Stylogaster eggs on blow flies attracted to millipede defence secretions in Tanzania, with a stab at summarising their biology (Diptera: Conopidae & Calliphoridae)

Abstract The genus Stylogaster Macquart (Diptera: Conopidae) is sister to the remainder of the Conopidae. While all other Conopidae are endoparasitoids of aculeate Hymenoptera, species of Stylogaster appear to be endoparasitoids of ‘orthopteroids’, as the only confirmed rearing records are from crickets and cockroaches. Many calyptrate flies have been observed with Stylogaster eggs attached, but since no Stylogaster have been reared from any dipterans, it is still unknown if these flies are hosts, results of accidental oviposition or carry the eggs to the actual hosts. In this study, we report our findings of Stylogaster eggs on blow flies (Calliphoridae) attracted to millipede defence secretions in Tanzania. Out of seven different species collected and a total of 301 specimens, only flies of the genus Tricyclea Wulp had Stylogaster eggs attached. Out of 133 Tricyclea collected, 32 (24%) had Stylogaster eggs attached and, with one exception, all eggs were attached to the abdomen. The lifecycle of Stylogaster is summarised and discussed with a particular focus on dipteran egg-carriers.


Introduction
The genus Stylogaster Macquart (Diptera: Conopidae) presents intriguing challenges with regard to the biology of its included species. The genus is remarkably distinct, with all species characterised by an extremely long, geniculate proboscis, elongate and tapering female terminalia and a harpoon-like anti-micropylar end of the egg (Brown et al. 2002). The genus has at times been placed in its own family (Stylogastridae, for example, by Séguy 1946, Rohdendorf 1964, Smith and Cunningham-Van Someren 1985, but has more recently been given subfamily status and recognised as sister taxon to the remainder of the Conopidae (Hennig 1966, McAlpine et al. 1989, Gibson et al. 2010, Gibson et al. 2012, Gibson and Skevington 2013. Currently, 125 species of Stylogaster are recognised in the world, with the main diversity in the Neotropical (73 species) and Afrotropical regions (42 species, including Madagascar), but species are also found in parts of North America, Asia, the Philippines, New Guinea, eastern Australia, Tasmania and New Caledonia (Smith 1967, Schneider 2010, Stuke 2012, Stuke 2017.
The natural history of Stylogaster is poorly understood and present evidence on the breeding biology is sparse, although host-seeking and apparently gravid females are frequently encountered (Rettenmeyer 1961a, Rettenmeyer 1961b. Females of Stylogaster ram harpoon-like eggs into potential hosts, using their characteristic elongated oviscapt (Lopes 1937, Kotrba 1997 and present evidence indicates that the larvae are internal parasitoids, living in the abdomen of the host (Smith and Cunningham-Van Someren 1985, Woodley and Judd 1998, Etzler 2019. For the Nearctic Stylogaster with cricket hosts, the larva pupates outside of the host Judd 1998, Etzler et al. 2020) and the host usually dies upon emergence of the larva or lives for only a short time thereafter (Etzler et al. 2020). Many Stylogaster species are facultative army ant and driver ant followers (Carpenter 1914, Bequaert 1922, Bequaert 1930, Aldrich 1930, Lopes 1937, Cohic 1948, Rettenmeyer 1961a, Rettenmeyer 1961b. The females search for potential hosts in front of the leading edge of the advancing raids, looking for potential hosts fleeing from the ants (Stuckenberg 1963, Smith 1967, Smith 1969, Smith and Cunningham-Van Someren 1985, Kotrba 1997, Couri and Pont 2006, Couri and Barros 2010. When not following army ants, adults of Stylogaster can be found hovering over sunlit paths in the forest understorey or feeding off the nectar on small white or yellow flowers (Rettenmeyer 1961a, Burt et al. 2014). The few hosts that are confirmed from actual rearing of Stylogaster specimens are all crickets and cockroaches (Smith and Cunningham-Van Someren 1985, Woodley and Judd 1998, Etzler et al. 2020. This stands in stark contrast to the fact that Stylogaster eggs have been found attached to several species of Diptera, none of which has been confirmed as hosts. These Diptera 'eggcarriers' are primarily calyptrate flies. In fact, there are only four observations of Stylogaster eggs on non-calyptrate flies, i.e. a single species (and specimen) each of Conopidae, Heleomyzidae, Lauxaniidae and Syrphidae (Berghe et al. 1956, Stuckenberg 1963, Smith and Cunningham-Van Someren 1985, Kotrba 1997. Rettenmeyer (1961a) was the first to mention Stylogaster ovipositing on Diptera, while studying army ants in Panama. He found Stylogaster eggs attached to tachinid flies of the genera Calodexia Wulp and Phasia Robineau-Desvoidy (as Androeuryops Beneway) following army ants and he observed Stylogaster ovipositing on unspecified "other insects", but he provided no rearing records. Firm evidence of actual parasitisation is still restricted to Smith and Cunningham-Van Someren (1985), who dissected larvae of Stylogaster from immature cockroaches, as well as from one cricket in Kenya; Woodley and Judd (1998), who reported Stylogaster biannulata (Say) as reared repeatedly from Gryllus rubens Scudder in USA, Florida; and Etzler et al. (2020), who reared Stylogaster neglecta Williston from the cricket Oecanthus nigricornis (Walker) sampled from Canada (southern Ontario) and USA (New York State). The presence of Stylogaster eggs on various calyptrate flies has led to speculation about possible dipteran hosts, but as no larvae of Stylogaster have been found within any fly and no adults have been reared, this is still uncertain (Rettenmeyer 1961a, Rettenmeyer 1961b, Smith 1967, Smith 1969, Smith 1974, Smith and Cunningham-Van Someren 1985, Couri and Pont 2006, Couri and Barros 2010, Couri et al. 2013, Couri et al. 2019.
In this paper, we aim at compiling and reviewing available data on Stylogaster biology, with a focus on what is known about hosts and egg-carriers. We are adding our own data on Stylogaster eggs found on calyptrate flies from specific sampling focused on flies attracted to millipede defence secretions in Udzungwa Mountains National Park, Tanzania and we discuss the lifecycle of Stylogaster with its possible host range, in order to stimulate further research into the biology of Stylogaster.

