BEER AS ATTRACTANT FOR ARION VULGARIS MOQUIN-TANDON , 1885 ( GASTROPODA : PULMONATA : ARIONIDAE )

Arion vulgaris Moquin-Tandon, an invasive slug, causes losses in crops all over Europe. Small farmers commonly use beer traps to control this pest. Our aim was to check which components of the volatile fraction of beer were important as slug attractants. Five beer brands available at the Polish market: Żubr, Warka Full, Karpackie Pils, Żywiec and Leżajsk Full, were used in the experiments. We examined the slugs’ behaviour in the laboratory, using a star-shaped olfactometer (two-hour tests), and in the field (three-day tests with traps). Particular beer brands attracted the slugs to various extent, depending on the composition of their volatile fraction: the number of slugs caught in the traps was positively correlated with the content of decanoic acid and negatively with that of acrylic acid N-hydroxysuccinimide ester and CO2. kEy woRds: Arion vulgaris, beer, attractant, pest, trap

Since the effectiveness of beer as slug attractant had already been partially proved (dankowska 2011(dankowska , Piechowicz et al. 2014)), in this study we decided to focus on determining which components of the volatile fraction of different beer brands could possibly play the role of attractants for A. vulgaris.

ANIMALS
In the experiments we used adult A. vulgaris caught in the park surrounding the Institute of Applied Biotechnology and Basic Sciences, University of Rzeszów.No means of chemical plant protection had been used in the park for one year before the slug collecting.No chemicals which could interfere with the slug's olfactory perception of the volatile fraction of the tested beer brands had been used before the experiment.

BEER AS ATTRACTANT
The attractiveness of five popular brands of beer for slugs was tested in the laboratory and in the field (Table 1).The volatile fraction of the beer brands was subject to chromatographic analysis.
Laboratory tests were performed in the Institute of Applied Biotechnology and Basic Sciences, University of Rzeszów, from July to October 2010 and 2013.Modified olfactometer OLF0001 (constructed by Andrzej Zienkiewicz Zakład Remontowo Montażowy Aparatury Laboratoryjnej, Toruń, Poland) was used for the analysis the olfactory preference of the slugs (Fig. 1).
Synthetic air, additionally purified by a carbon filter, was directed using the rotameter and a set of tubes Bev-A-Line IV (producer: USPlasticCorp, USA) into five of six arms of the olfactometer (the sixth arm was closed during the tests).Gas was pumped into each channel at the rate of 100 ± 5 ml/min.Before introducing into the olfactometer, the gas flowed through a flask of 50 ml (1 cm above the surface of the liquid), where a 20 ml sample of the studied beer brand was placed.The outlet of the gas carrying the Selection of the olfactometer arm with the smell of a given beer brand by the slugs was recorded.The position of the slugs in the olfactometer arena was recorded after 120 minutes of their stay in the olfactometer.Each time five individuals were placed in the arena (only vigorously moving slugs were chosen for the experiment).After the 120 minute period of the animals' stay in the middle of the olfactometer arena, none of them was seen to remain there.A total of 50 replicates of the tests were performed, using a total of 250 individuals of A. vulgaris.
Field tests were carried out in the park surrounding the Institute of Applied Biotechnology and Basic Sciences in Werynia (province Podkarpackie, Poland).Traps were placed at the following locations: 1. within 2 m of the drainage ditch, 2. in the middle of the park, and 3. on a small meadow.
Five traps (one for each beer brand) were placed at each location.They were home-made as described by haGnEll et al. (2006b).Bottles made of ethyl polyterephtalane (PET) of 1.5 L volume (producer GTX Hanex Plastic Sp. z o.o.), were used for the trap production.The upper part of the bottle was cut at 12 cm, reversed and fastened again to the bottom part of the bottle.Thus prepared traps were buried in the soil with their upper parts slightly below the soil surface, in order to prevent the slugs from injuring their soles on the sharp edges of the plastic.125 ml of beer was poured into each trap.After three days, the traps were brought to the laboratory where the slugs were counted and identified based on their reproductive anatomy.The test was performed in ten replicates, each using new traps and new portions of beer.
Besides, on each day of the experiment, at 9.00, in all the locations approximate environmental parameters: insolation [lux], temperature [°C], and relative humidity [%], were recorded using a Meter Environmental Conditions DT 8820 (producer: CEM, UK).

CHROMATOGRAPHIC ANALYSIS
The analysis of the volatile fraction of the beer brands was done using a gas chromatograph Varian 450 equipped with an MS detector 240 and a column Varian VS-5MS 30 m × 0.25 mm × 0.25 µm.Headspace Solid-Phase Microextraction (HS-SPME) fibre-PDMS 100 µm (microns) was used.Qualitative analysis was performed using the NIST 08 database.
The sample of beer (2 ml) was poured into a 5 ml glass flask, and then its volatile fraction was sampled using HS-SPME, for 30 min.at 37°C.In order to avoid the influence of changes in atmospheric pressure on the gas pressure in the samples of volatile fraction, they were collected at the same time.The SPME Holder with the analyte adsorbed on the fibre was placed in a gas chromatograph dispenser at 200°C, under helium flow of 1 ml/min.and splitless injection mode.The desorption process was carried out for 5 minutes and the analytes were assayed using the following temperature programme: 50°C for the first 5 minutes → temperature increase at 10°C /min.to 250°C maintained for 10 minutes → temperature increase at 20°C /min.to 300°C maintained for 10 minutes.Peaks identification was carried out using the NIST 08 database.

