Journal of Chemical Ecology

, Volume 36, Issue 11, pp 1207–1210

A Syrphid Fly Uses Olfactory Cues to Find a Non-Yellow Flower

Authors

  • Clara Primante
    • CREAF/Centre de Recerca Ecològica i Aplicacions ForestalsUniversitat Autònoma de Barcelona
    • Department of Plant SystematicsUniversity of Bayreuth
Rapid Communication

DOI: 10.1007/s10886-010-9871-6

Cite this article as:
Primante, C. & Dötterl, S. J Chem Ecol (2010) 36: 1207. doi:10.1007/s10886-010-9871-6

Abstract

Syrphid flies are frequent flower visitors, but little is known about the cues they use to find flowers. We determined the importance of visual and olfactory cues in a flight cage bioassay using Cirsium arvense (Asteraceae) flower heads and experienced Episyrphus balteatus (Diptera, Syrphidae). We tested the response of antennae of the flies to headspace inflorescence scent samples by using gas chromatography coupled to electroantennography (GC-EAD). The bioassays revealed that both sexes of experienced flies rely on olfactory, not visual, cues to find C. arvense flower heads. The GC-EAD measurements demonstrated that male and female flies have olfactory receptors for several of the compounds emitted by the inflorescences. These electroantennographic-active compounds may be responsible for the attraction of flies to the C. arvense flower heads. Among the compounds eliciting an antennal response are methyl salicylate and 2-phenylethanol, which were previously described as syrphid attractants. Overall, our study demonstrates for the first time that a syrphid fly uses olfactory and not visual cues to find a pollen/nectar host-plant.

Keywords

Floral attractionOlfactory cuesVisual cuesEpisyrphus balteatusCirsium arvenseBehavioral assayGC-EADAsteraceaeDiptera

Introduction

Syrphids (Diptera, Syrphidae) frequently visit flowers to feed on their pollen and nectar (Shi et al. 2009). It has been assumed that these flies primarily use vision to seek floral feeding sites, and flower color, size, and shape preferences of syrphids have been studied (e.g., Lunau and Wacht 1994; Sutherland et al. 1999; Lunau et al. 2005; Shi et al. 2009). Among the visual floral cues, color seems to be the most important, and syrphids repeatedly have been found to respond to yellow, and to prefer yellow over other colors (Lunau 1988; Sutherland et al. 1999). This color preference may help the flies find pollen that often is yellow (Lunau and Wacht 1994).

Besides visual plant cues, olfactory cues also may play a role in hoverfly attraction. Volatile organic compounds increase catches of hoverflies in yellow sticky traps (Zhu and Park 2005) and elicit searching behavior for oviposition sites (Harmel et al. 2007). However, almost nothing is known about the importance of olfactory compared to visual cues for finding nectar and pollen (Majetic et al. 2009).

Although yellow is highly attractive for syrphids, they also visit other colored flowers. The floral cues for finding non-yellow flowers are unknown. Cirsium arvense (L.) Scop. (Asteraceae), our study plant, has pink flower heads that are visited by a large variety of insects, among them syrphids (Theis et al. 2007). Episyrphus balteatus De Geer is the most abundant syrphid visitor (Primante and Dötterl unpublished data). In the present work, we focused on the relative importance of visual and olfactory cues of C. arvense flower heads for host-plant identification by experienced E. balteatus. Additionally, we tested headspace inflorescence scent samples on the antennae of flies by gas chromatography coupled to electroantennography (GC-EAD) in order to identify which compounds emitted by C. arvense inflorescences are perceived by the fly.

