Insect Visitors and Potential Pollinators of Orchis militaris (Orchidaceae) in Southern Belgium.

As part of a research project on the food deception strategy in Orchis militaris (L.), the objective of this study was to identify insect visitors and potential pollinators of this orchid species in Belgium. In 2013, insects were collected over 2 d per site in five localities distributed in Southern Belgium (Wallonia). A total of 104 insects belonging to 49 species were caught. Dipterans were the most abundant visitors (50% of total specimens), followed by Hymenopterans (32%). Rhingia campestris Meigen, Bombylius venosus Mikan, Apis mellifera (L.), and Bombus lapidarius (L.) were the most abundant species. Only five specimens bore one to more than 10 pollinia: four honeybees (A. mellifera) and one bumblebee worker (B. lapidarius). These two species should be considered as potential pollinators in the study area, but probably not confirmed ones.

Orchidaceae are one of the most diversified Angiosperm families, with more than 27,000 accepted names (The Plant List 2013). The majority of orchid species rely on biotic pollen vectors to reproduce (Nilsson 1992), so do 88% of flowering plant species (Ollerton et al. 2011). Usually, plants attract pollinators with floral signals that advertise rewards (Schiestl and Johnson 2013). However, some plant species also display attractive floral signals but don’t provide any reward (or low or wrong reward) (Jersáková et al. 2009); they are called deceptive species (Renner 2006). Deceptive pollination occurs in about 30% of orchid species (Jersáková et al. 2006, Claessens and Kleynen 2011), and the commonest systems involve the exploitation of the food-seeking behavior of the pollinators (food deception—Jersáková et al. 2006).

Under natural conditions, insect visits on rewardless flowers are rare events difficult to record (Cozzolino et al. 2005, Claessens and Kleynen 2011), making direct observation time consuming (Widmer et al. 2000). Indirect approaches have been developed to study more easily orchid-pollinator relationships: pollinaria carried by insects (caught on rewarding plants growing in the vicinity of studied orchid populations) allow the identification, through molecular techniques, of the corresponding orchid species (Widmer et al. 2000); pollinaria found on insects can also be identified through specific morphological features (Singer et al. 2008, Micheneau et al. 2009); scales and hairs found on pollinated stigmas can allow the identification of the corresponding insect species (Micheneau et al. 2009). Videotape observations are another alternative to direct observations (Micheneau et al. 2009, Lortie et al. 2012, Nakase and Suetsugu 2016), but observation-based techniques don’t allow accurate identification of numerous insects (especially dipterans and hymenopterans). The latter have to be collected and examined carefully in order to draw up accurately visitor and potential pollinator lists.

In the genus Orchis Tourn. ex L. (Orchidaceae), none of the species produce nectar and the majority depends on food deception to attract pollinators (Scopece et al. 2014). Visitor and pollinator guilds have been studied more or less extensively according to the species (e.g. van der Cingel 1995, Berger 2003, 2004; Claessens and Kleynen 2011). In O. militaris (L.), most data were collected in Austria by Vöth (1987) and in the United Kingdom by Farrell (1985). Other contributions come from Berger (2003, 2004), Ebert and Rennwald (1993), Petit (1998), Vöth (1999), etc. Among (altitudinal) regions, a plant species can be pollinated by different insect species. This spatial variation in pollinator identity is well known in various species, including orchids (e.g. Schatz et al. 2005, Gómez and Zamora 2006, González et al. 2014, Sun et al. 2014, Gross et al. 2016). The aim of this study was thus to determine the guilds of floral visitors and potential pollinators to O. militaris in a region that was not widely studied so far: Southern Belgium (Wallonia).

Materials and Methods

Study Species

O. militaris (Fig. 1) is a perennial herb with a Eurasian-Mediterranean distribution area (Kretzschmar et al. 2007). The species is confined to calcareous soils and occurs in sunny or partially shaded sites. It ranges from sea level to 2,200 m and it can grow in chalk grasslands, sand dunes, wet meadows, abandoned cultivated lands, scrub communities or open woodlands. O. militaris appears as a moderately long stem, generally varying from 20 to 45 cm, with a single inflorescence bearing 10–40 hermaphroditic flowers (Farrell 1985, Delforge 2012). It blooms from April to June (-July) (Delforge 2012). O. militaris is assumed to be self-compatible (Neiland and Wilcock 1998, Metsare et al. 2015) and non-autogamous (Delforge 2012), and it depends on the food deception strategy to attract pollinators (Claessens and Kleynen 2011). Low levels of fruit set (<30%) are common (Tremblay et al. 2005, Claessens and Kleynen 2011).