STATISTICAL ANALYSIS
The beer brands were ranked according to their attractiveness to slugs (mean frequency of selection by slugs) in the field and laboratory tests, and according to the mean relative concentration of the volatile substances.Then, the relationship between the two variables was analysed using the Spearman rank correlation test.
The Bernoulli scheme was used to check if in the laboratory tests the slugs chose randomly among the volatile fractions of beer with particular chemical composition.The experiment of placing five slugs which could potentially choose among five cells was run in 50 replicates.The success was that 0, 1, 2, 3, 4, 5 slugs, respectively, would choose the given beer brand and for each beer brand we would get the most probable number of choices.In order to determine whether the choice was random, we used the test of comparing the fractions.
The similarity of beer brands in terms of attractiveness for slugs in the laboratory was examined using cluster analysis with Ward's method and Taxicab metric.The Kruskal-Wallis test was used to check if the attractiveness of beer for the slugs differed among the locations.

LABORATORY TESTS
The number of selections and ranking of the slug preferences for the individual beer brands are shown in Table 2.The olfactometer channels with the components of the volatile fraction of Leżajsk Full and Żubr were chosen much more frequently than the channels with the smell of other beer brands, and the smell of Warka Full was the least frequently chosen (Table 3).
The cluster analysis of the number of choices of the individual beer brands by the slugs confirmed that the components of the volatile fraction of Warka Full were attractive to the slugs, similarly as the smell of Żywiec, while Karpackie Pils, Żubr and Leżajsk Full were preferred with varying frequency (Fig. 2).

FIELD TESTS
The number of individuals A. vulgaris caught in the traps with beer brands of different chemical composition of volatile fraction, the frequency of choices and ranking of the number of caught animals, are shown in Table 4.The greatest number of slugs was In all the locations the most numerous slugs were caught in the traps with Karpackie Pils, and the smallest number in the traps with Warka Full, while in the case of the traps placed in the middle of the park it was Żywiec.The Kruskal-Wallis test showed that for the traps with Warka Full and Leżajsk Full the distribution of the number of choices of the specific traps depended on their location (Table 5).Furthermore, the analyses of environmental conditions in all the locations showed that they differed in insolation, temperature and humidity (Table 6).

DISCUSSION
The slugs fall into beer traps much more often than into water-containing traps (haGnEll et al. 2006b, kaPPes et al. 2012, Piechowicz et al. 2014).There are thousands of beer brands in the world, and they can vary widely in their chemical composition.This was confirmed in our studies (Table 7).The differences in chemical composition between the beer brands resulted in different olfactory perception by and, consequently, attractiveness to the slugs (see also PERPètE & collin 2000PERPètE & collin , clark et al. 2011)).
Our aim was to check if the five beer brands tested might differ in their ability to attract A. vulgaris.We also attempted to identify the components of volatile fraction which might be responsible for inducing the slug's behavioural reaction.
During the olfactometry tests the olfactometer channel with the volatile fraction of Leżajsk Full was most frequently chosen by A. vulgaris (Table 2).In the field tests the greatest number of animals was caught in the traps with Karpackie Pils.In both kinds of tests the smallest proportion of slugs was attracted by the chemical components of Warka Full (Table 4).
The attractiveness of beer brands for slugs depends on the presence of specific chemicals.At least three ingredients of beer are currently known to affect the attractiveness for slugs: diacetyl (IUPAC name: butane-2,3-dione), acetoin (3-hydroxybutanone) and dihydroxyacetone (IUPAC name: 1,3-dihydroxypropan-2-one) (cranshaw 1997).None of these substances was detected in the studied brands.Nevertheless, it seems that in the field tests the number of slugs caught in the traps was positively correlated with the content of decanoic acid but negatively with acrylic acid, N-hydroxysuccinimide ester and CO 2 .The results of field tests are more difficult to interpret than the laboratory tests, because they may be influenced by additional environmen-tal factors (e.g.ambient temperature or lighting), which in turn affect the rate of beer decomposition (VandERhaEGEn et al. 2007, suáREz et al. 2011, RodRiGuEz-bEncoMo et al. 2012).They can influence the chemical content of the air in the traps.All these factors can be responsible for the variation in the distribution of the number of trap choices by the slugs (Table 5).However, the observed variation in the frequency of choices of individual traps by the slugs (Table 4) indicates that the trap location has a greater effect on the number of trapped animals than on their preferences.On the other hand, the laboratory tests were carried out on groups of individuals, not on single slugs, and the slugs may have been moving along paths laid by their conspecifics.According to our observations, individuals placed separately on the olfactometer arena often do not show any motor activity, which would be desirable during such studies.Our previous study (Piechowicz et al. 2014) demonstrated that there were no significant differences in preferences for the smell of environment in the olfactometer between groups of A. vulgaris, and slugs placed there separately.
The laboratory and field tests confirmed that the chemical composition of Leżajsk Full and Żubr in the laboratory tests, and Karpackie Pils in the field tests attracted more individuals of A. vulgaris than the smells of the remaining beer brands.In both kinds of tests the smell of Warka Full appeared to be the least attractive.The field tests indicate that the presence of decanoic acid in the volatile fraction caused an increase, while acrylic acid, N-hydroxysuccinimide ester and CO 2 -a decrease in the number of trapped slugs.Such observations were impossible to confirm in the laboratory tests.The field studies also confirmed that the variable distribution of the number of choices of traps with specific beer brands by the slugs depended on the traps' location.

Table 1 .
Beer brands used in the study

Table 2 .
Number and frequency of choices of olfactometer channel with smell of different beer brands by A. vul garis

Table 4 .
Number and frequency of individuals of A. vulgaris caught in traps with different beer brands and ranking of the number of trapped slugs caught in the traps located near the drainage ditch.

Table 5 .
Distribution of the number of choices of traps

Table 6 .
Environmental conditions at locations of field tests