Quo vadis Aethina tumida? Biology and control of small hive beetles

Wandering larvae can migrate long distance to reach the soil (>200 m, Sanford 1998). Pettis and Shimanuki (2000) showed that when soil surrounding hives in Florida was sandy and loose in consistency, most wandering larvae were found within 1 m of the hive and at a depth of no more than 10–20 cm. Laboratory data on more dense clay soils indicate that wandering larvae would most likely try and reach less compact soil in order to facilitate burrowing and pupation and increase pupation success (Ellis et al. 2004d; Meikle and Diaz 2012). SHB populations successfully developed in various types of soil, and vertical movement of larvae in the soil was not influenced by soil type (de Guzman et al. 2009).

In soil, the duration of SHB pupation can range between 2 weeks and 2 months (Lundie 1940; Schmolke 1974; Neumann et al. 2001a; Mürrle and Neumann 2004; Haque and Levot 2005; Meikle and Patt 2011; Meikle and Diaz 2012; Bernier et al. 2014) and abiotic factors are known to play a key role. Indeed, it is well known that both duration and success of SHB pupation are governed by temperature, soil texture and humidity (Lundie 1940; Schmolke 1974; Neumann et al. 2001a; Ellis et al. 2004d; Mürrle and Neumann 2004; Haque and Levot 2005; de Guzman and Frake 2007; de Guzman et al. 2009; Meikle and Patt 2011; Meikle and Diaz 2012; Bernier et al 2014). Sandy, moist soil and warm temperature (24.6 ± 1.3 °C) seem to provide good conditions with only 23 days needed to complete pupation and only 8.5 % mortality (Ellis et al. 2004d). There may be as many as six generations per year under moderate US and South African climatic conditions (cf. Neumann and Elzen 2004), but this could potentially be even higher under tropical conditions. Indeed, de Guzman and Frake (2007) showed that almost 16 complete life cycles could be achieved in a year, with constant 34 °C.

An entire life cycle of a SHB can be achieved in a single Eppendorf tube (Neumann et al. 2013). Therefore, intraspecific competition for space is rather unlikely in the soil, unless severe space limitations may interfere with the construction of SHB pupation chambers (e.g. in laboratory rearing, Mürrle and Neumann 2004). However, subterranean mating of SHB adults is most likely because they reach sexual maturity after 7 days (cf. Neumann and Elzen 2004) and often remain jointly underneath the soil surface without emerging for up to 35 days (Mürrle and Neumann 2004). What factors trigger adult SHB emergence from the soil is currently unknown, but light rain maybe involved (simulated with water and a manual pump sprayer in the laboratory, P. Neumann, unpublished data). The impact of salinity and pH value on SHB pupation are currently unknown.

The ability of entomopathogenic fungi to kill SHBs was already reported by Lundie (1940). Since, various fungi were isolated from and/or have proven effective against SHBs (Table III; Richards et al. 2005; Cabanillas and Elzen 2006; Muerrle et al. 2006; Ellis et al. 2010). Recently, isolates of both Metarhizium and Beauveria showed good efficacy against SHBs in laboratory and field assays (Leemon and McMahon 2009; Leemon 2012), where the Metarhizium isolates showed best results against larvae (three isolates were able to kill more than 70 % of larvae in 1 week), while the Beauveria isolates performed best against adult beetles (99–100 % adult SHB mortality in 2 weeks). Susceptibility of SHBs was also shown to entomopathogenic nematodes where several Heterorhabditis and Steinernema strains have provided adequate SHB control with 88–100 % mortality (Table III; de Guzman et al. 2009; Cabanillas and Elzen 2006; Ellis et al. 2010; Shapiro-Ilan et al. 2010; Cuthbertson et al. 2012). Three different Bacillus thuringiensis Berliner strains (B. thuringiensis var. aizawai, B. thuringiensis var. kurstaki and B. thuringiensis var. San Diego tenebrionis) showed no effect on reproductive success of SHBs (Buchholz et al. 2006). Ants are generally considered a potential biological control agent, and the ant Pheidole megacephala was identified as a key predator of larvae in Kenya (Torto et al. 2010a). Furthermore, a protozoan pathogen was discovered in the Malpighian tubules of adult SHBs (Wright and Steinkraus 2013), and Strauss et al. (2010) reported the storage mite Caloglyphus hughesi on laboratory-reared SHBs in South Africa.