Fig. 1.

O. militaris visited by A. mellifera (Wijre-akkers, Netherlands, 1 June 2010, photographer: Jean Claessens).

Study Sites

The study was carried out in 2013 in five Belgian populations (Table 1). According to Alexandersson and Ågren (1996), a population was defined as a group of inflorescences separated from their closest conspecific by at least 100 m. All populations were at least 5 km apart, except those of Visé, which were 800 m apart (but separated by a canal 140 m wide). The biotopes were various: recolonized slag heaps, semi-dry calcareous grassland, mesic grassland, thickets and deciduous forestry plantations.

Table 1.

Characteristics of O. militaris populations

Site no.	Municipality	Site name	Coordinates	Altitude (m)	Population size	Date of insect collections
A	Visé	Friche du canal Albert	50° 46′16″ N 5° 41′’04″ E	55	∼10,000	4–5 June 2013
B	Visé	Oseraie de Lanaye	50° 46′45″ N 5° 40′59″ E	65	∼1,500	6–7 June 2013
C	Virton	Contournement de Virton	49° 34′38″ N 5° 30′58″ E	250	∼150	16–17 June 2013
D	Musson	Crassier de Musson	49° 33′09″ N 5° 43′04″ E	290	∼230	27–28 May 2013
E	Rouvroy	Nature reserve ‘Raymond Mayné’	49° 30′34″ N 5° 28′53″ E	295	∼80	11–12 June 2013

The population size is given in terms of number of inflorescences.

Insect Collection

Field work took place during sunny and warm (at least 18 °C) days with little or no wind, between 8.00 a.m. and 8.00 p.m. between 27 May 27 and 17 June (Table 1). We made around 30 total hours of captures (180 10-min censuses), thus ∼6 h (36 10-min censuses) per population. Each population was studied during 2 consecutive days, and the 6 h were regularly distributed throughout the 2 days. During 10-min censuses, taking a standard route (100 m) across the population, we used a butterfly net to catch all insects that landed on O. militaris inflorescences. We put them in individual pill-boxes containing a piece of cotton wool soaked with ethyl acetate. Pill-boxes were conserved in a cool box (EDA Plastiques, Oyonnax, France) with cold accumulators (Kern Frio S.A., L’Hospitalet de Llobregat, Spain) and then in a freezer at −18 °C. When insects bore pollinia, we used morphological characters in order to distinguish the pollinia of O. militaris from those of other orchids found in the study sites and their surroundings [Anacamptis pyramidalis (L.) Rich., Neottia ovata (L.) Bluff & Fingerh., Ophrys apifera Huds., O Ophrys insectifera (L.), Orchis anthropophora (L.) All., and Platanthera chlorantha (Custer) Rchb.]. Insects were mounted and identified. Most of the specimens are kept at the Biodiversity Research Centre (Université catholique de Louvain); some dipterans are stored at the Royal Belgian Institute of Natural Sciences.

Results

We caught a total of 104 insects belonging to 49 species, 21 families and four orders (Table 2). Most of the specimens were dipterans (50%) and hymenopterans (32%). Rhingia campestris Meigen (14 individuals) and Bombylius venosus Mikan (10 individuals) were the most abundant dipterans; Apis mellifera (L.) (8 individuals) and Bombus lapidarius (L.) (6 individuals) were the most abundant hymenopterans. Some lepidopterans and coleopterans were also collected. Syrphids were the most diverse family with 12 different species. Pollinia (only from O. militaris) were only found in hymenopterans: four honeybee workers (A. mellifera) bore respectively 13, 12, 3 (attached to the clypeus) and 2 pollinia, and one bumblebee worker (B. lapidarius) bore 1 pollinium. A. mellifera was caught in three sites and B. lapidarius in four of the five sites. Because ethyl acetate fumes rapidly dissolve the viscidium glue (Peter and Johnson 2013), we failed to determine the position of pollinia on four of the five insects. A number of insects were also observed but not caught on O. militaris inflorescences: Clytus arietis (L.) (Coleoptera, Cerambycidae), Harmonia axyridis Pallas (Coleoptera, Coccinellidae), Eristalis sp. (Diptera, Syrphidae), Sarcophaga sp. (Diptera, Sarcophagidae), Pyrausta aurata Scopoli (Lepidoptera, Crambidae), Polyommatus icarus Rottemburg (Lepidoptera, Lycaenidae), Vanessa sp. (Lepidoptera, Nymphalidae), and Anthocharis cardamines (L.) (Lepidoptera, Pieridae).