Material and methods
Specimens of potential egg-carriers were collected (by TP) by placing injured or crushed local juliform millipedes in a white plastic tray or on a sheet of white cloth. The flies attracted to the millipedes where collected using a hand net and stored in 70% ethanol for further examination. All collections where made on 21 August 2018 at the same locality in Tanzania: Mizimu camp, Udzungwa Mountains National Park, Morogoro Region, which is montane rainforest at an altitude of 769 m a.s.l. (07°48'23.40"S; 36°51'7.29"E). All material for this project has been deposited at the Natural History Museum of Denmark.
Specimens were examined and identified in 70% ethanol, with some being pinned and airdried for imaging. For a reliable identification, male terminalia were dissected by making a cut between tergites 4 and 5, separating all of segment 5 plus the male terminalia and then isolating both sternite 5 and male terminalia from tergite 5. Male terminalia and sternite 5 were treated with 10% potassium hydroxide (KOH) for 24 hours at room temperature to macerate all soft tissue, then immersed in acetic acid, washed in distilled water, dehydrated in ethanol and transferred to glycerol for examination. After examination, all structures were stored in glycerol in a microvial pinned with the specimen. Each specimen with Stylogaster eggs was labelled with a unique identifier and the position of the eggs was recorded.
A series of photographs was taken using a Visionary Digital Imaging System with a Canon EOS 7D and stacked using Zerene Stacker version 2.0 (Zerene Systems LLC, Richland WA, USA). Superimposed photographs were edited using Adobe Photoshop® CS6 and GIMP 2.10. Drawings were digitally inked using Adobe Illustrator® CS6.
A thorough search was made in relevant literature for all records of Stylogaster hosts or egg-carriers and of other details of relevance for Stylogaster biology.