The impact of pathogens and predators may be less severe in the rather protected environment of social bee colonies due to host defence and sanitation (e.g. propolis, Evans and Spivak 2010) compared to food sources outside of colonies, but there are no comparative data in this regard. Likewise, the impact of natural enemies has never been compared between endemic and invasive populations. Therefore, release from natural enemies cannot be excluded as one key factor driving SHB invasion success.

Since SHBs are the only known species to be able to induce trophallactic feeding in honeybees (Ellis et al. 2002d), there is most likely no competitor with adult SHBs in colonies. In sharp contrast to A. tumida, the bee louse, Braula coeca, is not able to induce trophallaxis in honeybees. Instead, the bee louse takes advantage of two bees feeding each other (Morse and Nowogrodzki 1990). Sitting on the head or abdomen of a worker or the queen, the louse quickly moves forward and steals food during the food exchange between the two bees (Morse and Nowogrodzki 1990). When colonies are less well defended or honeybees abscond, SHBs can take advantage of the then unprotected food (brood, honey and pollen, see above). Such food will also be consumed by other species such as ants, which can therefore be both natural enemies (Arbogast et al. 2010) as well as competitors for reproduction (Spiewok et al. 2006b). Likewise, wax moths and SHBs can compete for the same resource and may negatively influence the reproductive success of each other (Spiewok et al. 2006b). On fruits and other food resources outside of colonies, Drosophilidae and other Nitidulidae would compete with SHBs (Buchholz et al. 2008).

In conclusion, our knowledge of natural enemies and competitors is still fairly limited, e.g. the role of tropical subterranean army ants for SHB pupation success (Kronauer 2009) and of nocturnal predators (Williams-Guillén et al. 2008) of flying adult SHBs is unknown.

A. tumida has been shown to act as a vector for Paenibacillus larvae, the causative agent of American foulbrood (AFB; de Graaf et al. 2013), with both adults and larvae becoming infected with spores when exposed to honeybee brood combs with clinical symptoms (Schäfer et al. 2010b). The subsequent field test, where honeybee colonies were infested with contaminated adult beetles, resulted in higher numbers of P. larvae spores in adult workers and honey (Schäfer et al. 2010b). Adult SHBs have also shown the potential to act as vectors for honeybee viruses (Eyer et al. 2009a, b). It has been demonstrated that adult SHBs can be infected with deformed wing virus (Eyer et al. 2009a) and sacbrood virus (Eyer et al. 2009b) via food-borne transmission.

During larval reproduction, combs become covered in a slimy material that is part honey and part larval in origin (Lundie 1940). From this “slime” and adult beetles, researchers isolated a yeast that appears to function as an attractant to SHB adults (Benda et al. 2008). The yeast, K. ohmeri, produces substances that are similar to bee pheromones, and a dough containing the yeast has been tried in traps to increase attractiveness to adult beetles (Arbogast et al. 2007; Torto et al. 2007a; Hayes et al. 2015). K. ohmeri has also been found in both managed and wild colonies of bumblebees (Graham et al 2011b). Organic acids used for mite control were shown to inhibit the growth of K ohmeri (Schäfer et al. 2009). While initially very promising as an attractant (Torto et al. 2005), additional research is needed to fully realize the potential of this yeast in the management of SHBs.

Besides control, it appears as if there is more of a relationship between K. ohmeri and the SHB than just the release of volatiles that are attractive to SHBs (Torto et al. 2007c). It may well be that there is a symbiotic relationship between the two. So contaminating the honey with K. ohmeri allows the yeast to grow by fermenting honey and may thereby provide some protein to supplement the SHB larval diet when feeding predominantly on honey. Moreover, there might be a human health risk with this yeast as well (Santino et al. 2013).

