Megastylus woelkei Humala

Andrei E. Humala, H. Marjolein Kruidhof & Jozef B. Woelke, 2017, New species of Megastylus (Hymenoptera: Ichneumonidae: Orthocentrinae) reared from larvae of Keroplatidae fungus gnats (Diptera) in a Dutch orchid greenhouse, Journal of Natural History 51 (1), pp. 82-95 : 85-93

Megastylus woelkei Humala , sp. nov.

( Figures 1 and 2 )

Diagnosis

This species differs from other known Megastylus species by the combination of the following characters: small size; fore wing length 2.3 mm; female antenna with 32 – 35 flagellomeres, first flagellomere 4.5 times as long as wide; ocular – ocellar distance 0.9 times maXimum diameter of lateral ocellus; first tergite 2.3 times as long as apically wide, spiracles at anterior 0.4 of tergite, sternite at about 0.5 of tergite length; second tergite as long as apically wide, without longitudinal striae.

From the related Holarctic Megastylus orbitator Schiødte , it can easily be distinguished by its smaller size, wider face, longer flagellum and malar space, comparatively short pubescence on the female flagellum, male face pale with a central dark spot, and the absence of pale markings along the frontal orbits in both seXes.

Remarks: one undescribed Megastylus species from Mexico, of which only males are represented in the available collections, is similar to this species in face colour pattern (with a central dark spot), but differs in yellowish inner eye orbits and pronotum, proportions and number of flagellomeres, weak occipital carina, more robust hind femur and larger size (fore wing 3.5 – 4.0 mm).

Description

Female. Body length 3.4 mm, fore wing length 2.3 mm ( Figure 1 (a)).

Head. Head as wide as high, nearly polished, sparsely punctate. Face at level of antennal sockets 0.7 times as wide as head width, its inner orbits slightly divergent ventrally ( Figure 1 (b)). Antenna long, with 33 flagellomeres, first flagellomere obliquely truncate, about 4.5 times as long as wide, second flagellomere 1.8 times as long as wide; flagellomeres from eight quadrate; flagellum densely covered with long erect setae. Frons smooth, nearly polished; margins of antennal sockets somewhat raised; face scarcely punctate and covered with sparse long hairs. Clypeus strongly convex, about 2.0 times as wide as high, its apical margin truncate. Malar space long, 2.8 times basal width of mandible, 0.35 times as long as face width, with distinct subocular sulcus as coriaceous strip between eye and base of mandible ( Figure 1 (c)); anterior tentorial pits large and open. Mandible slender, strongly tapered and turned inwards, small lower tooth much shorter than upper tooth; maXillary palps long. Temples moderately long, 0.5 times transverse diameter of compound eye; occipital carina complete. Ocelli of moderate size, ocular – ocellar distance 0.9 times maXimum diameter of lateral ocellus, postocellar line 0.8 times as long as maXimum diameter of lateral ocellus ( Figure 1 (d)).

Mesosoma. Mesosoma coriaceous, 1.8 times as long as high; epomia lacking. Epicnemial carina well developed, distant from front edge of mesopleuron and ending at mid-height ( Figure 1 (c)). Mesoscutum convex, with sparse small punctures bearing long setae; notaulus shallow and indistinct. Scutellum with lateral longitudinal carinae reaching its posterior end. Sternaulus weak and indistinct. Propodeum with lateral longitudinal carina present, two transverse carinae weakly developed, and other carinae reduced ( Figure 1 (e)). Propodeal spiracle circular, separated from lateral longitudinal carina by one diameter of spiracle. Fore wing with cu-a slightly proXimal to base of Rs&M by 0.2 times length of cu-a; vein 2 m-cu with two bullae, areolet absent, hind wing with cu-a slightly reclivous, Cu1 depigmented, closer to 1A than to M ( Figure 1 (f)). Legs long; fore coXa with transverse basal carina ( Figure 1 (c)); hind coXa matt, hind femur 5.3 times as long as wide, 0.8 times as long as hind tibia; hind basitarsus 0.4 times as long as hind tibia; hind tibia somewhat banded in apical half with depression on its dorsal surface; tibial spurs straight and slender, tarsal claws slender and weakly curved.