Table 2.

Identity and number of insects caught on O. militaris inflorescences

Order	Family	Taxon	A	B	C	D	E
Coleoptera	Elateridae	Indet.			1
	Scarabaeidae	Phyllopertha horticola (L. 1758)		2	2
		Valgus hemipterus (L. 1758)		2
Diptera	Agromyzidae	Indet.		1
	Asilidae	Neoitamus cf. socius (Loew 1871)					1
	Bombyliidae	Bombylius major (L. 1758)				3
		B. venosus (Mikan 1796)			3	1	6
	Empididae	Empis discolor (Loew 1856)					1
		Empis femorata (F. 1798)	1
		Empis livida (L. 1758)		1
	Hybotidae	Hybos sp.				1
	Scathophagidae	Scathophaga stercoraria (L. 1758)			1
	Syrphidae	Chrysotoxum cautum (Harris 1776)	3	1
		Episyrphus balteatus (De Geer 1776)	1
		Eupeodes nitens (Zetterstedt 1843)					1
		Helophilus pendulus (L. 1758)		1
		Merodon equestris (F. 1794)		1
		Platycheirus albimanus (F. 1781)	1		1
		Platycheirus peltatus (Meigen 1822)					1
Order	Family	Taxon	A	B	C	D	E
		Platycheirus scutatus (Meigen 1822)			1
		R. campestris (Meigen 1822)			4	3	7
		Sphaerophoria scripta (L. 1758)			1		1
		Syritta pipiens (L. 1758)	1	1	1
		Volucella bombylans (L. 1758)	1
Hymenoptera	Andrenidae	Andrena carantonica (Perez 1902)					1(♀)
		Andrena flavipes (Panzer 1799)			1(♀)
		Andrena nigroaenaea (Kirby 1802)					1(♀)
	Apidae	A. mellifera (L. 1758)	4(☿)	3(☿)			1(☿)
		Bombus hypnorum (L. 1758)		1(☿)			1(☿)
		B. lapidarius (L. 1758)	2(☿)		1(☿)	2(♀, ☿)	1(☿)
		Bombus lucorum (L. 1761)		1(♂)
		B. pascuorum (Scopoli 1763)		1(☿)	1(☿)
		Bombus pratorum (L. 1761)					4(3♂, 1 ☿)
		Bombus terrestris (L. 1758)	1(♀)
	Colletidae	Hylaeus confusus (Nylander 1852)			1(♂)
	Halictidae	Lasioglossum fulvicorne (Kirby 1802)					1(♀)
		Lasioglossum malachurum (Kirby 1802)	1(♀)
		Lasioglossum parvulum (Schenck 1853)					1(♀)
		Sphecodes sp.	1(♀)
	Megachilidae	O. bicolor (Schrank 1781)				1(♀)
Lepidoptera	Crambidae	Pyrausta sp.			1
	Geometridae	Siona lineata (Scopoli 1763)		1
	Hesperiidae	Erynnis tages (L. 1758)					1
		Carterocephalus palaemon (Pallas 1771)					2
	Nymphalidae	Aglais io (L. 1758)		1
		Lasiommata megera (L. 1767)			1
		Melitaea cinxia (L. 1758)		1
	Noctuidae	Euclidia glyphica (L. 1758)			1	1
	Pieridae	Pieris napi (L. 1758)			2

Formicidae were present in all the populations but weren’t caught. ♀ = female; ♂ = male; ☿ = worker.

