Welcome | News & Events | Cover Crops | Insectary Plantings | Strip Tillage | Links

Insectary Plantings

Enhancing Biological Control with Beneficial Insectary Plants

July, 1998

Introduction

Beneficial insectary planting refers to intentionally introducing flowering plants into agricultural ecosystems to increase pollen-and nectar-resources required by natural enemies of insect pests. Several species of natural enemies, including aphidophagous (aphid-feeding) hoverflies (Diptera: Syrphidae) and parasitic hymenoptera, depend on pollen and nectar for reproductive success and longevity (Schneider 1948, Jervais 1986). Surveys of weed and wild-plant compositions in agroecosystems have associated florally abundant, non-crop habitat with higher numbers of pollen and nectar-feeding natural enemies in and around farm fields (Cowgill 1989, Cowgill et al. 1993) and orchards (Leius 1967). However, due to agricultural practices such as frequent cultivation and herbicide applications, many farm fields have few flowering plants present which may limit the potential benefit of these insects for biological control.
Several studies have demonstrated the potential for establishing flowering plants in or around farm fields to attract natural enemies and enhance biological control of crop pests in adjacent fields (Kloen and Altieri 1990, White et al. 1995, Hickman et al. 1996). However, natural enemies are selective in their feeding and show preferences for certain plant species (Cowgill et al. 1993, Lunau 1994).

The goals of this research project were to asses the potential of a number of flowering plants as beneficial insectary plants and to evaluate strategies for using insectary plants to enhance biological control. Three field experiments were conducted in 1997 to evaluate the relative attractiveness of potential beneficial insectary plants; one additional on-farm experiment evaluated the effectiveness of interplanting two species of insectary plants in a broccoli crop to enhance biocontrol of aphids.

Project Objectives 1. Evaluate relative attractiveness of selected flowering plant species to adult aphidophagous hoverflies and parasitic hymenoptera. 2. To identify species of hoverflies associated with selected insectary plants. 3. To evaluate the potential of interplanting alyssum or cilantro in a broccoli field to attract hoverflies and suppress cabbage aphid populations.

Procedures
Relative Attractiveness Studies. The relative attractiveness of 11 flowering plants to aphidophagous hoverflies (Diptera: Syrphidae) and parasitic wasps (Hymenoptera: Ichneumonidae and Brachonidae) was evaluated at the Oregon State University Vegetable Research Farm. Flowers included seven annuals: annual alyssum (Lobularia maritima), calendula (Calendula officinalis), cilantro (Coriandrum sativa), mustard (Brassica juncea), Phacelia (Phacelia tanacetifolia), buckwheat (Fagopyrum esceulentum), and marigold (Tagetes patula ) and four perennials: yarrow (Achillea millefolium), Korean licorice mint (Agastache rugosa) fennel (Foeniculum vulgare) and perennial alyssum (Aurinia saxitalis). Flowering plants were grown in 1 m2 plots in a complete randomized-block design with four replications. Six of these plants were also evaluated for attractiveness to aphidophagous hoverflies at two organic farms (Denison Farm, Corvallis, and Persephone Farm, Lebanon) including, three annuals: annual alyssum, phacelia, and cilantro, and three perennials: yarrow, Korean licorice mint, and fennel. At the two on-farm sites, flowers were grown in 1 m2 plots in a complete randomized-block design with three replications.
Relative attractiveness of flowering plants to hoverflies was assessed by conducting timed observations of feeding-visit frequencies. Numerical abundance of parasitic hymenoptera (Brachonidae and Ichneumonidae) wasps was estimated by timed vacuum sampling. Blooming times of plant species varied, and evaluations were only made during blooming periods.
A sweep net was used to collect hoverflies associated with each insectary flower species; representatives of these hoverflies were sent to Germany and identified by a hoverfly systematist. This reference collection was then used to identify all hoverfly species collected in the experiments.

On-Farm Insectary Interplanting in Broccoli. Alyssum and cilantro were interplanted in a 30-acre commercial broccoli field to asses the potential to attract adult hoverflies and suppress aphid populations on broccoli. Plots were 60 x 60 ft and treatments included: alyssum interplanted with broccoli, cilantro interplanted with broccoli and a broccoli monoculture. The experimental design was a complete randomized block design with three replications of each set of treatments, with a 60 ft buffer zone of untreated broccoli plants between each treatment block. Alyssum and cilantro was transplanted at the same time the brocolli was transplanted into the field.
Adult hoverfly activity was monitored with yellow pan traps and hover fly egg and aphid densities on the broccoli was assessed by counts from 60 randomly pulled broccoli leaves within each treatment plot. All sampling was conducted on a weekly basis. Parasitism rates of the cabbage aphid by the aphid parasitoid, Dieretiella rapae, was determined by counting aphid "mummies" on the broccoli leaves.