Metasoma. First tergite of metasoma slightly constricted behind spiracles, 2.3 times as long as apically wide, coriaceous; glymma lacking; spiracles at anterior 0.4 of tergite ( Figure 1 (e)); dorsal and lateral longitudinal carinae lacking; first sternite fused to tergite, about 0.5 times as long as tergite length, in lateral view forming weakly rounded lobe; tergite II coriaceous, as long as apically wide; thyridium small, close to anterior margin. Remaining tergites more finely sculptured towards apeX; metasoma weakly compressed to apeX; tergites covered with sparse long hairs. Sternites II – V with sparse dark setiferous punctures, sclerites fuscous. Ovipositor straight and slender, its sheath as long as hind femur depth.

Colour. Fuscous. Mandibles, eXcept for reddish teeth, palps and tegula pale. Face brown, with light marks under antennal sockets. Scape and pedicel ventrally, first flagellomere, apical half of clypeus yellowish brown, rest of flagellum fuscous. Pronotum anteriorly and in hind corners, lower mesopleuron and mesepimeron reddish brown. Metasoma fuscous, tergite II brown with yellowish thyridium and apical margin; tergite III yellowish, more fuscous laterally; tergite IV with light brown medial spot anteriorly. Fore and mid legs yellowish; hind legs light brown. Hind tibia infuscate dorsally in apical 0.4. Wings hyaline, pterostigma and veins brown.

Male. Similar to female. Body length 3.0 mm, fore wing 2.4 mm ( Figure 2 (a)).

Head. Head width 1.1 times its height; head matt, sparsely punctate. Face at level of antennal sockets 0.55 times as wide as head width ( Figure 2 (b – d)). Antenna long, eXceeds body length, with 31 flagellomeres, first flagellomere about 5.7 times as long as wide, second flagellomere 2.8 times as long as wide; all flagellomeres longer than wide. Frons nearly polished, almost impunctate; face smooth, scarcely punctate. Clypeus convex, about 2.1 times as wide as high, apical margin nearly truncate. Malar space 2.5 times as long as basal width of mandible. Ocelli of moderate size, ocular – ocellar distance as long as maXimum diameter of lateral ocellus, postocellar distance 0.9 times as long as maXimum diameter of lateral ocellus ( Figure 2 (e)).

Mesosoma. Mesosoma length 1.7 times height. Legs long, slenderer than in females; hind coXa coriaceous, hind femur 6.0 times as long as wide, 0.7 times as long as hind tibia; hind basitarsus 0.45 times as long as hind tibia.

Metasoma. First tergite of metasoma slender, somewhat constricted behind spiracles, 2.8 times as long as apically wide, coriaceous, with sparse long hairs laterally; spiracles protruding, situated at anterior 0.4 of tergite; post-petiole somewhat inflated in middle and narrowed to apeX; first sternite 0.4 times as long as tergite length, in lateral view forming rounded lobe; second tergite 1.2 times as long as apically wide. Parameres strongly tapered to apeX ( Figure 2 (f)).

Colour. Colouration as in female eXcept lighter on pronotum and hind femur; tergites III – IV yellowish, more fuscous laterally; tergite V with light brown median spot along anterior margin. Head dark brown; face yellowish, eXcept for fuscous lower corners and elongated central spot in upper part close to antennal sockets, more dusky on facial margins and malar space ( Figure 2 (b – d)). Scape, pedicel and first flagellomere yellowish, rest of flagellum fuscous; mandibles eXcept for reddish teeth, palps and tegulae pale. Mesosoma brown, mesepimeron and pronotum eXcluding anterior part and hind corners yellowish; propleuron mostly brown; metasoma fuscous, eXcept for yellowish thyridium, parameres, and apical margins of tergite II, tergite III and tergites IV – V anteriorly. Legs yellowish, fore and mid coXae and trochanters pale. Hind femur yellowish brown, hind tibia more fuscous with light basal band and dark brown area apically on dorsal face.