Discussion

In this study, O. militaris attracted a variety of visitors, mainly hoverflies (32 individuals), bumblebees (16 individuals) and bee flies (13 individuals), probably through the display of general floral signals (typical for rewarding plants) such as showy flower colors, presence of a spur, etc. Rewarding plants attract insects and non-rewarding O. militaris individuals flowering nearby benefit from chance visits (Jersáková et al. 2009). All caught dipterans and lepidopterans have to be considered as visitors (sensu Schatz et al. 2005) since they bore no pollinia. To our knowledge, dipterans and butterflies are never cited as pollinators in the literature. The former visit various orchid species for nectar (absent in deceptive species like O. militaris, which exploit the food-seeking behavior of insects and lure visitors—Jersáková et al. 2006), substances secreted by the stigma, pollen or occasionally to find a shelter; lepidopterans seek nectar essentially in species with a long spur (Berger 2003), like Gymnadenia spp. or Platanthera spp. Caught and observed coleopterans were “simple visitors” (sensu Berger 2003); they apparently landed randomly on inflorescences and never actively visited flowers. A. mellifera was already known as a potential pollinator (i.e. insects collected in an inflorescence while carrying one or more pollinia—after Schatz et al. 2005) of O. militaris but it was apparently not the case of B. lapidarius, although this species was previously observed on O. militaris in different countries (Table 3). The high number of pollinia found on two of the four caught honeybees (13 and 12, respectively) indicates repeated visits to O. militaris.

Table 3.

Non-exhaustive list of insects observed on O. militaris inflorescences in different countries

Order	Family	Taxon	Country (ISO code)	References
Coleoptera	Cantharidae	Cantharis lateralis (L. 1758)	RU	12
		Cantharis rustica (Fallen 1807)	GB	9
	Cerambycidae	Pachyta quadrimaculata (L. 1758)	RU	10
		Rutpela maculata (Poda 1761)	GB	9
	Cetoniidae	T. hirta (Poda 1761)	RU, RU, AT	10*, 12*, 15*
	Chrysomelidae	Cassida viridis (L. 1758)	GB	9
		Smaragdina salicina (Scopoli 1763)	RU	12
	Malachiidae	Malachius bipustulatus (L. 1758)	RU	12
		Malachius coccineus (Waltl 1838)	RU	12
	Oedemeridae	Oedemera nobilis (Scopoli 1763)	RU	10, 12
Diptera	Bibionidae	Bibio marci (L. 1758)	GB	9
	Bombyliidae	B. major (L. 1758)	AT	13
		B. venosus (Mikan 1796)	AT	13
	Calliphoridae	Onesia sp.	GB	7
	Empididae	Empis tessellata (F. 1794)	FR, GB	3, 9
	Muscidae	Musca sp.	AT	13
		Thricops semicinereus (Wiedemann 1817)	GB	7
	Opomyzidae	Opomyza germinationis (L. 1758)	GB	7
	Scathophagidae	S. stercoraria (L. 1758)	GB	7
	Syrphidae	Baccha elongata (F. 1775)	GB	7
		C. cautum (Harris 1776)	GB	9
		Chrysotoxum octomaculatum (Curtis 1837)	AT	13
		E. balteatus (De Geer 1776)	GB, AT	9, 13
		Eristalis pertinax (Scopoli 1763)	GB	9
		Leucozona lucorum (L. 1758)	GB	7
		Melanostoma mellinum (L. 1758)	GB	7
		Melanostoma scalare (F. 1794)	GB	9
		Platycheirus ambiguus (Fallen 1817)	AT	13
		P. scutatus (Meigen 1822)	GB	7
		R. campestris (Meigen 1822)	GB	7
Hemiptera	Cercopidae	Cercopis vulnerata (Rossi 1807)	GB	9
	Scutelleridae	Eurygaster spp.	GB	9
Hymenoptera	Andrenidae	A. carantonica (Perez 1902)	AT	13
		Andrena cineraria (L. 1758)	RU	12*
		Andrena curvungula (Thomson 1870)	FR	8*
		A. enslinella (Stoeckhert 1924)	AT	13*, 14*
Order	Family	Taxon	Country (ISO code)	Reference
		A. flavipes (Panzer 1799)	AT	13
		Andrena hattorfiana (F. 1775)	AT	13*, 14*
		Andrena helvola (L. 1758)	AT	13
		Andrena minutula (Kirby 1802)	AT	13
		Andrena subopaca (Nylander 1848)	AT	13
		Andrena taraxaci (Giraud 1861)	AT	13
		Andrena tarsata (Nylander 1848)	AT	13
	Apidae	Anthophora aestivalis (Panzer 1801)	AT	13*
		Anthophora borealis (Morawitz 1864)	RU	10*, 12*
		A. mellifera (L. 1758)	FR, FR, NL, GB, AT, AT	2*, 3*, 5*, 9*, 13*, 14*
		B. lapidarius (L. 1758)	FR, GB, FR, GB, AT	3, 7, 8, 9, 13
		B. lucorum (L. 1761)	GB	7, 9
		B. pascuorum (Scopoli 1763)	FR, GB	3*, 9
		B. pratorum (L. 1761)	GB	1*, 9
		B. terrestris (L. 1758)	GB, AT	9*, 13
		Bombus vestalis (Geoffroy 1785)	GB	1*
		Ceratina callosa (F. 1794)	AT	13
		Ceratina cyanea (Kirby 1802)	AT	13
		Eucera sp.	FR	3, 4
		N. fabriciana (L. 1767)	BE	11*
		N. ruficornis (L. 1758)	BE	11*
		N. succincta (Panzer 1798)	BE	11*
		Tetralonia sp.	FR	3
	Formicidae	Myrmica ruginodis (Nylander 1846)	GB	7
	Halictidae	H. eurygnathus (Bluethgen 1931)	AT	13*, 14*
		Halictus simplex (Bluethgen 1923)	AT	13*, 14*
		Lasioglossum albipes (F. 1781)	AT	13
		Lasioglossum calceatum (Scopoli 1763)	AT	13
		Lasioglossum morio (F. 1793)	AT	13
		Lasioglossum nigripes (Lepeletier 1841)	AT	13
		Lasioglossum pauxillum (Schenck 1853)	AT	13
		Lasioglossum xanthopum (Kirby 1802)	AT	13
		Sphecodes ferruginatus (von Hagens 1882)	AT	14*
	Megachilidae	H. adunca (Panzer 1798)	BE	11*
		Osmia aurulenta (Panzer 1799)	AT	13*, 14*
		O. bicolor (Schrank 1781)	AT	13*
		Osmia rufa (L. 1758)	BE	11*
	Sphecidae	Ammophila sabulosa (L. 1758)	GB	9
		Sphex funerarius (Gussakovskij 1934)	GB	9
Order	Family	Taxon	Country (ISO code)	Reference
	Vespidae	Vespula vulgaris (L. 1758)	GB	9
Lepidoptera	Geometridae	Pseudopanthera macularia (L. 1758)	AT	13
	Hesperiidae	C. palaemon (Pallas 1771)	DE, AT	6, 13
	Lycaenidae	P. icarus (Rottemburg 1775)	DE	6
	Nymphalidae	A. io (L. 1758)	GB	9
		Boloria euphrosyne (L. 1758)	DE	6
	Papilionidae	Parnassius mnemosyne (L. 1758)	RU	12
	Pieridae	A. cardamines (L. 1758)	DE	6
		Colias hyale (L. 1758)	DE	6
		Gonepteryx rhamni (L. 1758)	DE	6
		Leptidea sinapis (L. 1758)	DE	6
		Pieris brassicae (L. 1758)	DE, GB, GB	6, 7, 9
		P. napi (L. 1758)	DE, GB, GB, AT	6, 7, 9, 13
		Pieris rapae (L. 1758)	GB	9