Results
Relative attractiveness studies. A total of 15 species of aphidophagous hoverflies were collected and identified. Although species varied somewhat among the 3 experimental sites, the six most common aphidophagous species of hoverflies caught in sweep nets were: Meliscaeva cinctella, Toxomerus marginatus, T. occidentalis, Sphaerophoria sulphuripes, S. pyrrhina, and Scaeva pyrastrib (Table 1). Attractiveness of flowering plants to hoverflies differed by dates and sites. Among early-season flowering species, cilantro was the most attractive to hoverflies followed by alyssum and buckwheat. Mustard, buckwheat, and Korean licorice mint were most attractive to parasitic hymenoptera. Among late-season flowers, fennel and Phacelia were most attractive to hoverflies, but attractiveness to parasitic wasps did not differ among evaluated plants.

Table 1. Hoverfly species present at experimental sites and associated flower hosts in 1997.

Hover fly species	Locationa	Flower hostb
Allograpta micrura	P	cl
Eupoedes fumipennis	P	fn
E. lapponicus	P O	ag ph
Melanostoma mellinum	P	cl
Meliscaeva cinctella	P D O	bk fn yr
Paragus variables	D	yr
Parasyrphus insolitus	D	ag
Scaeva pyrastri	P D	ph
Sphaerophoria sulphuripes	P D O	al au ca cl mu yr
S. opinator	P D O	fn ph yr
Toxomerus marginatus	P D O	al cl ph
T. occidentalis	P D O	ag al cl fn yr

a P = Persephone farm; D = Denison Farm; O = OSU Vegetable Research Farm. b al = alyssum; ag = agastache; au = aurinia; ca = calendula; cl = cilantro; fn = fennel; mu = mustard; ph = phacelia; yr = yarrow.

In this study, as in previous studies, foraging hoverflies exhibited a high degree of selectivity in their feeding (Table 2). Based on our fixed time-of-day observations of hoverfly feeding preferences, clearly certain plants evaluated in this study would be better suited as insectary plants. This approach, however, ignores differences in diurnal nectar flow among flowering plants, which would have a major impact on relative attractiveness based on a single period of observation. Also likely factors that could contribute to erroneous results is differences in hoverfly flight behavior under windy conditions. Although strong flyers, hoverflies have difficulty landing on flowers under windy conditions, and lower plants (such as alyssum) would experience less wind than taller plants like yarrow or fennel. Our research site at the OSU Research Farm was usually quite windy during the late morning (when we observed feeding behavior) and the winds would usually subside in late afternoon.

Table 2. Mean number of adult aphidophagous hoverflies (± SEM) observed visiting flowering plants per 2 min at Oregon State University Vegetable Research Farm in 1997.

	Date
Flower	7 June	14 July	24 July	30 July	13 Aug	21 Aug	29 Aug	2 Sept
Alyssum	3.8 ± 1.0aa	3.0 ± 1.1ab	1.3 ± 1.0b	1.0 ± 0.4ab	-	-	-	-
Buckwheat	3.3 ± 1.1a	3.0 ± 1.0ab	-	-	-	-	-	-
Calendula	0.8 ± 0.5b	2.3 ± 0.8ab	0.0± 0.0c	-	-	-	-	-
Cilantro	-b	4.0 ± 1.1a	5.3 ± 1.0a	2.0 ± 0.6a	-	-	-	-
Marigold	-	-	0.5 ± 0.5b	0.3 ± 0.3b	0.0 ± 0.0	0.0 ± 0.0	-	-
Mustard	1.0 ± 0.7b	4.0 ± 0.7a	-	-	-	-	-	-
Phacelia	-	-	-	-	1.5 ± 0.7	-	-	-
Agastache	-	-	-	0.0 ± 0.0	0.0 ± 0.0	0.0 ± 0.0	0.0 ± 0.0	0.0 ± 0.0
Aurinia	2.0 ± 0.9ab	1.3 ± 0.5b	0.8 ± 0.5b	0.3 ± 0.3b	-	-	-	-
Fennel	-	-	-	-	-	0.3 ± 0.1	0.00 ± 0.00	0.3 ± 0.3
Yarrow	-	-	-	0.3 ± 0.3b	0.3 ± 0.5	0.3 ± 0.3	0.5 ± 0.2	0.3 ± 0.3

aMeans followed by different letters within a column are significantly different at a = 0.05. b Dashes indicate plant was not > 50% in bloom. c Observations with 0.00 mean and 0.00 SEM are not included in analysis.

On-Farm Insectary Interplanting in Broccoli. In the interplanting study, Toxomerus occidentalis and T. marginatus were the two most abundant hoverflies caught in yellow pan traps. Throughout the season more T. occidentalis were caught in pan traps in the alyssum plots compared with controls (p < 0.05) (Fig. 1). Analysis of pan-trap catches by date suggest that increased hoverfly activity associated with alyssum plots extended into the adjacent field up to 15 m on the date of peak hoverfly catches.
Aphid sampling revealed no significant differences among mean estimates of aphid densities on the broccoli, however the overall density of aphids in the experimental blocks was considered a rather low level of aphid infestation. More hoverfly eggs were found in alyssum plots than in the control plots on the second to last sampling date (p< 0.05), when the aphid populations had began to increase (Fig. 2). We were not, however, able to identify the species of hoverfly eggs found on the broccoli leaves.
Evidence of an association between hoverfly egg laying and aphid densities suggests that egg laying did not occur until late season because aphid populations had not built up to a critical point earlier. Several published studies have documented a density-dependent oviposition response by some hoverfly species, however other species lay eggs on the leaves independent of aphid densities. No oviposition behavior has been reported for the hoverfly species trapped in our experiment.
Significantly higher numbers of parasitized aphids were found in the alyssum plots than either the cilantro or control plots (Fig. 3), with rates of parasitism nearly doubling. Other studies have shown that D. rapae, the parasitoid of the cabbage aphid, to be widely distributed so the effects demonstrated in this study could have implications across many areas.