Analogous to bark beetles (Byers 1989), pheromone-mediated aggregations of adult SHBs might be adaptive to overcome host defence (Neumann and Elzen 2004). Mustafa et al. (2006) showed SHB aggregation in laboratory plastic cages, and field studies have shown that SHB distribution among honeybee colonies at an apiary is different from a random distribution (Neumann and Elzen 2004; Spiewok et al. 2007). In a single A. m. scutellata colony, 491 adult SHBs were found, while all other colonies at the same apiary showed low infestation levels (Neumann and Elzen 2004). The maximum numbers of adult SHBs in European honeybee colonies so far recorded in functional colonies in the invasive ranges were 1071 in Umatilla, Florida (USA) and 2029 (P. Neumann, unpublished data) and 3027 (N. Annand, unpublished data) in Richmond, NSW, Australia. However, colony phenotypes (number of bees, amount of brood or stores) do not significantly influence infestation levels with adult SHBs (Spiewok et al. 2007), thereby indicating that cues other than host colony size and food stores are responsible for their attractiveness. Aggregation pheromones have been described for a variety of Nitidulidae species and are widely used as control agents (Petroski et al. 1994; James et al. 2000). We consider it very likely that a similar pheromone plays a role for long-range host finding and aggregations of SHBs. Observations that males tend to infest before females (Neumann and Elzen 2004) indicate that the aggregation pheromone might be male produced as in Carpophilus obsoletus and is attractive to both sexes (Petroski et al. 1994), but no sex-specific differences in SHB brain structure were found (Kollmann et al. 2015). Moreover, the sex ratios did not significantly differ between recently infested apiaries and longer infested ones in another study (Spiewok and Neumann 2012). Synergistic effects between food odours and aggregation pheromones for attracting SHB might also play a role as shown for Carpophilus lugubris (Lin et al. 1992). More research is needed to identify and evaluate the potential impact of different compounds such as aggregation pheromones, food volatiles or any synergism between pheromone and food volatiles on the short- and long-range dispersal and host finding of SHBs.

Most beetles captured in baited traps near La Crosse, FL, USA were captured during the spring and summer months, with peak captures during May and June (Arbogast et al. 2010). In St. Gabriel, Louisiana, USA, SHB populations varied throughout the year, with peak infestations observed in September and November (de Guzman et al. 2010). Similarly, SHB populations in Georgia seem also to rise by July/August and reach their peak by September/October (Berry 2009). In Louisiana, SHB abundance was significantly correlated with the proportion of hot days, but not with the proportions of cool, dry or humid days or the percentage of days with rainfall (de Guzman et al. 2010). On the other hand, trap captures in Kenya indicated that SHB was present in honeybee colonies in low numbers all year round, but it was most abundant during the rainy season and with over 80 % trapped during this period (Torto et al. 2010a). Likewise, traps installed in Kenya in front of infested hives over an entire year showed that more wandering larvae were leaving infested colonies associated with the “kusi monsoon” (Arbogast et al. 2012). In contrast, in Nigeria, both adult and larval SHBs were more often found inside colonies (N = 437) during the dry season (28,057 adults, 5210 larvae) compared to the wet season (7869 adults, 831 larvae; Lawal and Banjo 2008).

In South Africa, mated adult SHB females provided in the laboratory with adequate food did not readily oviposit in July/August unless incubated at 30 °C for a week (P. Neumann, unpublished data). Similarly, Australian beekeepers reported almost no damage by SHB larvae in June/July/August (Rhodes and McCorkell 2007). In Australia, SHB populations in hives followed a cyclical pattern, which peaked in late autumn then declined through winter to a minimum in late spring (Annand 2011b). Most SHB movements in NSW, Australia occurred in April and May; however, a major spike was also observed in October (Annand 2011b). Differences between years have also been reported, e.g. SHB numbers in Gauteng, South Africa were considerably higher in winter 2011 compared to winter 2010 (Strauss et al. 2013).

In conclusion, several observations suggest considerable seasonal variation in SHB reproduction. Therefore, it appears as if season might also play an important role, e.g. by triggering host colony migration (Hepburn and Radloff 1998), fostering SHB reproduction (e.g. pupation) and inducing high SHB mortality over winter (Schäfer et al. 2011), with subsequent consequences of higher/lower infestation levels (Spiewok et al. 2007).

