 * MCPA = 125 * dichloroprop = 185). They are herbicides generally used for broadleaf weed control.

The chrysomelid beetles, G. viridula and G. polygoni were chosen as model species. They seem to be relevant for many reasons. First of all, these insects are characterized by low mobility. Also, oligophagous preferences for Polygonaceae spp. allow to choose these insects as possible biological control agents for their host plant.

The sprayer was a 12 l, hand-held Kwazar (Kwazar Corporation S.C.) with a lance length from 0.6 to 1.2m, and a double filtration system. The analysis of variance means were separated using LSD Dunnet’s test.

The field experiments.

1. Five chosen plants (in 5 replicates) were sprayed only once with a recommended solution of the test chemical in the beginning of May. Five control plants were sprayed only with water. The field studies were carried out over whole growing season; from spring to autumn. The plants length and larval abundance were recorded weekly. Captured larvae were counted precisely and then released. For the larvae gathering, 25 full sweep nets, what resulted in 5 tested plants, were done.

Laboratory experiments.

Insects from the field were taken for experiments. The material collected this way was analysed in the laboratory and used for rearing. The following assays were carried out in the laboratory.

1. In the survival test eggs laid by the Gastroidea spp. females on the leaves were used. The experiment was conducted with four replicates on Petri dishes supplied with filter paper. For comparison, a control was also set up. Leaves with eggs were soaked once in a solution of chemicals recommended by producers as suitable to control Rumex spp. In the control, leaves were soaked only in water. Filter paper and leaves used by larvae were changed daily. The observations were continued to the moment of adult emergence.

2. Forty-eight hour tests were done for L1, L2, and L3 larvae, as well as for adult insects. Only young larvae of L2, and L3, immediately after shedding, were investigated. The experiment was conducted with four replicates on Petri dishes supplied with filter paper. For comparison, a control was set up. With 48h tests, measured amounts of leaves, soaked in a solution of the suitable chemical, were put into the Petri dishes. The leaf weight was checked after 24 and 48 hours.

RESULTS

The control plants grown up in experiment years from April to late July (Fig. 1). Then, they started to die which seems to be normal in their natural habitat. Aminopielik seemed to be the strongest herbicide. All treated plants died in the beginning of June.

Moderate herbicide effectiveness characterized Harmony, Refine, and Glean. Plants were damaged by those chemicals, but were still green. The smallest pesticide activity was recorded in case of Superselectyl. From late July, when all control and treated plants were growing up again, no large differences were recorded, which suggests that herbicides only partly restricted development of the tested plants (Fig.1).

The number of Gastroidea spp. larvae is presented in Fig. 2. The control larvae were not stressed by any herbicide; were developing, and reached 398, 234, and 108 individuals in three consecutive stages. Aminopielik apparently changed the leaf taste, and/or odour, which restricted feeding of Gastroidea spp. larvae, causing the highest avoidance of the plant. From early June the Aminopielik plants were not present, as well as Gastroidea spp. larvae. Superselectyl seemed to be not too dangerous to insects, but also slightly lessened the abundance of larvae in consecutive stages. The number of larvae caught in the third generation was similar.

Correlation was detected and statistically significant (r=0.72) when Aminopielik was applied on the plants (Fig. 3). The weed biomass and larval population declined.

The first 24h test of adult insects was not statistically significant (Table 1). For the second one, insects in all treatments ate significantly less food compared to the control. The same trend was recorded for the 48h total test. This suggest that beetles at least were repelled from the odour of herbicides. In case of Aminopielik, insects ate 54% less leaves in comparision to insects in the control. It is important to recognize that this calculation is based on experiments undertaken in a controlled environment.

Survival of the larvae was significantly reduced when the larvae were put on the treated leaves (Table 2). The herbicide treatment could have changed the nutritional status of the plant, thus causing the increased mortality. The survival of Gastroidea spp. hatched from eggs, was significantly affected by herbicide treatment of the host plant, especially by Aminopielik, Glean, and Harmony. On unsprayed plants, 79 % of the individuals survived to the adult stage. The effects of herbicides on plants in the laboratory may differ greatly from the field situation where plants tend to be much more resistant. It is important to recognize that this calculation is based on experiments undertaken in a controlled environment. Significant mortality of Gastroidea spp. from the recommended dose of herbicides were observed. Furthermore, sublethally exposed individuals may exhibit a decreased fertility, which will affect the population size in the next generation.

The first 24h and total 48h tests showed significant results in all of the treatments (Table 3). Especially in case of Aminopielik, and Glean L1, larvae ate less provided food compared to insects in the control. Larvae fed with Aminopielik, Glean, and Harmony-treated food had less survival. For Harmony, 87% of the consecutive developmental stages died.

Significant results were recorded for all treatments in the first 24h test, as well as for total 48h experiment (Table 4). For Aminopielik and Refine, less adult emergence was observed.