Fig.1. Mean number of female T. occidentalis caught in pan traps in treatment and control plots from 22 June to 25 July, 1997.

Fig. 2. Mean number of hover fly eggs found on broccoli leaves in treatment and control plots on 16 July and 23 July, 1997.

Fig. 3. The percentage of aphids parasitized by Diaeretiella rapae on broccoli leaves in treatment and control plots from 25 June to 23 July, 1997.

Conclusions
This work has shown clear evidence of selectivity of potential insectary flowers by both hoverflies and parasitic hymenoptera. Some flowers appear to be strongly attractive to these natural enemies, some were flowers distinctly less attractive, and some exhibited no attractiveness whatsoever. Interpretation of these data must be tempered by the limitations of a single time-of-day sampling period, which could miss key periods of feeding behavior by some hoverfly species. We can conclude, however, that insectary plants in our trials clearly serving as food resources for hoverflies include annual alyssum, cilantro, buckwheat, mustard, phacelia, fennel and yarrow.
We have also shown a differential attraction among hoverfly species to various insectary plants. In addition, we have identified the predominant species of hoverflies ocurring in western Oregon and have developed a taxonomic reference collection housed in the Department of Horticulture at OSU. This collection is available to anyone interested in hoverfly identification.
In the on-farm interplanting study, there was strong evidence that annual alyssum attracts more hoverflies into the experimental plot areas and more hoverfly eggs are layed on the broccoli plants. There was also an "alyssum" effect extending up to 45 feet outside the edge of the treatment blocks which increased numbers of hoverflies trapped.
In this trial, there was no apparent effect of the interplanted flowers on abundance of aphids on the broccoli plants. Aphid population levels were generally low throughout the experimental plots.
These experiments clearly demonstrate the potential of insectary plantings for increasing natural enemy abundance and behavioral response (increased egg laying and parasitism) in a commercial scale field setting. Our experiments have not, however, shown evidence of pest reduction through the use of insectary plants. Merely increasing natural enemy abundance or even actually increasing parasitism rates (as in the case of increased rates of parasitism of the cabbage aphid by D. rapae in this experiment) doesn't necessary translate to an "biologically significant" level of improved pest control which reduces or eliminates the need for insecticidal control. Larger scale experiments will be required in which whole fields will Many questions remain concerning the development of practical, "farmer-friendly" recommendations for using insectary plants to enhance biological pest control.

Acknowledgments
Special thanks are extended to Persephone Farm, Denison Farm, and Stahlbush Island Farm for their participation in this project. We also thank Christian Kassebeer for his invaluable taxonomic identification of hoverflies. This project was funded in part by the Organic Farming Research Foundation, the Oregon Department of Agriculture, Center for Applied Agricultural Research, and the USDA Sustainable Agriculture Research and Education Program.

References Cited
Cowgill, S. 1989. The role of non-crop habitats on hoverfly (Diptera: Syrphidae) foraging on arable land. Brighton Crop Protection Conference. 1103-1108.
Cowgill, S. E., S. D. Wratten and N. W. Sotherton. 1993a. The effect of weeds on the numbers of hoverfly (Diptera: Syrphidae) adults and the distribution and composition of their eggs in winter wheat. Ann. Appl. Biol. 123: 499-515.
Hickman, J. and S. D. Wratten. 1996. Use of Phacelia tanacetifolia strips to enhance biological control of aphids by hoverfly larvae in cereal fields. J. Econ. Entomol. 89: 832-840.
Jervis, M. A., N. A. C. Kidd, M. G. Fitton, T. Huddleston and H. A. Dawah. 1993. Flower visiting by hymenopteran parasitoids. J. Nat. Hist. 27: 67-106.
Kloen, H. and M. Altieri. 1990. Effect of mustard (Brassica hirta) as a non-crop plant on competition and insect pests in broccoli. Crop Protection. 9: 90-96.
Leius, K. 1967. Influence of wild flowers on parasitism of tent caterpillar and coddling moth. Can. Entomol. 99:444-446.
Lunau, K. and S Wacht. 1994. Optical releasers of innate proboscis extension in the hoverfly Eristalis tenax L. (Diptera: Syrphidae). J. Comp. Physiol. 174:575-579.
White, A. J., S. D. Wratten, N. A. Berry and U. Weigmann. 1995. Habitat manipulation to enhance biological control of brassica pests by hoverflies (Diptera: Syrphidae). J. Econ. Entomol. 88: 1171-1176.