Within the literature, several different definitions have been put forward ranging from scavengers (Hepburn and Radloff 1998) over parasites (Neumann and Elzen 2004) to symbionts (Ellis and Hepburn 2006). Data have emerged that SHB actually fit several ecological definitions. They can feed on dead animals (scavenger; honeybees; conspecifics = cannibalism (Neumann et al. 2001b); beef schnitzel (Buchholz et al. 2008)) and live animals (predator; Pirk and Neumann 2013) as well as on fruits (phytophagous (Keller 2002; Ellis et al. 2002c; Buchholz et al. 2008; Arbogast et al. 2009a, 2009b, 2010). Moreover, they are able to steal food from their hosts via trophallactic mimicry (kleptoparasite; Ellis et al. 2002d; Ellis 2005c; Neumann et al. 2015). In conclusion, the authors agree with the line of thoughts initially developed by Arbogast et al. (2009b): A. tumida can be considered as an ecological generalist.

The two new SHB records from West Africa (Benin and Burkina Faso, Figure 3) are well in line with previous records from sub-Saharan Africa (El-Niweiri et al. 2008; Neumann and Ellis 2008). In Madagascar (Figure 3), A. tumida appears to be widespread and well established since it was found in all honeybee colonies of three districts from the “Highlands” and on the eastern coast (Rasolofoarivao et al. 2013). Since there are no reports of any introductions or of any serious damage to local colonies, it seems as if the SHB is endemic to Madagascar.

In its native range, the SHB is usually only a minor pest (Pirk and Yusuf 2015), because successful reproduction appears most often in weak, stressed honeybee colonies or in recently abandoned ones and is far less common in strong colonies (Lundie 1940; Schmolke 1974; Hepburn and Radloff 1998; Neumann and Elzen 2004). Strong African honeybee colonies can usually prevent or postpone successful beetle reproduction (Neumann and Elzen 2004). In such colonies, SHBs usually have to wait until absconding (non-reproductive swarming) for various reasons, possibly including heavy SHB infestations (Hepburn and Radloff 1998; Neumann and Elzen 2004). However, there might be regional differences within its endemic range. Indeed, while SHBs appear to be rare in Kenya (Torto et al. 2010a) and in Uganda (Kugonza et al. 2009), it seems to be the most common honeybee pest in Nigeria (40.5 % of hives infested with adults, 9.6 % with larvae; Lawal and Banjo 2007), where a total of 28,057 adult SHB were sampled from 437 colonies (Lawal and Banjo 2008). Accordingly, the SHB seems to be a bigger problem for beekeepers in West Africa compared to East Africa or the Horn of Africa (Mutsaers 2006; Akinwande et al. 2013). During South African winter, SHB pupation success is rather unlikely, at least in the Eastern and Western Cape provinces and in the Highveld.

Knowledge about the likelihood of different invasion pathways is essential to limit the further spread of the SHB via adequate legislation and border control measures. SHBs can potentially reach new shores via trade associated with apiculture, including package bees, queen cages, whole honeybee colonies as well as apicultural tools, combs and processed or non-pressed wax. Imports of Bombus spp. and stingless bees might also be an alternative route as well as import of fruits and soil (e.g. for potted plants with soil, FERA 2014; Chauzat et al. 2015). Introductions near major harbours (Charleston, USA, Hood 2000; Gioia Tauro, Italy, Mutinelli et al. 2014) or on islands (Jamaica (FERA 2010), Hawaii (Connor 2011a, b), Cuba (Peña et al. 2014), Philippines (Brion 2015)) indicate that trade via ships may play a role (possibly via fruit import, H. Smith, personal communication).

In almost all cases, migratory beekeeping and or active dispersal of SHB might be involved (Québec, Canada (Evans et al. 2003, 2008; Giovenazzo and Boucher 2010); Coahuila, Mexico (Del Valle Molina 2007)). In Australia, hives that moved from New South Wales into Victoria in 2003 and in Victoria in 2005 as well as from Kimberley to Perth in 2008 were found to be carrying SHB (N. Annand, R. Spooner-Hart, personal communication). In May 2011, adult SHBs were found in a conservation park on the South Australia–Victoria border on illegally moved hives and, in May 2012, SHB beetles and larvae were found in hives and supplementary pollen in close-by Naracoorte, South Australia (N. Annand, R. Spooner-Hart, personal communication). National and international apicultural trade seems to play a key role too (Alberta, Canada, Australian package bees, Lounsberry et al. 2010; Itay-Al-Baroud, Egypt, import of colonies from Ethiopia? A. M. Mostafa, personal communication). Survival of adult and/or immature SHB life stages obviously depends on both storage conditions during transport and food availability. Moreover, inspections prior to and after trade (e.g. border controls) should also be considered. However, rather unlikely routes such as processed wax (Manitoba, Canada, cf. Neumann and Elzen 2004) and queen cages (Portugal 2004: Murilhas 2004; Northern Territory, Australia 2010: R. Spooner-Hart, personal communication) have also been reported, although beetle survival in the former may be compromised and inspections prior to export should be rather efficient in the latter case.