Variation. Flagellum with 32 – 35 flagellomeres in females and 28 – 32 flagellomeres in males. Fore wing sometimes with single indistinct bulla in 2m-cu, hind wing with cu-a and abscissa of Cu1 between M and cu-a forming single arched vein; one female specimen with sternite I reaching level of spiracles only. Colouration of propleuron and mesopleuron varying from fuscous to reddish-brown in both sexes, usually mesosoma darker dorsally than ventrally. Some male paratypes pale yellow on entire tergite III. In some female paratypes clypeus entirely yellowish; face, marks along notauli and tip of scutellum brownish. Hind femur and tibia range from yellowish to brown; male face varying from entirely pale to light brown; fuscous spot in centre of face sometimes enlarged as vertical mark connecting frons and clypeus ( Figure 2 (b – d)), or sometimes totally reduced.

Material examined

Holotype female ( BMNH), The Netherlands, Bleiswijk, 52.004771°N, 4.516682°E, orchid greenhouse , emerged 13 July 2015 eX larva of fungus gnat ( Diptera : Keroplatidae ) (leg. J. B. Woelke, H.M. Kruidhof and M. van Twist); Paratypes 12 females and 12 males ( BMNH, ZISP, FRI, RMNH), same label, but dates of emergence 10 July 2015, 13 – 16 July 2015.

Biology

Koinobiont, solitary larval endoparasitoid. All specimens of this species were reared from keroplatid larvae ( Figure 3 (a)), which were found between the bark and roots in pots of the orchid hybrid cultivars of ‘ Cambria ’. Three species of Keroplatidae were recorded in the orchid nursery: Lyprauta chacoensis , Lyprauta cambria ( Figure 3 (d)) and Proceroplatus trinidadensis . Larvae of the new Megastylus species consume host tissues almost completely ( Figure 3 (a)). Cocoons are tightly woven; thin and semi-translucent, light brown in colour, 4.5 – 4.7 mm in length and 1.3 mm in diameter ( Figure 3 (b)).

Distribution

The Netherlands, orchid greenhouses. This species presumably originates from the Neotropics as its probable hosts also originate from there.

Etymology

The species is named after its first collector and co-author of this article Jozef B. Woelke.

Discussion

Orchid growers face problems with root damage, especially damage to the root tips that results in increased branching of the roots just above the affected area. They attribute this damage to the action of Keroplatidae larvae, based on the frequent presence of a keroplatid larva in the vicinity of damaged roots ( Chandler and Pijnakker 2009; Pijnakker et al. 2010; Pijnakker and Leman 2013). The biology of keroplatid larvae is still largely unknown. All known species can be divided into two groups: larvae that are predaceous, spinning toXic diffuse nets with a low pH in order to catch prey, and larvae that are fungal spore feeders, spinning a sheet-like web to catch spores of polyporous fungi ( Aiello and Jolivet 1996; Matile 1996, 1997). The keroplatid larvae found during this survey all spun diffuse nets. It was also observed that they were actively hunting in the rearing cups when Carpoglyphus lactis mites and Sciaridae larvae were added. Even though the occurrence of orchid root damage seems to be related to the presence of keroplatid larvae, these larvae have not been observed to be actively feeding on the roots. The diet of keroplatid larvae as well as the eXact relationship between the keroplatid larvae found in orchid nurseries and orchid root damage require further investigation.

