Unusually for Ichneumonidae, Trogus lapidator emerges through a hole in the pupal wing case of its papilionid butterfly host that is made largely by a liquid secretion that softens and disintegrates the host tissue. The mandibles are deployed to help spread the secretion, but only towards the very end of the emergence process are they used (and then only in a minor way) to enlarge the hole. Links to video clips showing the emergence of T. lapidator are provided. Photographs illustrating the nature of emergence holes left in Lepidoptera pupae by a range of Ichneumonidae and some Chalcidoidea are presented and discussed, contrasting with the emergence hole left by Trogus and close allies.

Pupal cuticle, eclosion, cuticular disintegration, staining, cap-cutting, mandibular structure

One of the general functions of the mandibles of adult Hymenoptera is to aid emergence from the cocoon or other pupation site, although the wide range of variation of mandibular structure suggests that this is not always achieved in the same way, and indeed often indicates other functions for the mandibles too (undoubtedly including nest-building, prey manipulation, accessing hosts, and feeding; uses of the mandibles that are well-known in many relevant taxa). Many Ichneumonoidea pupate in well-defined cocoons, from which adults of various groups emerge by cutting a cap or strips with great precision, but others chew more or less irregular holes for the purpose. In contrast with Braconidae (the small subfamily Meteorideinae and a small and unusual subtribe, Aspidobraconina, of Braconinae aside), many Ichneumonidae emerge as adults from Lepidoptera pupae, whether as true pupal parasitoids (some Pimplinae, many Ichneumoninae, a few Cryptinae), or as larva-pupal parasitoids (Metopiinae, most Anomaloninae, many Ichneumoninae, occasional Campopleginae). In these cases it is usual for adult emergence to be either through a roughly shaped chewed hole, typically subapical, around which a scattering of bits of host cuticle can be found, or in many cases though the hole left by a more neatly cut and more or less detached apical cap (sometimes partly assisted by the host’s dorsal ecdysal sutures).

Trogus lapidator (Fabricius) is a widespread specialist koinobiont parasitoid of the Swallowtail Butterfly Papilio machaon Linnaeus (Papilionidae) in Europe, with rare confirmed records from the more restricted Papilio alexanor Esper (e.g. Sanetra 1998). The European nominal taxon Trogus violaceus (Mocsáry), which is confined to Corsica and Sardinia where it parasitizes both P. machaon and P. hospiton Guenée, is regarded as a synonym of T. lapidator by Wahl and Sime (2006). Trogus is a genus of about twelve species (Wahl and Sime 2006) of larva-pupal parasitoids of Papilio occurring in the Palaearctic and New World, and all probably have similar biology, invariably emerging as adults through the wing case of the host pupa. Closely related taxa such as Psilomastax pyramidalis Tischbein, parasitizing Apatura spp. (Nymphalidae: Apaturinae) make similar emergence holes, presumably in the same way, though in the case of Psilomastax the emergence hole is substantially more ventral. According to Sime and Wahl (2002) in their cladistic analysis of the Callajoppa genus-group, emergence through the wing case (their character 58) is unique to the Trogus-subgroup, which contains several additional genera.

During filming of the life history of Papilio machaon and its parasitoids by the second and third authors (included in https://youtu.be/PsU5rqeIXBg?list=PLLIrjN9bA2GwtZiIlVdmHwOcTUzk6q-6V, https://youtu.be/4K6F_pr1Om4?list=PLLIrjN9bA2GyWrineNq9vnKCct17m1ewH, https://youtu.be/GdBKDTYR57o?list=PLLIrjN9bA2GzkK_g-Tbq9odQYgt_mX1BS, https://youtu.be/6qygQ62nQWQ?list=PLLIrjN9bA2GxY-BAR-nHRUfcADsXSRytF and https://youtu.be/fchB-w67vTg?list=PLLIrjN9bA2Gwd7MyVv2im_Sbq-jly4Kl_) as part of their project to document the natural history of the French department of Var, the remarkable sequences showing the emergence of adult Trogus lapidator (https://www.youtube.com/watch?v=GrhKhsj00BE and https://www.youtube.com/watch?v=y_p8QcnHgjs) presented and discussed in this paper were brought to the attention of the first author. In order to set this in context, a brief illustrated overview of emergence from pupae of Lepidoptera by Ichneumonidae is given. When possible we have paid particular attention to those parasitizing butterflies which, by often being relatively exposed (and sometimes needing to persist for much of the year), may be expected to sometimes have relatively hard thick cuticle similar to that of P. machaon. Several groups of Chalcidoidea also emerge as adults from the pupae of Lepidoptera, and very brief mention is made of some of these.