Genetic tools enable to trace back the origin of invasive populations, which can be helpful to better mitigate future introductions. SHB mt-DNA sequence analyses of SHBs from the USA and South Africa indicate that the populations on both continents belong to a single species, although it is not clear whether a single or multiple introductions occurred (Evans et al. 2000, 2003). SHBs in Australia have a different origin than beetles in North America and the initial North American beetles shared the same source (Evans et al. 2008; Lounsberry et al. 2010). The outbreaks in Quebec, Canada appear to originate from the USA (Evans et al. 2003, 2008), and all have been found close to the US border. The SHBs confirmed in Alberta (Canada) in 2006 were a new introduction via Australian package bees (Lounsberry et al. 2010).

Taken together, the current evidence clearly shows that international apicultural trade and/or migratory beekeeping were involved in all invasion cases, where clear signs could be found. So far, there is no evidence whatsoever that imports of soil and fruits have resulted in any invasion of this pest. From an economic point of view, it appears therefore prudent to focus on legislation and even stricter control of international bee trade to limit the future global spread of SHBs.

Diagnosis of A. tumida in the laboratory is either based on morphological criteria (Neumann et al. 2013) or by molecular identification (Evans et al. 2000, 2008; Ward et al. 2007; Lounsberry et al. 2010). The life stages of A. tumida are shown in Figure 1. Adult SHBs are 5–7 mm long and 2.5–3.5 mm wide. Head, thorax and abdomen are well separated. Key features of A. tumida are its “club-shaped” antennae and its elytra, which are smaller than the abdomen, so the end of the abdomen is exposed. Special attention should be given to body coloration, because freshly emerged adult A. tumida look very similar to C. luteus (Neumann and Ritter 2004), thereby bearing the possibility of false positive results. SHB larvae grow to ~1-cm long, are creamy white and can easily be distinguished from the wax moth larvae (Galleria mellonella) due to the presence of three pairs of long forelegs, a row of spines on the dorsal side of each body segment and two large spines protruding from the rear and in the field the absence of webbing in the combs.

Several molecular techniques to diagnose SHBs have been described (Table II; Evans et al. 2000, 2008; Ward et al. 2007; Lounsberry et al. 2010). Real-time PCR in conjunction with an automated DNA extraction protocol was used to screen hive debris for the presence of SHBs (Ward et al. 2007). The method was shown to be able to detect DNA from SHB eggs, larvae and adult specimens collected from Africa, Australia and North America and to successfully detect SHB DNA extracted from spiked and naturally infested debris. Evans et al. (2008) described 15 microsatellite loci, which are polymorphic in the native as well as in the introduced ranges of this species, showing from two to 22 alleles. Such polymorphic microsatellite loci for SHB are necessary to map the increase in distributions, but they are also useful to explore the SHB mating system and local gene flow patterns. These loci have also proven useful in mapping the movement of SHBs in the Americas (Lounsberry et al. 2010). Especially if only eggs or larvae are available, the molecular techniques are essential as these live stages are impossible to be diagnosed visually. Another advantage is that samples can be collected over time and as long as they stored appropriate molecular analyses are always possible. For example, screening of hive debris (Ward et al. 2007) was used in a surveillance study in Spain to evaluate the possible presence of SHB in the country (Cepero et al. 2014). However, our limited knowledge of the A. tumida genome currently constrains the full diagnostic power of molecular tools. Indeed, limited resolution and false negative results are likely (e.g. using conventional primers; Evans et al. 2008). Therefore, it appears crucial to get the A. tumida genome sequenced to foster the usage of molecular tools for both research and diagnostics.