Methods and Materials

Plant Material and Volatile Collection

The Cirsium arvense plants were from the Ecological Botanical Garden of the University of Bayreuth. C. arvense is a dioecious plant. Staminate and pistillate flower heads emit the same scent compounds, although the total amount of scent is higher in staminate flower heads (Theis et al. 2007; Primante and Dötterl unpublished data). The petals of both sexes reflect light mainly in the blue (max. 450 nm) and red range (650–700 nm) (Primante and Dötterl unpublished data). We, therefore, did not discriminate between flower sexes. Flowering branches were cut in the field and placed in water in the laboratory for immediate scent collection. Four to seven flower heads of C. arvense were enclosed in a polyester oven bag (Toppits, Germany), and over an 8 h period the emitted volatiles were trapped in an adsorbent tube filled with 20 mg of a 1:1 mixture of Tenax-TA (mesh 60–80, Supelco) and Carbotrap (mesh 20–40, Supelco) (Dötterl et al. 2005). Volatiles were eluted with 60 μl of acetone (SupraSolv, Merck KgaA, Germany) to obtain 6 odor samples for the electrophysiological experiments.

Electrophysiological Experiments and Chemical Analyses

Electrophysiological analyses of the floral scent samples were performed with the GC-EAD system as described by Dötterl et al. (2005). Antennae from 23 females and 11 males of E. balteatus (one antenna per individual, one run per antenna) were tested between July and August 2008 on our 6 odor samples (3–5 female and 1–5 male antennae per sample). To identify the EAD-active compounds, 1 μl of the scent samples was analyzed on a Varian 3800 gas chromatograph fitted with a 1079 injector and a ZB-5 column (5% phenyl polysiloxane; length, 60 m; inner diam, 0.25 mm; film thickness, 0.25 μm; Phenomenex) and a Varian Saturn 2000 mass spectrometer (Dötterl et al. 2005). Component identification was carried out using the NIST 08 mass spectral database or MassFinder 3, and was confirmed by comparison of mass spectra and retention times with those of authentic standards.

Behavioral Assay

To determine the relative importance of olfactory and visual cues for finding flowering C. arvense by flies experienced in foraging, we conducted two-choice bioassays in a flight cage at the end of summer 2009.

Pupae of E. balteatus were provided by Katz Biotech Ag (Baruth, Germany) and kept in a small gauze tent (60 × 60 × 60 cm) at 23°C until hatching. Immediately after the adults hatched, they all were transferred to a flight cage (7.20 × 3.60 × 2.20 m) that was set up in a greenhouse. The population of 55 female and 30 male adult flies were fed on fresh flowers of C. arvense, but all floral material was removed from the cage at least 12 h before conducting a bioassay.

We performed two two-choice bioassays to assess fly attraction to floral cues: 1) Visual only vs. visual and olfactory cues combined, and 2) Olfactory only vs. visual and olfactory cues combined. In the assays, flies were offered flowering branches (with 30–40 flower heads each) in quartz glass cylinders constructed to present either visual or olfactory cues, or both. The basic cylinder for testing attraction to olfactory and visual cues in combination consisted of a transparent quartz glass cap and body and a sleeve of Macrolon®, which connected and sealed the cap and body. The Macrolon® sleeve had 60 holes (diam 0.2 cm), arranged in three horizontal lines to allow diffusion of floral scents. The cylinders were mounted on a black polyvinyl chloride (PVC) disc (diam 11 cm) that was attached to a square wooden table. A tube coupled the cylinder to a membrane pump (flow 1 l min−1; G12/01EB, Rietschle Thomas, Puchheim, Germany). A modified transparent cylinder without holes and without connection to a pump was used for testing visual attraction only. A cylinder with holes and the pump, but painted black with semi-matte varnish, was used for testing olfactory attraction only.

The two test cylinders were set up 2 m apart for each of the two bioassays. Each test was conducted for 40 min, and 20 min after beginning the test, the position of the cylinders was exchanged. The behavior of the flies was classified as “approaching” when flies hovered in front of the cylinder but did not land, and “landing” when flies contacted the glass cylinder. To assure that an individual fly was counted only once in a specific test, responding flies were captured.