The three keroplatid species found in Dutch orchid nurseries: Lyprauta chacoensis , Lyprauta cambria ( Figure 3 (d)) and Proceroplatus trinidadensis , are not of Dutch origin ( Evenhuis 2006; Chandler and Pijnakker 2009; Pijnakker et al. 2010). From each of the genera Lyprauta Edwards ( Diptera : Keroplatidae ) and Proceroplatus Edwards ( Diptera : Keroplatidae ) only a single species occurs in the Palaearctic region ( Evenhuis 2006). Twenty-siX species of the genus Lyprauta have been described worldwide, of which nine species are found in the Neotropics including L. chacoensis ( Evenhuis 2006) . Of the genus Proceroplatus , 37 species are known worldwide, and 24 species are found in the Neotropics including P. trinidadensis ( Evenhuis 2006) . The eXact origin of L. cambria is unknown, because this species was first discovered in an orchid nursery in The Netherlands. However, this species is also believed to originate from the Neotropics ( Chandler and Pijnakker 2009).

As previously mentioned, three keroplatid species occur in the orchid nursery where Megastylus woelkei sp. nov. was found. Because the larvae of all these species look very similar ( Figure 3 (a)), they could not be separated and identified to species level, and all of them could be considered as a suitable host. The most abundant keroplatid species recovered from a light trap in the orchid nursery where M. woelkei was first discovered was L. chacoensis , followed by L. cambria . Also, a single specimen of P. trinidadensis was recovered. It seems likely that all keroplatid species were parasitized, because at the time wasps were abundant in the greenhouse almost no adult keroplatids were monitored in the traps. As a consequence, the wasp population also declined.

The cosmopolitan genus Megastylus Schiødte ( Hymenoptera : Ichneumonidae : Orthocentrinae ) comprises 35 described species and is divided into two subgenera: Megastylus with 27 species and Dicolus Förster with eight species ( Yu et al. 2012; Humala 2014). However, only the Holarctic fauna has been studied, and many species of this genus are not described yet. Little is known about the hosts of Megastylus species but all known hosts of Megastylus species are restricted to the family Keroplatidae . Published host records are: Megastylus caseyi Ashmead and Megastylus orbitator Schiødte in North Carolina, USA, were reared from Orfelia inops Coquillett ( Townes 1945; Dasch 1992); Megastylus amoenus Dasch was reared from Macrocera sp. in Tennessee, USA ( Dasch 1992); Megastylus cruentator Schiødte was reared from Platyura sp. ( Mansbridge 1933); Megastylus panamensis Wahl was reared in Panama from Orfelia billuus Matile (Wahl 1996) .

The distribution of M. woelkei in Dutch greenhouses is not yet fully known. From the 10 orchid nurseries that were included in the survey, M. woelkei was only detected in a single nursery. However, approXimately 8 months after the first discovery of M. woelkei this parasitoid was also detected in another Dutch orchid nursery that had not participated in the survey. Interestingly, this second nursery has not recently eXchanged any plant material with the first nursery.

Moreover, the origin of M. woelkei is unknown and difficult to verify. The most likely option is that it has the same origin as the Keroplatidae host species, which are all believed to originate from the Neotropics ( Evenhuis 2006; Chandler and Pijnakker 2009). How eXactly these keroplatid species arrived in The Netherlands is unclear. They may have arrived with orchid plant material imported from the Neotropics. However, they could have been introduced with other ornamental species as well, as apart from Dutch orchid greenhouses keroplatids have also been found to occur on Anthurium (Araceae) , Gerbera (Asteraceae) and other potted plants in Dutch greenhouses ( Pijnakker et al. 2010). It may therefore well be that plant material imported to The Netherlands from the Neotropics also contained keroplatid larvae that were parasitized by M. woelkei . Moreover, the chance of discovering a new Megastylus species that originates from the Neotropics is relatively high. The West Palaearctic Megastylus species are rather well known ( Förster 1871; van Rossem 1974, 1983), but the Neotropical fauna of the genus Megastylus , as well as all orthocentrines, are still mostly unstudied ( Veijalainen et al. 2012, 2013). Until now, only two Neotropical species of Megastylus have been described: Megastylus panamensis from Panama ( Wahl 1996) and Megastylus kasparyani Humala from MeXico ( Humala 2014), and the number of Megastylus morphospecies registered in the Neotropics is higher than the number of described species in the world ( Veijalainen et al. 2012). Hopefully, future studies of the Neotropical fauna will allow the newly described species to be found in its natural habitat and its origin can be clarified.

