Nontarget catches of traps with chemical lures may refer to the flower‐visitation, probable pollination, and feeding of bush crickets (Ensifera: Tettigoniidae)

Abstract The diurnal bees, lepidopterans, and other pollinators are among the most studied flower‐visiting insect taxa. They mostly play distinct functions in temperate grasslands and ecotones of grassland‐forest mosaics (such as in forest steppes). Although orthopterans are widely distributed in these habitats, however, their flower visitation is nearly unknown, especially in the temperate zone. During the development of traps with chemical lures to catch Lepidoptera pests, large numbers of Orthoptera were caught that provide a chance for studying the flower visitation and odor and indirectly the host plant preference of seven temperate zone Tettigoniidae species. Data on the attractivity of isoamyl alcohol‐based semisynthetic lures for Meconema thalassinum and efficiency of phenylacetaldehyde‐based lures on Leptophyes albovittata and Phaneroptera falcata were reported for the first time. Additionally, analysis of nature photos collected from internet sources, as part of a passive citizen science also supports the revealed preference of these species. Based on photos, the studied orthopterans mainly visit Asteraceae species including the most preferred Tanacetum vulgare, Pulicaria dysenterica, Achillea millefolium, Solidago canadensis, and Centaurea scabiosa. Based on catches of volatile traps, the first data were recorded on the attractivity of phenylacetaldehyde‐ and isoamyl alcohol‐based lures on three temperate zone Orthoptera species. Results of a passive citizen science study strengthen these results that may increase the knowledge on the host plant and habitat preference of Orthoptera species.


| INTRODUC TI ON
Pollination is one of the basic requirements of the sexual reproduction of angiosperm plants. The delivery of pollen can happen in several ways, but the role of pollinators, especially insects, is outstanding (Katumo et al., 2022;Kevan, 1999). Beyond pollination, there are many aspects of flower visitation in insect-plant interactions since plants, especially flowers, may provide food source, habitat, shelter, and site for reproduction and oviposition for insects.
Among the flower-visiting insects, the diurnal bees, lepidopterans, and other pollinators are relatively well studied (Hoehn et al., 2008;Kevan et al., 1990). Nevertheless, other less-known taxa such as beetles, orthopterans, hoverflies, and nocturnal lepidopterans (Breadmore & Kirk, 1998;McCall & Irwin, 2006) also may have essential role in the pollination, and have direct or indirect effects on flower adaptation and community dynamics (Frame, 2003;Krupnick et al., 1999;McCall & Irwin, 2006). Many of them are rather known as pests e.g. some Orthoptera, and many noctuids (Haile et al., 2021). The species of the Orthoptera order can provide mutualistic ecological services including seed dispersal (Duthie et al., 2006;Suetsugu, 2018), but they are rarely considered capable of pollination .
There are more than 28,500 known Orthoptera species worldwide (Cigliano et al., 2023;Myers et al., 2000;. They occur in grasslands and ecotones of grassland-forest mosaics (e.g. in forest steppes) in the temperate zone (Nagy et al., 2019) and species-rich assemblages inhabit woodlands, especially in tropical and subtropical areas . They may be herbivores (e.g. species of Caelifera order and many members of Tettigoniidae family (Ensifera)), predators (as some Tettigoniidae e.g. Saga spp.), and omnivores such as the most species of Ensifera order including species of Grylloidea and many other groups (Sanam et al., 2021;Xu et al., 2013). Due to their large biomass, herbivore species play an important role in the food webs of grasslands and beyond that, some flower-visiting species are described as pollinators.
Orthoptera pollination is a little studied phenomenon, and we have valid data on it mainly from the tropical regions, although the connection between orthopterans and flowers originated from the late Jurassic and early Cretaceous periods (Song et al., 2015).
According to Krenn et al. (2016) obligatory flower-visiting orthopterans are rare, and all belong to the Ensifera order. Some Australian species of the Zaprochilinae subfamily feed on flowers but do not pollinate them (Kevan & Baker, 1983). Contrarily, some neotropical crickets and endemic wetas of New Zealand (species of Anostostomatidae and Rhaphidophoridae families) are probable pollinators. Later Lord et al. (2013) Hugel et al. (2010) described Glomeremus orchidophilus (Hugel), a formerly unknown species from the island of Mauritius and Reunion (Mascarene Island in the Indian Ocean), which is a single pollinator of the Angraecum cadetii Bos. orchid . The most intensive studies on Orthoptera pollination were carried out in the Indo-Malayan region during recent years (Tan et al., 2019; when two main types of flowervisiting orthopterans were described. One of them is the group of floriphilic species (which prefer flowers as food sources) containing species of the Phaneropterinae subfamily (e.g Phaneroptera brevis Serville). P. brevis adults and nymphs are commonly seen feeding on the inflorescences of Bidens pilosa Lin. (Asteraceae), a plant native to tropical America and naturalized throughout disturbed areas of warm temperate and tropical regions of the World (Strother & Weedon, 2006). The other group consist of opportunist polyphagous species, which are florivores or facultative florivores. Some of them are known pests (e.g. Valanga nigricornis Bur. and Xenocatantops humilis Ser.), while other Tettigoniidae species (e.g. some Conocephalus spp. and Tremellia timah Gor.-Tan) mainly feed on leaves of bushes and trees and feeding on flowers occasionally (Higginson et al., 2015).
The flower visitation of temperate zone orthopterans is nearly unknown and there is no published data on this topic, and the host plant preference of the species is also mostly unknown. Using data collected by traps baited with chemical lures designed for plant protection monitoring Szanyi et al., 2019) and analyzing nature photos of the potential flower-visiting species here we provide the first data on the flower-visiting, probable pollination, and host plant preferences of seven Tettigoniidae species living in Central Europe. Beyond the scientific novelty, our results help to protect and manage the general pollinator communities of the temperate zone.  (Boros, 1964;Simon, 1952), the most important relict habitats and species have become extinct after water regulation in the 19th and 20th cen-  (Szanyi et al., 2021). The most valuable grasslands are the remains of humid meadows maintaining a large amount of the former species-rich assemblages indicated by orthopterans (Szanyi et al., 2017).