Results
From the flies attracted to the wounded millipedes, a total of 301 calliphorid flies belonging to three different genera were collected and examined: Phumosia Robineau-Desvoidy (one species), Hemigymnochaeta Corti (four species) and Tricyclea Wulp (two species). Eggs of Stylogaster were only found on the two Tricyclea species: Tricyclea fasciata (Macquart) and Tricyclea sp. A, which were also the most numerous flies in the sample as 44% of the calliphorids belonged to Tricyclea. In total, 133 specimens of Tricyclea were collected, of which 32 (24%) had Stylogaster eggs attached and a total of 48 Stylogaster eggs were counted ( Table 1). The number of eggs attached per individual ranged from one to five eggs, with an average of 1.4 eggs attached per individual. The distribution was as follows: 21 (44%) individuals with only one egg attached, nine (19%) with two eggs, zero with three eggs, one (2%) with four and one (2%) with five eggs attached. There was no significant difference in the number of eggs attached per individual between male and female flies (Fisher's Exact Test: T. fasciata (N = 23) p = 1, T. sp. A (N = 9) p = 0.57) or between the two Tricyclea species (Fisher's Exact Test: (N = 32) p = 0.6). Stylogaster eggs were predominantly found attached to female Tricyclea, although this was non-significant (Yates's chi-square (1, N = 165) = 0.28, p = 0.6), with 21 (65.6%) females and 11 (34.4%) males of the total of 32 Tricyclea individuals collected with eggs attached. However, there is a marked difference at the species-level, where more females carry eggs in T. fasciata (Yates's chi-square (1, N = 48) = 3.85, p = 0.05), while more males carry eggs in T. sp. A, although the latter difference is non-significant (Yates's chi-square (1, N = 117) = 0.17, p = 0.7).

Number of specimens collected per species
Stylogaster eggs on carriers according to sex Stylogaster eggs were only found attached to the posterior part of the abdomen of both male and female carriers ( Fig. 1) and mostly to the ventral surface, with the single exception of an egg found on the thorax of a female, which also had one egg on the abdomen (Fly B1 in Fig. 2, Suppl. material 1). Eggs were most often attached to tergite 5 and the terminalia (Fig. 2, Suppl. material 1).  genera and 48 species. The Calliphoridae have 48 observations, for three genera and 10 species. In the Neotropical region, the only records are from Tachinidae, with 17 observations for two genera and seven species. Around half of the calyptrate species recorded with Stylogaster eggs have more than one record per species and the eggs are primarily found on female flies (  Records of Stylogaster egg-carriers and hosts worldwide. Confirmed hosts with rearing records are underlined. Obs. = total number of observations for a given carrier species. For each egg-carrier the sex is given followed by the number of observations, for example, ♀:3 ♂:1 ?:1 for three females, one male and one specimen of unknown sex with one or more Stylogaster eggs attached.
Observations are a total of all observations from references given. For taxa not identified to species level, the taxa follow the original identification and are put in quotations marks, for example, 'Lauxanidae sp. ' from Stuckenberg (1963 The Stylogaster species identified from attached eggs and host or egg-carrier data are compiled in Table 3. Number of Stylogaster eggs for specific body parts of calyptrate eggcarriers is summarszed for each genus in Table 4. Average of Stylogaster eggs per fly for calyptrate egg-carriers is presented in Table 5 and the proportion of calyptrate flies with Stylogaster eggs versus the total number of calyptrate flies collected is presented in Table 6.
The Stylogaster species identified from attached eggs and host or egg-carrier. References as in Table 2. For taxa not identified to species level, the taxa follow the original identification and are put in quotations marks, for example, 'S. cf. ornatipes Kröber, 1914' from Lopes (1937. [ Lopes (1937) did not find any S. stylata eggs attached to Blattodea, but mentioned several observations of attacks from S. stylata. Genus misspelled as "Suilla". Given as "Dichaetomyia albinita Stein". Given as "Pseudobdellia subsetosa Curran", here interpreted as Helina subsetosa Curran, 1938, with the valid name Pseudohelina nigritarsis (Jaennicke, 1867). Smith and Cunningham-Van Someren (1985) observed S. westwoodi attacking cockroaches, but they did not find any eggs on the cockroach in which they found a Stylogaster larva. Smith (1967)