1,  Bateman and Rudall (2014); 2, Berger (2003); 3, Berger (2004); 4, Berger (2010); 5, Claessens and Kleynen (2011); 6, Ebert and Rennwald (1993); 7, Farrell (1985); 8, Godfery (1933); 9, Harding (1996); 10, Krivosheev et al. (2009); 11, Petit (1998); 12, Shamigulova (2012); 13, Vöth (1987); 14, Vöth (1999); 15, Vöth (2003). *potential pollinator sensu Schatz et al. (2005).

Several factors can explain why an insect is only a visitor and not a pollinator of a plant species (Pellmyr 2002; Shivanna and Tandon 2014). In the case of O. militaris, the behavior and or the morphology of visitors on the orchid flowers normally prevents these insects from picking up pollinia. To do so, an insect first has to press the spur opening. Then the bursicle’s membrane breaks and the viscidia are released and glued to the insect’s head (Claessens and Kleynen 2011). As long as individuals from a given insect species are not documented as having collected at least one pollinium, they should only be considered as ineffective visitors. This categorization could be temporary or permanent; for example, some insect groups, like orthopterans (but see Micheneau et al. 2010), usually are never considered as pollinators. Caught insects bearing pollinia are potential pollinators and could be either incidental (transporting pollinia by chance) or confirmed ones. A potential pollinator becomes confirmed when the observer has seen the removal of one or more pollinia from a flower and the deposit of pollen on the stigma of another flower from the same species (Schatz et al. 2005; but see Berger (2006) who details the limits of this definition).