| Sampling area
The Orthoptera fauna of the Bereg Region was formerly discussed considering both the Hungarian and Ukrainian parts of the region . The assemblages of the most common habitat types were also described by Szanyi et al. (2017).

| Sampling with traps
The surveys were carried out from 20th July to 19th October 2014, and from 12th July to 25th October 2015. CSALOMON® VARL+ funnel traps (www.csalo montr aps.com; Figure 2a) were placed in margin of the Velyka Dobron' Forest. In the experiments, two synthetic phenylacetaldehyde-based (PHEN1 and PHEN2) and a semisynthetic bisexual isoamyl alcohol-based lure (SBL, see also e.g. Szanyi et al., 2022) were used parallelly with unbaited control traps (UNB).
The semi-synthetic baits were made from 4 mL polypropylene tubes containing dental rolls soaked in a scent made of isoamyl alcohol, acetic acid, and red wine (1:1:1, 3 mL). The bait mixture F I G U R E 1 Location of the sampling site near Velyka Dobron' village at a forest margin (empty black rectangular) in Transcarpathia (West Ukraine).
evaporated through a 4 mm diameter opening . The SBL lure imitates the odor of tree saps containing acetic acid as a decomposition product of sugars and potentially attracts species living in arboreal habitats.
In 2014, two separate polyethylene bag (PEbag) dispensers were used to bait PHEN lure traps . In these dispensers, a dental roll was included in 1.5 × 1.5 cm polyethylene bags. One of the dispensers contained mixture of phenylacetaldehyde and (E)anethol (1:1) (PHEN1), while the other held phenylacetaldehyde, eugenol, and benzyl acetate (1:1:1) (PHEN2). In 2015, only the bait PHEN2 was used. The compounds evaporated through the wall of the 0.2 mm thick PEbag. PHEN lures contain compounds common in flower scents and they are potentially attractive for flower-visiting insects.
The Vaportape® II pesticide strip, designed particularly for insect traps, was used to kill the trapped insects. The tested trap types (SBL, PHEN and UNB) were used in four repetitions, thus 12 (4 × 3) traps were placed on trees 20 m apart, at 1.8-2.0 m height. The traps were checked and emptied once a week and were also rotated weekly to reduce the bias caused by the location.
Since the tested volatiles are not species specific they attracted a wide range non-target taxa beyond the targeted Lepidoptera pests.
In the caught material there were many non-target Lepidoptera  and Diptera species (Katona et al., 2020). Bumblebees (Hymenoptera: Apidae) containing new taxa for the studied area, and some orthopterans were also caught.
Orthopterans were identified at species level based on the keys of Harz (1957Harz ( , 1975Harz ( , 1960, and the number of specimens caught was determined for each sample. Considering the nomenclature of the species the Orthoptera Species File online database was followed (Cigliano et al., 2023).

| Photo collection and processing
In the case of species attracted by volatile traps, a passive citizen science study was carried out. Photos of flower visitations of the species were collected using Google search following the study of

| Data analysis
The attractivity of the tested baits was compared using the total number and the mean number of caught individuals (individuals/ trap/sample). The assumptions of parametric tests were tested with Q-Q plots (normality) and Levene-test (homogeneity of variances).
Since our data did not fulfill these assumptions, the nonparametric Kruskall-Wallis test was used. If it showed significant differences, treatments were compared with Mann-Whitney U-test.
During the photo analysis, the flower visitation of a given   (Table 1).
Phaneroptera falcata living in bushy habitats and forest edges was the most abundant species in both studied years (Table 1). In 2014, the phenylacetaldehyde-based lures (PHEN) attracted significantly more specimens than both the isoamyl alcohol-based semisynthetic lures (SBL) and unbaited traps. SBL baits did not show a significant effect compared to the unbaited control. In 2015, the same trend could be seen but the difference was significant only between the two baits (PHEN and SBL; Figure 3).
In the case of M. thalassinum, SBL lure was significantly more efficient than the PHEN lures in both studied years. Because of relatively high catches of unbaited traps, the SBL lure could not attract significantly more individuals than empty traps in 2014. Contrary in 2015, when the number of total catches was higher the difference was significant (Figure 3).