Very interesting results were achieved from the L3 larvae experiment (Table 5). Even though L3 larvae seem to be quite resistant to any unpalatable foods, significant differences were observed. For all treatments of the first and second test, as well as for the total 48h test, much less food was eaten than in the control. For Aminopielik, the food consumed in the treatment was only 23% of that consumed in the control. Also, difficult emergence of adults was recorded. Additionally, for the Superselectyl treatment, significantly less adult abundance was observed.

DISCUSSION

The most important phytophagous insect occurring on R. confertus was G. viridula (Coleoptera, Chrysomelidae). Feeding of the beetle imagines proceeded in the form of biting out holes of different sizes in the plant leaves. At the beginning, the larvae skeletonized the leaves and then also made holes [16]. Under natural site conditions G. viridula has damaged the vegetative weight of mossy sorrel to a large extent. Whittaker et al. [29] described this species as a very important biological factor, as well as part of an integrated programme for regulating the development of undesirable plants. G. viridula is considered as temperature-dependant insect [4]. G. polygoni plays a smaller role in destroying leaves, mainly because of its smaller population on sorrels. However, feeding of larvae and adult individuals was observed on these weeds. Two generations of this insect species were noted in a year, which is in good agreement with results of the other r esearchers [10, 17].

Herbicide-treated plants were significantly less attractive food in comparison to the control. This suggest that beetles at least were repelled from the odour of herbicides. Significant mortality of Gastroidea spp. from the recommended dose of herbicides were recorded. Survival of the larvae was significantly reduced when the larvae were put on the treated leaves. Furthermore, sublethally exposed individuals may exhibit a decreased fertility, which will affect the population size in the next generation. The herbicide treatment may change the physiological status of the plant, for example altering water and nutrients contents [11, 28]. Speight and Whittaker [23] studied the controlling ability of G. viridula on R. obtusifolius and found that this beetle was unable to improve the control of the weed in combination with the herbicide Asulam under natural conditions. Sotherton and Moreby [22] found that Gastroidea spp. often were absent from host patches in fields sp rayed with herbicide the previous year, indicating that the dispersal to host patches or the ability to locate hosts is limited. However, it was indicated that in concert with relatively low dosages of herbicide, it would have the ability to control R. obtusifolius effectively. Speight and Whittaker [23] studied the combination of the herbicide Asulam and the herbivore G. viridula on R. obtusifolius. They found that the herbivore was largely unaffected by the herbicide, but also that this beetle was not able to reduce the performance of the weed.

So far, the integration of chemical and biological control is scarcely exploited in Europe, even though it has the potential to reduce pesticide use significantly. Furthermore, insects are subjected to direct toxic effects from herbicides and other pesticides. Sotherton [20] found a reduced survival of first instar larvae of G. polygoni after 2,4-D application to eggs or host plant leaves. Speight and Whittaker [23] reported no effect of Asulam on survival or rate of development of G. viridula, but following ingestion of contaminated leaves, the fecundity of female beetles was reduced. Patnaik et al. [13] found significant mortality of water hyacinth weevil exposed to 2,4-D, paraquat, and diquat.

CONCLUSIONS

1. Mossy sorrel (R. confertus), a weed of the polygonum family (Polygonaceae), was injured by G. viridula and G. polygoni.

2. In the field, plants development was restricted only partly by herbicide treatment. Plants developed new leaf rossetes by the end of the summer.

3. Investigated herbicides significantly impacted the insects development, reducing their survival and increasing mortality of the Gastroidea spp. Insects avoided feeding on herbicide treated plants.

4. Insufficient results of herbicides applying may suggest biological agents as the most appropriate to control R. confertus population.

1. Boczek J. 1996. Stan i perspektywy walki biologicznej z chwastami [State and perspectives of biological control of weeds]. Post. Nauk Roln., 4, 77-89 [in Polish].

2. Cavers P.B., Harper J.L. 1964. Biological flora of the British Isles, Rumex obtusifolius L. and Rumex crispus L. Ecol., 52, 737-766.

3. Hengchen L., Hoy C.W., Head G. 1993. Olfactory response of larvae diamondback moth (Lepidoptera: Plutellidae) to permethrin formulations. Environ. Entomol., 22(5), 1096-1102.

4. Honek A., Jarosik V. and Martinkova Z. 2003. Effect of temperature on development and reproduction in Gastrophysa viridula (Coleoptera: Chrysomelidae). Eur. J. Entomol. 100: 295-300.

5. Jędruszczak M. 1998. Niektóre ekologiczne skutki ochrony przed chwastami [Some ecological results of weeds control], Puławy, 78-84 [in Polish].

6. Kjaer C. 1994. Sublethal effects of chlorosulfuron on black bindweed (Polygonum convolvulus L.). Weed Res., 34, 453-459.

