Introduction

Species of Achrysocharoides Girault (Hymenoptera: Eulophidae) are small parasitic wasps with transparent non-pigmented wings (Figs 1–6, 9). The short postmarginal vein in the fore wing is characteristic for the genus and the shape of the fore wing can be used to distinguish males of some species, but otherwise wings have been disregarded as non-informative neutral entities in this genus (e.g. Askew and Ruse 1974; Kamijo 1991). Recently wings in this group were discovered to display patterns with stable structural colours (Fig. 7), comparable to other insect groups with colourful wings such as butterflies (Shevtsova et al. 2011). These wing interference patterns (WIPs) become visible when transparent insect wings are seen against a dark background, and are most distinctive in small species with exceptionally thin wing membranes.

WIPs as a morphological character are so new that very little is known about the significance of these patterns for their bearers or for entomologists studying them, although they have already proven useful for generic-level classification in Eulophidae (Hansson 2011). The application of WIPs as a species character was first used in a study including two cases of cryptic species in Achrysocharoides (Hansson and Shevtsova 2010), where the initial species separation was based solely on male WIPs. However, data showing the usefulness of WIPs were withheld pending the publication of Shevtsova et al. (2011) where a general background to these patterns was outlined. In order to expand the knowledge of WIP diversity and to prove the usefulness of these patterns for studies at the species level it is important to link the information from these two publications. To further enhance this knowledge we also describe two new Achrysocharoides species with distinct WIPs.

The two new species of Achrysocharoides described here are from North America and the genus was initially recorded from this region by Miller (1962), as the genus Enaysma Delucchi, including six new species from Canada which were placed in the same subgenus (Pentenaysma Graham). Yoshimoto (1977) synonymized Enaysma with Achrysocharoides, and added nine species (six newly described) to the six described by Miller. He also separated the 15 species into two newly created species groups, thus abandoning the division into subgenera. The latest comprehensive treatment of North American Achrysocharoides is by Kamijo (1991), who treated 18 species, including four new species and one new synonym, separated into five species groups, two of which were newly created. Hansson and Shevtsova (2010) added two new species to the North American fauna, increasing the total to 20 species. With the two new species described here this total is now 22, equal to the number of species in Europe. Worldwide, including the two new species described here, 56 species of Achrysocharoides are known. The majority (ten) of the remaining species are from Japan (Kamijo 1990a, b), thus establishing the main distribution of Achrysocharoides as the northern hemisphere.

Material and methods

The observation and documentation of WIPs do not require a special light source and can be done on any dry specimen with intact wings arranged against a dark background. However, to make the illustrations comparable all photos in this paper as well as in Shevtsova et al. (2011) are of wings removed from the specimens and horizontally arranged, and with the same magnification (6×). To achieve this, the wings are flattened between a glass slide and a glass cover slip on top of the wings. The underside of the glass slide is stained with a drop of black ink to make the background pitch black and homogeneous (this was proposed by J. Kjærandsen). In a few cases where the wings could not be properly flattened the slide was slightly tilted so that the pattern in a non-flattened area, e.g. in a wrinkle, became visible and could be documented. This area was then manually combined in Adobe Photoshop with the initial horizontal photo of the wing, thus showing the complete pattern. A Nikon SMZ1000 stereomicroscope and 5MP Nikon DS-L1 camera were used to take photos of the wings at different focus levels, and Helicon Focus Pro version 4.75 software was used to merge them into a single image. WIPs are usually too shiny for the camera to balance brightness automatically and therefore the brightness was individually adjusted in Adobe Photoshop. Subsequent editing included cleaning and cropping of the photo. After the fore and hind wings were documented they were glued back to the card with the original specimen, which retained the second pair of wings for future observations – structural colours disappear on glued or slide mounted wings (Figs 1–6). The images of transparent wings in this paper are from temporary slide preparations with wings mounted in a water-soluble clear gel. The scanning electron microscopy (SEM) images (Figs 11, 13, 15, 17, 66–71, 82–87) are from uncoated specimens on their original card mountings. The photos were taken in low vacuum mode via a backscattered electron detector on a JEOL® JSM 5600LV microscope.