Known hosts of Stylogaster
The only hosts of Stylogaster that are confirmed from actual rearing records are cockroaches (Blattodea) and crickets (Orthoptera, Gryllidae) from the Nearctic and Afrotropics (Table 3). Three Stylogaster species have been reared from crickets: the Nearctic S. biannulata (Say) and S. neglecta Williston Judd 1998, Etzler et al. 2020) and the Afrotropical S. westwoodi Smith (Smith and Cunningham- Van Someren 1985). One species has been reared from cockroaches: the Afrotropical S. varifrons Malloch (Smith and Cunningham-Van Someren 1985). Besides these rearing records, there are only two other records of Stylogaster eggs attached to non-dipterans, which are those of Lopes (1937), who reported one egg attached to a cockroach (Chorisoneura sp.) and one egg attached to an undetermined orthopteran. There has been some confusion about other records of eggs on crickets and cockroaches, i.e. Ferrar (1987) referring to Rettenmeyer (1961b) as confirmed records of eggs and Taber and Maloney (2006) referring to Stuckenberg (1963), but in both cases, these works do not provide any new records and refer to the records from Lopes (1937). The almost total lack of records of Stylogaster eggs attached to cockroaches and crickets, as compared to the numerous records of eggs on calyptrate flies, could be due to a lack of coordinated search efforts, as suggested by the study of Etzler et al. (2020), where many crickets with Stylogaster eggs and larvae were collected after targeted sampling. Another explanation could be that eggs have been overlooked or that eggs detach after some time, as one observation suggests (Smith and Cunningham-Van Someren 1985). An explanation could also be that hosts are induced to express a 'grave-digging' behaviour before dying, as has been documented for larvae of some Conopidae in their hymenopteran hosts (Müller 1994, Rasmussen and Cameron 2004, Malfi et al. 2014, which would make parasitised hosts less prone to being collected.

Stylogaster egg-carriers
Stylogaster eggs have been found attached to several different dipterans of the families Anthomyiidae, Calliphoridae, Heleomyzidae, Lauxaniidae, Muscidae, Rhiniidae, Syrphidae, Tachinidae, even Conopidae (a Stylogaster!) and eggs have also been found on a spider ( Table 2). The non-calyptrate records are all single specimens and could easily be explained away as accidental egg-impaling by Stylogaster females. In contrast to this, a large range of calyptrate flies have many records of Stylogaster eggs, especially species of Muscidae, Calliphoridae and Tachinidae, some even from the same collection event (Table  6). However, despite the presence of eggs, no Stylogaster larva has ever been recovered inside a calyptrate fly; this could be due a lack of a coordinated effort of dissecting the flies carrying the eggs, as all authors remove the Stylogaster eggs without dissection of the carrier flies. The only attempt at dissecting known egg-carrier flies is by Rettenmeyer (1961a), who dissected 20 females of Calodexia without Stylogaster eggs, but with abnormal abdomens and without finding any Stylogaster larvae. Therefore, there is no firm evidence that these flies are regular or occasional hosts or even if they are hosts at all. It has been speculated that the calyptrate flies instead are used to transport the eggs to the final host or food source or that the flies just happen to share the same appearance or habitat as the host of Stylogaster and, therefore, accidentally become impaled with eggs (Smith 1967, Ferrar 1987, Couri and Pont 2006, Couri and Barros 2010, Stuke 2012, Couri et al. 2013, but the evidence to support this remains circumstantial. As noted by Stuckenberg (1963), the Afrotropical egg-carriers seem to be mostly yellowishbrown and forest dwelling. Almost all the dipteran egg-carriers share a yellowish-brown abdomen with black or dark stripes (Table 2). This could indicate that this pattern somehow triggers Stylogaster to oviposit, either because the flies are potential hosts or because they resemble the actual Stylogaster host. However, the crickets and cockroaches, so far recorded as hosts, do not have this pattern on their abdomen.