7. Kovalev O.V., Zaitzev V.F. 1996. A new theoretical approach to the selection of promising agents for biological weed control. Proc. IX Int. Symp. Biol. Contr. Weeds, Stellenbosch South Africa, 283-285.

8. Latowski K. 1993. Study of the synanthropic flora of the Balkan Peninsula. Wiad. Bot., 37 (3-4), 71-72.

9. Marocchi G. 1989. New Problems in weed control in Italy. Proc. VII Int. Symp. Biol. Contr. Weeds, Rome, Italy, 633-637.

10. Marocchi G. 1994. A useful insect for weed control. Vita in Campagna 12 (6), 50.

11. Oka I.N., Pimentel D. 1979. Ecological effects of 2,4-D herbicide: Increased corn pest problems. Contr. Centr. Res. Insr. Agric. Bogor., 49, 1-17.

12. Paspatis E.A., 1987. Chemical, cultural and biological control of Oxalis pes -caprae in vineyards in Greece. Proceedings of a meeting of the EC Experts Group, Dublin, 1985, 27-29.

13. Patnaik N.C., Mohapatra L.N., Ghode M.K., and Mishra K. 1987. Effects of some weedicides on the water hyacinth weevil (Neochetina bruchi Hustache) and its weed host. Plant Prot. Bull., 39(3), 23-26.

14. Piesik D. 2000. The occurrence of Gastroidea viridula Deg. and Gastroidea polygoni L. on Rumex confertus Willd. as biological representatives of weed population control. J. Plant Prot. Res., 40(3), 219 - 230.

15. Piesik D. 2001. The occurrence of Hypera rumicis L. (Col., Curculionidae) on Rumex confertus Willd. as an interesting opening for biological weed population control. J. Plant Prot. Res., 41(3), 209 - 215.

16. Piesik D. 2004. Uszkodzenia powodowane przez owady zasiedlające Rumex confertus Willd. [Damages caused by insects on Rumex confertus Willd.]. Acta Sci. Pol. – in press [in Polish].

17. Rakhimberdyeva N.A., Shodiev A. 1989. The knotweed leafbeetle (Alticinae, Chrysomelidae) - phytophage of knotgrass. Uzb. Biol. Zh., 1, 43-44.

18. Rechinger K.H., 1984. Rumex (Polygonaceae) in Australia: a reconsideration. Nuytsia 5 (1), 75-122.

19. Sotherton N.W. 1982a. Observation on biology and ecology of the chrysomelid beetle Gastrophysa polygoni in cereal fields. Ecol. Entomol., 7, 197-206.

20. Sotherton N.W. 1982b. Predation of a chrysomelid beetle (Gastrophysa polygoni) in cereals by polyphagous predators. Ann. Appl. Biol., 101, 196-199.

21. Sotherton N.W. 1982c. The effects of herbicides on the chrysomelid beetle Gastrophysa polygoni (L.) in the laboratory and field. Z. Angew. Entomol., 94, 446-451.

22. Sotherton N.W, Moreby S.J. 1992. The importance of beneficial arthropods other than natural enemies in cereal fields. Aspects Appl. Biol., 31, 11-18.

23. Speight R.I., Whittaker J.B. 1987. Interactions between the chrysomelid beetle Gastrophysa viridula, the weed Rumex obtusifolius and herbicide asulam. J. Appl. Ecol., 24, 119-129.

24. Spencer N.R. 1980. Exploration for biotic agents for the control of Rumex crispus. Proc. Int. Symp. Biol. Contr. Weeds, Brisbane, Australia 1980, 125-151.

25. Thacker J.R.M., Hickman J.M. 1990. Techniques for investigating the routes of exposure of carabid beetles to pesticides. In: Stark, N.E. (ed)., The role of ground beetles in ecological and environmental studies, Intercept, Andower, UK, 105-113.

26. Vickerman G.P., Sotherton N.W. 1983. Effects of some foliar fungicides on the chrysomelid beetle Gastrophysa polygoni (L.). Pest. Sci., 14, 405-411.

27. Watson A.K., Wymore L.A. 1989. Biological control, a component of integrated weed management. Proc. VII Int. Symp. Biol. Contr. Weeds, Rome Italy. 101-106.

28. White T.C.R. 1984. The abundance of invertebrate herbivores in relation to the availability of nitrogen in stressed food plants. Oecologia, 63. 90-105.

29. Whittaker J.B., Ellistone J., Patrick C.K. 1979. The dynamics of a chrysomelid beetle, Gastrophysa viridula, in a hazardous natural habitat. J. Anim. Ecol., 48 (3). 973-986.

Responses to this article, comments are invited and should be submitted within three months of the publication of the article. If accepted for publication, they will be published in the chapter headed ‘Discussions’ in each series and hyperlinked to the article.