Morphological abbreviations and acronyms

HE = height of eye; HW = height of fore wing; LG = length of gaster; LM = length of marginal vein; LW = length of fore wing, measured from base of marginal vein to apex of wing; MM = length of mesosoma; MS = malar space; OOL = distance between one posterior ocellus and eye; PM = length of postmarginal vein; POL = distance between posterior ocelli; POO = distance between posterior ocelli and occipital margin; ST = length of stigmal vein; WH = width of head; WM = width of mouth; WT = width of thorax. For illustrations of the morphological terms see http://www.neotropicaleulophidae.com/.

Collection acronyms, for the deposition of type material: BMNH = Natural History Museum, London, England; CAES = Connecticut Agricultural Experiment Station, New Haven, U.S.A; CNC = Canadian National Collection of Insects, Ottawa, Canada.

Results and discussion

The paper by Hansson and Shevtsova (2010) included two cryptic Achrysocharoides species from Acer, Achrysocharoides platanoidae Hansson & Shevtsova from Acer platanoides and Achrysocharoides acerianus (Askew) from Achrysocharoides pseudoplatanus, and two cryptic speciesfrom Robinia pseudoacacia, Achrysocharoides robiniae and Achrysocharoides robinicolus, both described in that paper. The transparent wings in these four species are very similar and identical between males and females (Figs 1–4). Nevertheless the initial differences distinguishing these cryptic species were found in the wing morphology through distinct WIPs, which visualize uneven thickness of the wing membrane through different interference colours (Shevtsova et al. 2011).

In both cryptic cases only one of the species displays a distinct species specific WIP, and in males only, while conspecific females and both sexes of the other cryptic species have similar WIPs. In the two Achrysocharoides species associated with Acer only males of Achrysocharoides platanoidae have a distinctive WIP with an eye-catching blue spot in the upper-apical corner of the fore wing (Figs 18–21). The female WIP of Achrysocharoides platanoidae displays no such spot (Figs 22–23) and is very similar to Achrysocharoides acerianus, which has the same WIP in both sexes (Figs 24–27). In the two other cryptic species, associated with Robinia pseudoacacia, only males of Achrysocharoides robiniae display a very characteristic WIP with a large ovate spot below the marginal vein. The male WIP also has an extended and usually green triangular area in the medio-apical part of the fore wing (Figs 28–33). In the female WIP the triangular area is usually less pronounced than in males and the submarginal ovate spot is significantly smaller (Figs 34–35). As the female does not display the characteristic features in these patterns as distinctly as the male, it can be confused with the female of Achrysocharoides robinicolus, which has the same WIP in both sexes (Figs 36, 37).

The two North American species described here, Achrysocharoides maieri and Achrysocharoides serotinae, are known only from, and are probably confined to, Phyllonorycter species on Prunus. Males can be distinguished through easy-to-see differences in the external morphology, e.g. the shape of the head (Figs 57, 67, 73, 83) but females are not so distinct and display less divergent characters (Figs 56, 66, 72, 82). Even though the wings of Achrysocharoides maieri and Achrysocharoides serotinae appear very similar in transparent mode (similar to Figs 1–6) the WIPs in these species are distinct and specific. Apart from being useful in the discrimination of the females, in this case WIPs are also useful for the association of otherwise dimorphic males and females of the same species. The external morphology in these species exhibits a pronounced sexual dimorphism and as they share the same host it is not obvious which females and males are conspecific. However, there is one important character they have in common – WIPs, which are identical in both sexes but different between the species. Achrysocharoides maieri has a WIP with wide coloured cross bands on the fore wing (Figs 64–65), and Achrysocharoides serotinae has a quite featureless almost unicoloured WIP (Figs 80–81).