Observations on Trogus lapidator, which has up to three annual generations in the region, were made and filmed during spring and summer 2013 in a large garden (1.5 hectares) in southern France (Var: Callas), where natural populations of the swallowtail butterfly Papilio machaon and various of its parasitoids, including T. lapidator, occurred. In some cases ovipositions by captive T. lapidator into middle instar host larvae had been witnessed and, following their pupation a week later and after the unwitnessed emergence of an adult T. lapidator several weeks later from a different pupa, two hosts were watched carefully for signs of adult emergence. The emergence process took place in late morning over a period of about 45 minutes and both cases were filmed with a Canon XL2 with a 20× L IS zoomobjectif XL 5.4-108mm with extender 72mm close-up lens 500D. The footage is registered on mini DV60 tapes. From the footage videos were made that can be seen at https://www.youtube.com/watch?v=GrhKhsj00BE and https://www.youtube.com/watch?v=y_p8QcnHgjs. Stills taken from these and another video resulting from a similar process were used for Figs 1–4 and 85.

Figure 1–10. 

1–4 Sequence in the emergence process of Trogus lapidator (Fabricius) from Papilio machaon Linnaeus (Papilionidae). 5–10 Host pupa following emergence of parasitoid. 5–7 Trogus lapidator from Papilio machaon 8, 9 Psilomastax pyramidalis Tischbein from Apatura iris (Linnaeus) (Nymphalidae) 10 Amblyjoppa proteus (Christ) from Deilephila elpenor (Linnaeus) (Sphingidae).

The National Museums of Scotland (NMS) houses a large collection of western Palaearctic Ichneumonoidea, made special by the quantity of reared material it contains, particularly in respect of parasitoids of Lepidoptera. A survey of emergence patterns of adults from Lepidoptera pupae seen in the three main subfamilies of Ichneumonidae with this habit (Ichneumoninae, Anomaloninae and Metopiinae) as well as lesser groups (Pimplinae: Pimplini, and the few relevant taxa of Campopleginae and Cryptinae) was undertaken from this resource, with particular attention to emergence from butterfly hosts. We illustrate a particularly large number involving Ichneumoninae, as this subfamily has proved to be the most variable. The survey was briefly extended to Chalcididae and Pteromalidae. Digital photographs of host Lepidoptera pupal remains from which Ichneumonidae (or Chalcidoidea) had emerged were taken using a Nikon D80 with Medical Nikkor 120mm lens attachment.

Quite apart from the video evidence (Trogus lapidator 1st emerging – 16 September 2013; https://www.youtube.com/watch?v=GrhKhsj00BE and Trogus lapidator 2^nd emerging - 16 September 2013; https://www.youtube.com/watch?v=y_p8QcnHgjs) and Figs 1–4, it is seen from the host remains through which Trogus and the closely related Psilomastax have emerged long before (Figs 5–9) that the pupal cuticle has been degraded by a wet substance, not only because of the ‘molten’ appearance around the edges of the emergence hole but also because it is more or less extensively stained. Unsurprisingly, papilionid pupae in BMNH from which Holcojoppa species (another Trogus-subgroup genus) have emerged have the same appearance (G. Broad, pers. comm.). We have observed generally sharper edges to the emergence holes of other ichneumonids, but staining would be expected to be less noticeable in the tough dark brown (subterranean) pupae typical of many large moths. Despite a moderate incidence of staining (see below), especially in Ichneumoninae, we have searched hard to see damage comparable with that caused by the Trogus-subgroup but have found very little sign of it elsewhere, even in the same tribe Heresiarchini (Figs 10–12) or in the relatively tough and often pale-coloured pupae (many of which are exposed and sometimes of long duration) of many butterflies from which a variety of ichneumonids have emerged. Thus we conclude that in general emergence from host pupae depends, as the existing perception has it, essentially on the action of the adult parasitoid’s mandibles, and that the Trogus-subgroup is exceptional. However, elsewhere in Ichneumoninae staining is rather widely seen, especially in some of the lighter-coloured butterfly pupal remains examined, such as Nymphalis polychloros (Linnaeus) parasitized by Hoplismenus terrificus (Wesmael) (Figs 13, 14), and some (Fig. 15) (but not all; Fig. 16) Coenonympha pamphilus (Linnaeus) parasitized by Hoplismenus bispinatorius (Thunberg), as well as in several Ichneumon species (Figs 18, 21, 23–25) although again without consistency (compare Figs 25 and 26). There are also a few weak examples from Listrodromini (Figs 38, 40). It is hard to be absolutely dismissive of the possibility of help from biochemicals in view of these clear signs of rather even staining around the edges of the emergence hole, especially in cap-cutting taxa. On the other hand, both in Ichneumoninae and in other subfamilies, it is possible that this can arise coincidentally (Fig. 40 shows an unusually stained Celastrina argiolus (Linnaeus) parasitized by Lystrodromus nycthemerus (Gravenhorst), with numerous others being free of staining; and Fig. 59 is the only example seen of staining in Anomaloninae, affecting only one area of the pupa of Zegris eupheme (Esper) from which Clypeocampulum lubricum (Atanasov) has emerged). While staining may result only incidentally, for there are always liable to be fluids accompanying the emergence of parasitoids (see Fig. 85, showing fluid in the hole being chewed by Pteromalus puparum, despite the typically clean appearance of the hole left in the (different) pupal cuticle once dry in Fig. 84), it points to a possibly interesting study awaiting anyone with the necessary patience, or luck, to investigate these phenomena in action. Chalcididae often leave strong stains behind them (Fig. 83) but this may just represent excess fluid consequent on their unusually small size in relation to the size of the pupa parasitized. The state of the host when arrested by the parasitoid is another factor likely to have a bearing on staining, as it might be expected that histolysed hosts could potentially leak quite a lot of fluid during parasitoid emergence, and in a rather inconsistent way.