Monitoring of sentinel apiaries that are located in zones at risk for SHB introduction (e.g. seaport or airport) is recommended to detect SHB infestation at an early stage in order to eradicate it (Chauzat et al. 2015), and several countries have initiated this method (Australia (Annand 2011b), Italy (Mutinelli et al. 2014), Switzerland (Elena di Labio, personal communication)). The most rudimentary and time-consuming method to estimate the infestation level of a colony in the field is screening of the entire colony (Spiewok et al. 2007; Neumann and Hoffmann 2008). However, this method is also highly accurate, if done by specially trained personnel and therefore the method of choice for new introductions with low infestation rates. A very fast, cheap and easy quantitative diagnosis for SHB in the field is the use of diagnostic strips made of corrugated plastic (Schäfer et al. 2008, 2010a), which are placed on the bottom board via the entrance, without the need to open or manipulate the colonies. A test at five apiaries in Richmond, NSW, Australia showed that the number of SHBs found in the transparent strips correlated significantly with the total number of SHBs in the hives (detected by screening of the entire colony), and the average efficacy was 34.5 % (Schäfer et al. 2008). It is recommended to leave the diagnostic strips inside the colonies for at least 48 h to give the beetles a chance to find the refuges and it is not recommended to use them during cold seasons, when bees form winter clusters, because adult SHBs might leave the bottom board seeking shelter in bee clusters (Figure 3; Pettis and Shimanuki 2000).

The most accurate diagnosis method is to kill the entire colony with bees and the pest followed by a visual inspection, including dissection of each frame (Neumann and Hoffmann 2008; Neumann et al. 2013). However, this can of course not be applied systematically in the field. Nevertheless, this method seems to be “gold standard”, because its efficacy is virtually 100 % if applied correctly and can thus be used to validate other diagnostic methods. It could also be applied for wild swarms and other research purposes.

Efficient control of SHBs should not rely on a single method; it requires pest management decisions based on adequate diagnosis methods to assess which control measures will lead to maximum efficacy and simultaneously produce results with the fewest environmental impacts. Integrated pest management (IPM) should be considered (Ellis 2005a, b; Hood 2011), using a combination of all available control methods (mechanical, biological and chemical) in a responsible matter. Manual removal of SHBs (screening of entire colonies, see above) could also be regarded as a control option, but it is very labour-intensive. Different kinds of SHB traps can be installed in or outside of the colonies that should be checked regularly during apiary visits. The functional principle of most SHB traps is creating a refuge or space, which is large enough for SHBs and too small for bees to pass through at once to enter the trap which contains a reservoir for the killing product (veterinary medicine, oil, diatomaceous earth, etc.). Another way of trapping is currently in use in Australia, where kitchen cleaning wipes, placed on top of the frames, are shred into fibres by the bees and the beetles become trapped in the fibrous material (Zacchetti 2015).

Among the products that are used to control V. destructor mites, CheckMite+^TM strips were shown to also have broad toxicity against SHBs, killing both larvae and adults, when placed underneath corrugated cardboard or plastic sheets on the bottom board (Ellis and Delaplane 2007; Neumann and Hoffmann 2008). However, when wandering larvae were exposed to CheckMite+^TM for 24 h, they still burrowed successfully, which will minimize the efficacy to control SHB larvae in the field (Ellis and Delaplane 2007). Another easy to install bottom board trap (Beetle Barn^TM with CheckMite+^TM) also showed high efficacy against adult SHBs and the position of the trap, whether in the front or the back of the bottom board, had no effect on the number of captured SHBs (Bernier et al. 2015). In Australia, a refuge trap was developed comprising of two-piece rigid plastic shells encasing a fipronil-treated corrugated cardboard (Apithor^TM) insert (Levot 2008; Levot and Somerville 2012). In a 36-day-long field trial conducted in a beetle-infested apiary at Richmond, NSW, Australia, live adult beetles were eliminated from hives containing Apithor^TM while beetle numbers increased by approximately 20 % in co-located control hives The residues in honey ripened inside the test hives over 1 month while the devices were in place did not exceed 1 μg/kg, and no ill effects on the bees were observed (Levot 2012). Despite earlier observations of lively SHBs emerging from a hive, unscathed by a formic acid treatment (Amrine and Noel 2006), Schäfer et al. (2009) tested the effects of formic (60 %) and acetic acid (70 %) treatments on SHB in nucleus boxes with two honey/pollen frames each. Formic acid decreased the number of SHB larvae-infested areas on the combs, and acetic acid treatments showed higher SHB adult mortality compared to controls (Schäfer et al. 2009). However, further evaluation of evaporating formic acid in field colonies did not significantly increase mortality in SHBs, and thus neither formic nor acetic acid treatments are suitable to control SHBs (Buchholz et al. 2011).