More than 90 insect species (half are hymenopterans) are known to visit the flowers of O. militaris (Table 3). Not all these flower visitors are effective at pollen delivery. One-fourth (almost exclusively hymenopterans, with the exception of the coleopteran Tropinota hirta Poda) have to be considered as potential pollinators. Among them, some species were caught in this study but bore no pollinia (Apidae: Bombus pascuorum Scopoli, B. pratorum (L.), and B. terrestris (L.); Megachilidae: Osmia bicolor Schrank). The low sample size per species (in average, two specimens) could explain why we did not observe pollinia on these species. Moreover, a few potential pollinators from Table 3 are not present or extremely rare in Belgium: Andrena enslinella Stoeckhert (Andrenidae: Rasmont et al. 2013), Anthophora aestivalis Panzer, A. borealis Morawitz (Apidae: Rasmont 2014), and Halictus eurygnathus Bluethgen (Halictidae: Pauly et al. 2016); their role in O. militaris pollination is probably very low in this country. In addition, in the study site “Oseraie de Lanaye” (Table 1), Petit (1998) found specimens from three species of Nomada Scopoli (Apidae: Nomada fabriciana (L.), Nomada ruficornis (L.), and Nomada succincta Panzer) and Hoplitis adunca Panzer (Megachilidae) with O. militaris pollinia. However, none of these species was observed in our study. Finally, as mentioned earlier, A. mellifera was the only species bearing pollinia in common in this study (Table 2) and in previous ones (Table 3).

According to Vöth (1987), confirmed pollinators (at least in Lower Austria) are short-tongued hymenopterans from the genera Andrena (F.) (Andrenidae: A. enslinella and A. hattorfiana F.) and Halictus Latreille (Halictidae: H. eurygnathus and H. simplex Bluethgen). A. hattorfiana (Rasmont et al. 2013), H. eurygnathus, and H. simplex (Pauly et al. 2016) are found in Belgium, but these species were not observed on O. militaris during the censuses. As mentioned in the previous paragraph, A. enslinella is absent from the Belgian territory (Rasmont et al. 2013). Vöth’s conclusion is based upon the presence of pollinia on insects and the similarity between spur length (and lip length) and proboscis length (and body length). According to this interpretation, pollinia adhesion and subsequent transport would be unadapted to the morphology of long-tongued bees (in Vöth’s study, Apidae: A. mellifera, B. lapidarius, and B. terrestris) because they reach the end of the spur before the viscidium adheres to the insect (Vöth 1987, van der Cingel 1995). However, real pollination events are not described by Vöth (1987) and his conclusion is thus mostly putative. Bateman and Rudall (2014) hypothesized that bumblebees (B. pratorum and B. vestalis Geoffroy) could be confirmed pollinators in a British population. In Vöth’s study, the two mentioned bumblebee species (B. lapidarius and B. terrestris) weren’t pollinators because no specimen bore pollinia (but see Harding 1996) and their proboscis was longer than the spur. B. lapidarius and B. terrestris are long-tongued bees but this is not the case of the short-tongued B. pratorum and B. vestalis (Goulson 2010, Bateman and Rudall 2014). The results of Vöth (1987) and those of Bateman and Rudall (2014) suggest the importance of short-tongued (non-)Apidae bees as confirmed pollinators of O. militaris. In general, long-tongued bees, like A. mellifera and B. lapidarius in this study, probably remove pollinia accidentally during their visits. In certain cases, the number of removing events can be high but specimens bearing numerous pollinia should be rare, although we mostly observed honeybees bearing many pollinia. The function of short-tongued bees should be examined in more details through direct observation of pollination events and study of pollination efficiency. The status of incidental pollinators attributed to long-tongued bees could also be confirmed through experiments involving A. mellifera or B. terrestris (e.g. hives provided by Biobest, Westerlo, Belgium). Direct field observations and captures also remain of the greatest importance to determine the role of insect visitors and study the visitor spectrum of a plant species.