Overall, even from this very sparse survey within a narrow zoogeographical region, there is considerable evidence that the form, toughness, concealment and duration of the host pupa plays a major role in determining how specialist ichneumonids emerge from it, although the trends within and between ichneumonid tribes and subfamilies outlined above might repay a more detailed and wider survey. While the toughness of the host pupa correlates quite strongly with its duration, and the strongest pupae tend to be cap-cut, there is not complete correspondence (for example, Agrypon polyxenae ecloses by chewing a hole (Figs 61, 64) in very tough exposed pupae that must persist unprotected for ten or more months in the field). In the subfamily Ichneumoninae both koinobiont and idiobiont taxa are found, but there seems to be no consistent difference in the mode of emergence between them, although we have had fewer examples of the latter category to explore.

It is rather obvious that some correlation between mandibular structure and the kind of exit holes made would be expected, and indeed this is rather easily seen — Trogus, in particular, has quite small coarsely punctate but otherwise unremarkable mandibles; the teeth are blunt and although the upper tooth is the larger the two teeth are in the same plane. But it is well beyond the scope of the present paper to attempt more than cursory notes on variation in mandibular structure in relation to eclosion, especially bearing in mind the minimal extent of the data. Also, the mandibular structure of the Chalcidoidea mentioned is fundamentally different from that of ichnemonids and cannot be compared. Nevertheless it seems clear that the best cap-cutters in the Ichneumonidae (e.g. Listrodromini (Figs 38–40), Hoplismenus (Figs 14–16) and Eurylabus (Fig. 54), as well as some Ichneumon and close relatives) have a relatively long and acute upper tooth and a reduced and sometimes inflexed lower tooth; and some of the crudest emergence holes are made by the taxa with untwisted mandibles and the least difference in size of the two teeth, such as Goedartia (Fig. 55) and Pimpla (Fig. 74). Even within a genus, such as in Ichneumon (Figs 17–29), this trend appears to hold up moderately well. However, it is not always the case: for example the phaeogenine Herpestomus brunnicornis (Gravenhorst) has the sharp upper tooth substantially the longer and yet exhibits predominantly type 2 emergence (Fig. 42). Another phaeogenine, Heterischnus, shows a consistent detaching of an apicoventral cap (Fig. 43), presumably in some way connected with its strongly curved, narrow and unidentate mandibles. Among Anomaloninae, the type 1 cap-cutting taxa seem to have the upper tooth on the whole slightly longer than in the hole-chewing taxa, and perhaps more particularly to have a stronger flange along the lower margin of the mandible. For some taxa, evolutionary pressure on the form of the mandible to enable emergence from the host’s substrate (or cocoon) must trump the barrier of the mere host pupal cuticle. This presumably accounts for the form of the widened blunt mandible of Chasmias and the crudity with which it leaves the host (Figs 36, 37) en route to chewing through a tough plant stem. Another case is the mess made of the host pupa (Fig. 56) by Cotiheresiarchesbefore it tackles the hard cocoon of its host; in this case the whole face is unusual, with a protruding plate-like clypeus as well as blunt, unidentate and curved mandibles, which presumably aid its escape from the host’s tough obovoid cocoon through a small circular hole (much as that made by the moth when it emerges, and possibly exploiting the same prepared facility; the cocoon in Fig. 56 has been artificially torn open). For others, such as Pimplini, not much specialisation of the rather broadly tapering and small toothed subequidentate mandible for eclosion seems to have occurred, although the arguably most specialised genus of those examined, Apechthis, which tends to cut strips (visible in both Figs 72 and 73) and emerges from relatively neat holes, does have an out-turned lower rim of the mandible if not quite a flange. An interesting difference is seen in the two species of Metopius examined (Figs 70 and 71), which belong to two subgenera with significantly different mandibular structure. Although Metopius (Peltocharus) dentatus (Fabricius) (Fig. 70), which has a tough host cocoon in addition to its pupa to contend with, has a slightly longer upper tooth the two teeth are in the same plane, while in M. (Tylopius) leiopygus Foerster (Fig. 71) the mandible is twisted, with the small lower tooth inflexed, so that the mandibles present as scissor-like blades. This brief survey by no means covers the range of mandibular structure seen in Ichneumonidae, and in fact most of the truly bizarre structures known in the family are not included, although probably most have nothing to do with Lepidoptera pupal hosts. Nevertheless, it seems clear that among taxa eclosing from Lepidoptera pupae there are strong links between the form of the mandibles and the kind of egress opening that is made, allowing also for any need to break through whatever additional protection the host pupa may habitually have, with some predictive possibilities. The extent to which these characteristics have evolved to meet (sometimes extreme) autecological requirements, rather than being rooted in phylogeny, is an unresolved question of a familiar kind, but the more that is understood about the basic biology of the host-parasitoid relationships the better such conundrums can be resolved.