Kanga and Somorin (2011) assessed the susceptibility of SHBs to 14 selected insecticides and four insect growth regulators in the laboratory. Their results indicated that SHBs are selectively susceptible to several classes of insecticides. Chemicals are always the first choice in controlling a new pest because they are mostly easy to use and effective, but there are reported problems, for example with the residues of coumaphos present in wax (Mullin et al. 2010), which can contaminate honey and besides the product does not seem to work well in cooler temperatures (Mostafa and Williams 2000). Moreover, SHBs may develop resistance to any chemical used for their control because of their high mobility and fecundity (De Guzman et al. 2011). In conclusion, chemicals are fraught with difficulties in the end.

A diverse range of alternatives to chemical control have been developed and tested in laboratory and field assays. There have been mixed results with alternations of the hive entrance. The use of entrance reducers had a significant effect on the average number of invading SHBs (Frake et al. 2009), whereas an upper hive entrance was not effective against SHB invasion and resulted in a reduction in bee brood, demonstrating their infeasibility (Hood 2004; Hood and Miller 2005). The use of slaked lime and diatomaceous earth was tested in the laboratory and in honeybee field colonies (Buchholz et al. 2009). In diagnostic tray traps in the field, about a third of the adults of infested colonies were caught with slaked lime and diatomaceous earth (Fossil Shield® FS 90.0s) and caused 100 % adult mortality inside the traps, where more than 50 % died within 48 h (Buchholz et al. 2009). Cribb et al. (2013) performed laboratory tests with diatomaceous earth as killing agent inside top frame bar traps (Beetle Eater^TM) and reported 100 % SHB mortality compared to 8.6 % mortality in controls without diatomaceous earth. In these and similar traps (Beetle Blaster^TM), the reservoirs are usually filled with oil, in which entering SHBs eventually drown (Bernier et al. 2015) which is also the case in many bottom board traps (e.g. West Trap^TM, Freeman Beetle Trap^TM). A bait, consisting of pollen dough (mixture of pollen and honey) that was conditioned either by the feeding of adult SHBs or by inoculation with the yeast K. ohmeri (Benda et al. 2008), has been used successfully in bottom board traps in the USA and Kenya to monitor SHB populations (Torto et al. 2007b, 2010a).

All these traps only provide adequate control when the position of the trap in the hive is adapted to the environmental situation. This is important, as the bottom board is a suitable location for SHB traps during the hot season, but during the winter, SHB will be found close to or among heat-producing clustering bees (Schäfer et al. 2011). At present, in-hive control devices appear to be most appropriate for removal of SHBs (Annand 2011b). Traps that capture wandering larvae at a colony scale (Arbogast et al. 2012) are recommended primarily as research tools (Neumann et al. 2013), as damage has already occurred when wandering larvae leave the colonies.

When reproduction has occurred, SHB larvae and pupae may be present in the soil around the hive. Since the pyrethroid permethrin (in the USA sold as GardStar®) applied around colonies achieved some success at killing beetle larvae and pupae (cf. Neumann and Elzen 2004), pyrethroid ground drenching is widely used for SHB control (e.g. cypermethrin and tetramethrin were used in Italy; Mutinelli et al. 2014), but there are also problems with the use of ground drenches. They may protect individual hives on site but do little to prevent the spread of SHBs (Hood 2000), and they need to be applied around the hives in all directions (about 180–360 cm) to maximize their efficacy (Pettis and Shimanuki 2000). Moreover, during the reproductive season of SHBs, larvae are continuously leaving the hives to pupate in the ground (Torto et al. 2010a), which makes continuous treating necessary.

Three different light types (white bulb, black light insect light) were tested as attractants, indicating that none of these light sources are attractive to adult SHBs (Neumann and Elzen 2004). But lights have been used successfully in the honey house to attract wandering larvae, as they are positively phototactic (Somerville 2003).