F I G U R E 3
Mean number of caught Orthoptera specimens (individuals/trap/sample) attracted by different baits in 2014 (black bars) and 2015 (white bars). Letters show significant differences based on the Mann-Whitney U test (p < .05).
The preferences of C. fuscus showed the same pattern as M.
thalassinum. They were not affected by PHEN baits but showed a clear preference for SBL baits in both studied years. The absolute efficiency of SBL baits against C. fuscus could not be proved, because of relatively high catches of unbaited traps (Figure 3).
In the case of R. nitidula, the SBL baits attracted the most individuals, and the unbaited traps were the less effective, but these differences were not significant in any cases (Figure 3).
Leptophyes albovittata showed a clear preference to the phenylacetaldehyde-based lures (PHEN), while the SBL lure had no influence on it in both years (Figure 3).
Considering T. viridissima and P. griseoaptera the total number of individuals caught was not enough for statistical analysis. T. viridissima was attracted by SBL baits, while P. griseoaptera was caught only in 2015 by both types of traps (Table 1).

| Results of passive citizen science
During the web search, 307 photos of the seven studied species were found. In 189 photos (61.6%), insects are staying on flowers or feeding with them. In 149 pictures 70 plant species belonging to 25 families could be identified at species level (  pests. Each caught species is a common and sometimes abundant member of the local fauna, and they are all widely distributed eurytopic, Eurasian species (Heller et al., 1998;Nagy & Rácz, 2007;Szanyi et al., 2017).
Although each sampled species belongs to the thamnobiont life form (Nagy & Rácz, 2007), they have different habitat preferences.
M. thalassinum, T. viridissima, and P. griseoaptera prefer shrubby vegetation of forest edges (Diekotter et al., 2005), and the first two Temperate zone tettigonoids are mainly mixophagous. They feed both on vegetative and generative parts of different host plants and predate small arthropods parallelly, but we know little about their host plant and prey preferences. Among the first "canopy-living" group of species, M. thalassinum showed a significant preference to SBL bait. This is the first report on volatiles attractive to M. thalassinum, which help us to understand the its food preference. Harz (1957) reported that this small bush cricket is mainly carnivorous and predates small caterpillars and aphids living in the canopy and only secondarily feeds on green parts of the trees. Our result suggests that we should rethink our knowledge since the plant parts, which have a similar chemical character as our SBL lure (e.g. Rosa spp., Prunus spp.; pherobase.com), may have larger importance in its feeding.
The VARL+ traps used are designed for catching large lepidopterans such as Noctuidae and Pyralidae species; however, they are not efficient for as large insects as T. viridissima, and the position of the traps was far from the layer used by flightless P. griseoaptera, which can explain their low abundances in the traps. Both of them have been reported as omnivorous species mostly preying on caterpillars, aphids, small moths, and other small invertebrates (Harz, 1975;Lupu, 2007 Ruspolia nitidula has the same habitat preference as C. fuscus, and it was also attracted by the SBL lure, but in this case, the PHENs have a lower effect. Since there were no significant differences, the exploration of the effect of the lures on these two species needs further investigation.
In the case of L. albovittata Harz (1957) and Marini et al. (2009) reported that they like to stay on shrubs such as blackthorn (Prunus Phaneroptera brevis is one of the few known floriphilic Orthoptera species. It lives in tropical South-East Asia (Strother & Weedon, 2006), while in Hungary there are two Phaneroptera species (P. falcata and P. nana Fieber) can be found. They are omnivorous species (Harz, 1960) without any detailed data on their host plants in Europe. Here, we pro- of PHEN baits that also support our findings (Jürgens et al., 2003).
In the passive citizen science study, large ratio of P. falcata and L.
albovittata photos show these species on flowers, which refer to the flower-visiting habit of them, that support our findings. In the case of P. falcata one of the preferred plants Achillea millefolium contains eugenol (Anderson, 2003;Shawl et al., 2002), while on some pictures, this bush cricket could be seen on Syringa vulgaris containing phenylacetaldehyde (Brunke et al., 1992;Mookherjee et al., 1990) and Mirabilis jalapa containing eugenol, phenylacetaldehyde, and benzyl acetate together (Levin et al., 2001).
In the case of other plant species appearing in the photos, the presence of the compounds is not known.
These results show that our knowledge of the host plant and habitat preference of these relatively well-known species needs revision. Further investigations can clarify the feeding habits and ecological requirements of these and other related species and help to understand their role in local ecosystems.

Szabolcs Szanyi's research was financed by the National Research
Development and Innovation Office (NKFIH, grant PD 138329).

CO N FLI C T O F I NTE R E S T S TATE M E NT
The authors declare no competing interests.

DATA AVA I L A B I L I T Y S TAT E M E N T
The datasets generated during and/or analyzed during the current study are available in the ZENODO repository, Nagy et al. (2022).