The origin of the presumed secretion through which the host pupal cuticle is degraded by the eclosing Trogus adult is unknown. Although it would not explain how the necessary agent reaches the site of emergence, it would be worth investigating whether the extremely coarsely sculptured metasomal tergites of the Trogus-subgroup (which are unlike those of other ichneumonines, and whose morphology appears not to have been explained) may have a glandular, secretory function, as it is difficult to see what other (externally visible) morphological feature of Trogus might be involved, although its mandible is unusually coarsely punctate. Otherwise, the glossa and other submandibular aspects of the mouthparts seem rather enlarged in Trogus, but this might be a modification for merely spreading the fluid.

As an extension of this survey, we examined extralimital pupae of very large saturniid moths from which rather smaller solitary ichneumonids had emerged, on the one hand an unidentified koinobiont Agrypon (=Trichionotus) species (Anomaloninae) from a Mexican saturniid, Copaxa multifenestrata Herrich-Schäffer (Figs 86–88), and on the other hand an unidentified idiobiont Xanthopimpla sp. (Pimplinae) from Attacus atlas (Linnaeus) in Thailand (Figs 89, 90). In the first (Agrypon) the Copaxapupa was very malodorous some days after after the parasitoid’s emergence, and when opened (Fig. 88) the now-hard but messy remains of the host (apparently very unstructured; probably fully histolysed) could be seen to have been separated (perhaps within a membrane) from the area that had been occupied by the pupating parasitoid, which did not appear to have been isolated from them by any clear cocoon and may have resided in the ecdysal space of the host pupa. The staining seen in Fig. 87 seems more likely to have been a result of incidental spillage from wet host content than to indicate biochemical degradation of the host pupal chitin. In the case of Xanthopimpla the parasitoid appeared to have scraped its way through considerable distances of accreted material in the Atticus pupa, leaving a more or less smooth tunnel through a very hard substance, without resulting in any clear staining of the cuticle (Fig. 89) that might have indicated biochemical assistance (although most of its arduous journey to the surface did not involve the host cuticle, Xanthopimpla has basally broad but evenly narrowed mandibles which are strongly twisted so that the small ventral tooth is not involved in the powerful pickaxe-like anterior presentation). In contrast to the smelly Copaxa remains, those of the Attacus were set hard and (at least by the time they were received by MRS) inoffensive with a faintly sweet smell, perhaps indicating that some suppression of microorganisms had occurred, possibly in the manner reported for Pimpla by Führer and Willers (1986). But whether or not pimplines such as Pimpla and Xanthopimpla may provide interesting antibacterials, they do not seem to degrade chitin or other impediments to adult emergence biochemically.

It should be noted that no consideration has been given here to emergence of adult Ichneumonidae through the host pupal (or puparial) cuticle of Diptera (e.g. as by Diplazontinae which have specialized mandibles: Rotheray 1981), or the Orthocentrus-group of Orthocentrinae (in which mandibles are severely reduced).

We are especially grateful to the very many Lepidopterists who have donated reared parasitoids to MRS for NMS, without whose contribution that collection could not have sustained this survey. We are also indebted to Erich Diller, Klaus Horstmann, Matthias Riedel, Heinz Schnee and Martin Schwarz for some of the determinations of Ichneumonidae. Gavin Broad’s careful eye and helpful comments in reviewing the MS are also much appreciated.