Results and Discussion

In the bioassays, E. balteatus preferred a combination of both cue modalities over visual cues but not over olfactory cues. When testing a combination of visual and olfactory cues against visual cues, 24 flies were attracted by the combination of both cues and no flies were attracted by the visual cues (chi square observed vs. expected test: χ2 = 24.0; df = 1; P < 0. 001) (Table 1). When testing a combination of visual and olfactory cues against olfactory cues, both types of cylinders had the same attractiveness (chi square observed vs. expected test: χ2 = 0.36, df = 1; P = 0.55) (Table 1). These experiments demonstrate that experienced flies primarily use olfactory cues for seeking C. arvense flower heads, whereas visual cues do not play a significant role. We did not determine whether flies would respond to visual cues in the absence of C. arvense odors. C. arvense flowers reflect light in the blue range of light, and blue was somewhat attractive to E. balteatus (Sutherland et al. 1999).
Table 1

Attractiveness of combined visual and olfactory cues in comparison to decoupled visual (Test 1) and olfactory (Test 2) cues for experienced female and male Episyrphus balteatus. *The sex of one fly was not determined

 

VISUAL

vs.

VISUAL + OLFACTORY

Test 1

Landing

Approaching

 

Landing

Approaching

 

0

0

 

18 (11♀; 7♂)

6 (5♀; 1♂)

 

OLFACTORY

vs.

VISUAL + OLFACTORY

Test 2

Landing

Approaching

 

Landing

Approaching

 

7 (3♀; 4♂)

4 (4♀)

 

10 (8♀; 1♂)*

4 (3♀)*

The GC-EAD measurements revealed the candidate molecules responsible for attraction of flies. Four compounds occurring in the scent of C. arvense flower heads (Theis et al. 2007), consistently elicited antennal responses, in more than 50% of tested antennae from both female and male flies: phenylacetaldehyde, methyl salicylate, dimethyl salicylate, and pyranoid linalool oxide (Fig. 1). Less consistent responses were found for the two coeluting compounds methyl benzoate, which was not described by Theis et al. (2007), and linalool (47% response rate); and for 2-phenylethanol (41% response rate). Although we did not test the isolated EAD-active compounds for attractiveness, our results, and those of other researchers, led us to hypothesize that at least methyl salicylate and 2-phenylethanol are attractants for E. balteatus. These compounds were emitted in abundant amounts from the flower heads used in this study (Fig. 1), and have been described previously as syrphid attractants (Zhu and Park 2005). In a field experiment, they increased the attractiveness of yellow sticky cards (Zhu and Park 2005).
https://static-content.springer.com/image/art%3A10.1007%2Fs10886-010-9871-6/MediaObjects/10886_2010_9871_Fig1_HTML.gif
Fig. 1

Examples of coupled gas chromatographic and electroantennographic detection (GC-EAD) of a Cirsium arvense inflorescence scent sample using antennae of female and male Episyrphus balteatus. a: phenylacetaldehyde; b: (E + Z)-pyranoid linalool oxide; c: methyl salicylate; d: dimethyl salicylate. The abundant compounds acetophenone (1), (E)-furanoid linalool oxide (2), coeluting linalool and methyl benzoate (3) and 2-phenylethanol (4) did not elicit responses. The responses marked with an asterisk are artifacts as shown in the box where we present an enlarged section of the FID and corresponding antennal response of the male. The antennal response did not occur simultaneously with FID peak 1

Recently, Majetic et al. (2009) found that augmentation of inflorescences of Hesperis matronalis with scent (collected at night from H. matronalis) increased visitation by syrphids, whereas the color of this color-polymorphic plant (white vs. purple) did not influence the visitation rate of flies. Their study also showed that syrphid flies respond to floral scent, but they performed the experiments with manipulated inflorescences. Our study demonstrates for the first time that a syrphid fly uses olfactory and not visual cues to find an unmanipulated pollen/nectar host-plant.

Acknowledgments

Gregor Aas helped with plant material and provided the greenhouse and Irmgard Schäffler participated in useful discussions. Lynn Amanda Staff, Joseph Woodring, and two anonymous referees gave valuable comments on an earlier version of the manuscript. CP was supported by a grant (AP2006-03067) from Ministerio de Educación y Ciencia-Spain.

Copyright information

© Springer Science+Business Media, LLC 2010