Additional examples of Achrysocharoides species with distinct and sexually dimorphic WIPs are Achrysocharoides butus (Walker) (Figs 38-43) and Achrysocharoides latreilleii (Curtis) (Figs 46-49) where characteristic and specific WIPs, again, are confined to males. Female WIPs of these two species are similar (Figs 44–45, 50–51), and as in females of Achrysocharoides platanoidae and Achrysocharoides acerianus (Figs 22–23, 26–27), and Achrysocharoides robiniae and Achrysocharoides robinicolus (Figs 34–35, 37), WIPs are not useful for species recognition. The WIP of male Achrysocharoides butus is similar to that of male Achrysocharoides platanoidae because the apical margin of the fore wing has a blue spot in both species. However, in Achrysocharoides butus this spot is prolonged and reaches along a major part of the apical margin (Figs 38–43) whereas in Achrysocharoides platanoidae the spot is short and confined to the upper-apical corner of the fore wing (Figs 18–21). The male of Achrysocharoides latreilleii is distinct not only in the truncate shape of the fore wing but also in its WIP (Figs 46–49). The basal 2/3 of the fore wing is the thickest part of the wing membrane and due to its micromorphology reflects very weak interference colours (Shevtsova et al. 2011). The apical part of the fore wing, and a small submarginal spot located in the corner between marginal and stigmal veins, are brightly coloured. The potential of WIPs as a character for separating species can be further demonstrated through two species where only male WIPs are known. Achrysocharoides albiscapus (Delucchi) has a WIP similar to that of Achrysocharoides latreilleii, but differs in having the basal 2/3 of the fore wing completely transparent without colour reflections and no submarginal colour spot (Figs 52–55). The shape of the fore wing is also different between males of these two species. The other species is undescribed, from Arizona, USA (specimen in CNC), and we have only seen a single male. This specimen has a distinctive WIP which emphasizes very unusual shapes of both fore and hind wings. The WIP includes a blue spot in the upper-apical corner of the fore wing (Fig. 8), comparable to Achrysocharoides platanoidae (Figs 18–21) but with the blue spot smaller and differently shaped.

Similar to other morphological characters there is a certain intraspecific variation in WIPs (Figs 18–55), but the species specific traits nevertheless remain clearly recognizable and are reliable for species separation. The intraspecific variation in WIPs can be divided into two types, variation in colour and in shape of patterns. Variation in colour is basically size-dependent – the thickness of the wing membrane usually varies with the size of the specimen and there is a general shift of the hues in WIPs from larger to smaller specimens. Variation in the shapes of pattern outlines of conspecific WIPs is not apparently size dependent but reflects individual differences between specimens - the overall pattern nevertheless remains the same.

Wing interference patterns are due to structural organization patterns of the wing membrane where areas of different thickness reflect certain interference colours (Shevtsova et al. 2011). We have found that the uneven thickness of the wing membrane also can be demonstrated and authenticated through the contrast in grey scale SEM images of uncoated wings. The SEM images created through back-scattered electrons (BSEs) visualize specific patterns on wings (Figs 11, 13, 15, 17). These patterns fully correspond to the approximate mapping of the wing thickness based on WIPs where the thickness of the wing membrane at any point can be estimated by the reflected interference colour (Shevtsova et al. 2011). The thickness gradient as seen through grey scale gradients in SEM images is due to specific properties of BSEs which have the escape depth of up to hundreds of nanometers (Egerton 2005). This means that the signal comes from a sample depth in the range comparable to membrane thickness in wings producing bright WIPs, i.e. 100–600 nm. In uncoated wings the primary (incident) electrons are scattered inside the membrane and reflected as BSEs to the back-scatter detector. In thick areas of the membrane the amount of BSEs is large, resulting in a strong signal, while thin areas of the membrane produce fewer BSEs and a weaker signal, thus displaying light and dark grey hues respectively. If the wings are coated with platinum or gold, the resulting picture is completely different due to secondary electrons (SEs) which are generated only within a very small distance below the surface as the escape depth of SEs is less than two nanometers (Egerton 2005), thus displaying the surface of the specimen rather than the underlying structure. In Shevtsova et al. (2011) secondary electron images were used to illustrate the microstructures of the wing surface, such as the ridges of membrane corrugations with rows of setae.

The clarification of the two cryptic species on Acer spp. (Hansson and Shevtsova 2010, Shevtsova et al. 2011) requires a correction of the molecular information deposited in Genbank. At the time of the publication of Lopez-Vaamonde et al. (2005) the identity of the Achrysocharoides species associated with Acer spp. was not clear, and “Achrysocharoides acerianus ex Acer platanoides” and “Achrysocharoides sp. ex Acer pseudoplatanus” in Lopez-Vaamonde et al. (2005) are Achrysocharoides platanoidae and Achrysocharoides acerianus respectively, which is confirmed here with new molecular analyses compared to data of “Achrysocharoides sp.” and “A. acerianus” in Genbank. Our new sequences include CO1, 18S, 28S and will be deposited in Genbank.

Species descriptions