Other possible sources of origin of M. woelkei seem less plausible, for the reasons outlined hereafter. One possibility would be that parasitized keroplatid larvae were brought to the greenhouse together with substrate imported for use in orchid pots. The three most frequently used substrates (in miXtures) are bark, coco chips and sphagnum. Bark used in Dutch orchid nurseries predominantly originates from Portugal and Spain, and bark from these two countries is commonly treated against insects by steaming. Therefore, the chance that M. woelkei has arrived with Portuguese or Spanish bark is relatively low. A small proportion of the bark is from Sweden. Although Swedish bark is untreated and therefore could contain parasitized keroplatid larvae, the chance of discovering a new Megastylus species from Sweden is rather low because of the relatively well-studied Swedish Megastylus fauna (van Rossem 1983). Coco chips are imported from Sri Lanka and are dried and pressed into blocks before transport, making survival of insects in this substrate unlikely. Sphagnum, which is imported from Chile, is also dried and pressed before transport, and is irradiated upon arrival to kill plant seeds.

Another theoretical possibility would be that M. woelkei occurs locally in The Netherlands, entered the orchid nurseries through windows that are opened on warm days, and effectively located keroplatid larvae to parasitize. However, this scenario seems rather unlikely. Moreover, the Dutch Megastylus fauna is well studied (van Rossem 1974, 1983), so the chance of discovering an unknown species here is relatively low.

So far, no effective biological control agent against keroplatid larvae has been identified. Adults of Dalotia (formerly Atheta ) coriaria Kraatz ( Coleoptera : Staphylinidae ), the predatory mites Macrocheles robustulus Berlese (Acari: Macrochelidae ), Hypoaspis miles Berlese and Hypoaspis aculeifer Canestrini (Acari: Laelapidae ) and the entomopathogenic nematodes Steinernema feltiae Filipjev (Nematoda: Steinernematidae ) and Heterorhabditis bacteriophora Poinar (Nematoda: Heterorhabditidae ) are not sufficiently effective. They do not persist in bark substrate and/or do not attack the larger keroplatid larvae, which may be due to the oXalic acid secreted by these larvae ( Pijnakker and Ramakers 2010; Pijnakker and Leman 2013). Megastylus woelkei may be used as a biological control agent against keroplatids in orchid nurseries as their presence was preceded by a strong decline in keroplatid adults in the greenhouse where M. woelkei was first discovered. More research is needed on the biology of Keroplatidae , on rearing methods of Megastylus wasps, and on the effectiveness of M. woelkei as a biological control agent against keroplatids in orchid nurseries.

Figures

Figure 1. Megastylus woelkei sp. nov. ι holotype female: (a) whole insectι scale bar 0.5 mm; (b) head (anterior view) ι scale bar 0.2 mm; (c) head and mesosoma (latero-ventral view) ι scale bar 0.5 mm; (d) head (dorsal view) ι scale bar 0.2 mm; (e) propodeum and tergites 1 – 5 of metasoma (dorsal view) ι scale bar 0.5 mm; (f) wingsι scale bar 0.5 mm.

Figure 2. Megastylus woelkei sp. nov. ι paratype male: (a) whole insectι scale bar 0.5 mm; (b – d) variation of face colour (anterior view) ι scale bar 0.2 mm; (e) head (dorsal view) ι scale bar 0.2 mm; (f) parameres (lateral view) ι scale bar 0.2 mm.

Figure 3. (a) Keroplatid larva parasitized by Megastylus woelkei sp. nov. ι inside the keroplatid larva the larva of M. woelkei is visible; (b) cocoon of M. woelkei sp. nov. from which the adult wasp has already emergedι scale bar 1.0 mm; (c) keroplatid pupa from orchid nurseryι scale bar 1.0 mm; (d) female of Lyprauta cambria ι one of the three probable host species for M. woelkei sp. nov. ι scale bar 1.0 mm.