Flying SHBs have been monitored with baited traps (cf. Neumann and Elzen 2004). The traps were baited with honey, pollen, brood and live adult honeybees in several compositions to test their attractiveness to SHBs. But as beetles could enter and then escape, these traps were ineffective for control. Later, Arbogast et al. (2007) used refined traps that lure and capture SHBs using a bait consisting of pollen dough (mixture of pollen and honey) conditioned by allowing male SHBs to feed on it for 3 days were tested successfully in the field. In their study, they also found that significantly more SHBs were captured if the traps were located in the shade compared to sunny locations (Arbogast et al. 2007). Torto et al. (2007a) tested the attractiveness of the same diet inoculated with the yeast K. ohmeri, isolated from the beetle. Further tests with K. ohmeri-inoculated pollen dough-baited traps confirmed the positive influence of shade and showed a negative correlation with distance to bee hives on the frequency of capture (Arbogast et al. 2009a). To investigate the response of adult A. tumida to visual stimuli, de Guzman et al. (2011) tested the influence of black and white colour and the height on pole trap efficiency. The average catch in white traps was significantly higher than that of black traps, probably because white is more reflective than black, and traps positioned at the same height as colony entrances showed the highest catch numbers in the field. As these traps generally do not catch high numbers of beetles and honeybee hives seem to lure the majority of the beetles, further improvements are necessary (de Guzman et al. 2011), e.g. capturing and killing SHBs prior to hive entry will bear obvious advantages. Using external baited lures throughout the peak movement times (autumn and spring), capturing SHBs outside of the hives when flying into apiaries may provide a safe, easy control option (Annand 2011b). Such traps would be especially useful if we aim to control SHB numbers outside of managed apiaries, e.g. to safeguard bumblebees and stingless bees as well as wild or feral honeybees. In conclusion, these traps allow continuous observations that provide a relative measure of A. tumida migration (Neumann et al. 2013), and there is still hope to develop a lure that is more attractive to SHBs than the colonies of honeybees or other bees.

Good beekeeping management is an essential tool in the fight against SHB damage (Westervelt 2005; Hood 2011). As the larval stage is most harmful to the colonies, measurements should be focused on limiting the reproductive success of A. tumida in the colony. If strong colonies are infested, the adult beetles are pursuing a “sit-and-wait” reproductive strategy (see above), as they are confined by guard bees or hiding from the bees in areas of the colony, where food for larvae is rare. Only if the situation changes and the beetles are free from captivity do they take the chance and try to lay eggs close to food. Especially if working on the hive creates a narrow space that is secured from bee aggression and pollen or brood is available. This could happen for example when there is no bee space between two brood combs. The impact of the apiary location in the shade or in the sun is unclear. While Arbogast et al. (2007, 2009a) found more SHBs in lured traps outside hives in the shade, Ellis and Delaplane (2006) found that the location in a shaded area or in an open field with full sun had no significant effect on the number of SHB entering the colonies. In Australia, large beekeeping operations around Sydney moved their operations over the Blue Mountains where conditions are cooler and less humid, which has proven the most effective SHB management strategy (N. Annand and R. Spooner-Hart, personal communication).

If honey is harvested, the SHB eggs and larvae on such combs are released from the bees and free to develop; therefore, it is recommended to extract the honey immediately (at least within 2 days) and if this is not possible cold storage or at low relative humidity should be considered as temperatures below 15 °C and relative humidity less than 34 % prevented egg survival (Annand 2011b). Hives where bees do not exhibit good hygienic behaviour (attempt to remove beetle adults or larvae) should be re-queened or replaced with bees that do (Fletcher and Cook 2005). Feeding pollen substitute patties is problematic as enhanced SHB reproduction may occur (Westervelt et al. 2001), but in winter the risk is lower, as then the conditions are unfavourable for SHBs (Hood 2009). Some beekeepers in the USA use small colonies (nucleus) with Checkmite+^TM strips as death traps for SHBs, believing that small colonies may be especially attractive to them (Jacobson 2005). However, there is no evidence for smaller or queenless colonies being more attractive (Spiewok et al. 2007; Annand 2011a). Nevertheless, it was shown that SHB levels, which are harmless to full-sized colonies, may have an impact on smaller ones (Annand 2011a; Mustafa et al. 2014). As hives containing very low bee numbers are vulnerable to SHB reproduction and subsequent damage, beekeepers need to regularly check and remove hives that are weak or declining. Removing susceptible hives will reduce opportunities for SHB reproduction and, therefore, help to minimize population expansion within the apiary (Annand 2011b).