Stylogaster egg placement
Females of Stylogaster predominantly attach their eggs to the abdomen of the host (Lopes 1937, Smith and Cunningham-Van Someren 1985, Woodley and Judd 1998, Etzler 2019 and only eggs attached to the abdomen develop successfully (Etzler 2019). This, combined with the fact that the Stylogaster larva, as with other Conopidae (Brown et al. 2002, Stuke 2017, develops in the abdomen of the host, would indicate that the eggs must be attached to the host abdomen for a successful development and that eggs attached elsewhere would be misplaced and unsuccessful.
The distribution of Stylogaster eggs on the calyptrate flies appears to vary between genera (data too sparse to allow assessment per species). Taking into consideration the different proportions of surface area for head (17.5%), thorax (54.1%) and abdomen (28.4%), Stuckenberg (1963) and Smith (1967) found the attached Stylogaster eggs to be randomly distributed on the bodies of the flies, for example, from the muscid genera Dichaetomyia Malloch, Dimorphia Malloch and Pyrellina Malloch. This agrees reasonably well with our data compiled for Muscidae, where most eggs are placed on the thorax (62%), followed by the abdomen (19%) and head (15%) (Table 4). However, the Calliphoridae and Tachinidae have most Stylogaster eggs attached to the abdomen, with 89% and 65%, respectively ( Table 4). The present material of Tricyclea spp. shows a strong concentration of Stylogaster eggs on the postero-ventral part of the abdomen, which would make sense if the Stylogaster female attacks a flying potential carrier from behind. This is also in agreement with Rettenmeyer (1961a), who reported Stylogaster eggs to be concentrated on the abdomen of females of Calodexia spp. and one male and female of Phasia ecitonis ( Table 4). The known Neotropical carriers are almost exclusively female Tachinidae with a host-seeking behaviour associated with foraging army ants and with hosts amongst Orthoptera, Blattodea and Heteroptera, which are attacked as they flee from the ants (Rettenmeyer 1961a, Rettenmeyer 1961b, Brown et al. 2010. If species of Stylogaster share one or more hosts with species of Calodexia, a possible scenario would be Stylogaster females accidentally impaling females of Calodexia when both are darting after an orthopteran or a cockroach fleeing from the foraging ants.

Oviposition strategy
Parasitoids, like Stylogaster, with a direct deposition strategy, produce a small number of eggs and often tend to be oligo-or monophagous. The clutch size of the Afrotropical species of Stylogaster is about 60-128 eggs (Stuckenberg 1963, Smith 1967, the Neotropical S. stylosa carries 120 eggs (Kotrba 1997) and the Neartic S. neglecta carries the most eggs with around 155 eggs per female (Taber and Maloney 2006). The modest number of eggs is likely related to a high rate of successful parasitisations, i.e. the gravid female allocates more energy to host seeking and egg deposition in order to secure the offspring an optimal developmental environment, rather than to increased egg production, as is common in parasitoids with an indirect oviposition strategy, where eggs have a lower chance of being picked up by a suitable host. It would, therefore, appear likely that Stylogaster females could afford to 'waste' only very few eggs on non-host impaling. Etzler (2019) observed that mature Stylogaster larvae sharing a host appeared to be smaller than larvae that did not and only a few crickets had multiple larvae. This indicates that one or only a few eggs per host is the optimal strategy, as smaller larvae from shared hosts would produce smaller adult flies with a lower fitness (Etzler 2019). However, this could vary with host size, as the two cockroaches examined by Smith and Cunningham-Van Someren (1985) both had multiple eggs and larvae. Calyptrates with many observations of Stylogaster eggs show an average number of 1.45 eggs per fly (Table 5). This compares with 1.25 Stylogaster larvae per cricket found by Etzler (2019), who does not provide an estimate of number of eggs per cricket.

Larval biology
It is not known how the Stylogaster larva enters the host (Fig. 3). Some have suggested that the larva enters through the extrusible sac at the anti-micropylar end (Fig. 3D2), as the barbed part of the egg, which is stabbed into the host, is presumed too heavily sclerotised for the larva to exit. Other records indicate that the larva may emerge from the blunt micropylar end of the egg, facing away from the host (Fig. 3D1, Rettenmeyer 1961a, Rettenmeyer 1961b, Stuckenberg 1963. This is further supported by records of larvae inside attached eggs, placed with their head towards the blunt micropylar end of the egg (Fig. 3D, Smith and Cunningham-Van Someren 1985, Couri et al. 2013), and Stuckenberg (1963 found an empty egg attached to Dichaetomyia quadrata (Wiedemann) with the micropylar end "irregularly broken open" and interpreted this as a hatched egg. If the larva hatches from the egg through the end facing away from the host, especially for eggs that are attached to the ventral part of the host, it would appear that there is a high risk that the emerging larva would fall off the host. That would support the hypothesis of the calyptrate flies functioning as egg-carriers rather than true hosts.

Biology of hosts/egg-carriers
The two species of Tricyclea Wulp found to be impaled by Stylogaster eggs in the present study have a remarkably similar -and notably high -rate of infection (23-24%) ( Table 1). This matches the overall infection rate of the cricket Oecanthus nigricornis, which is the known host of the Nearctic S. neglecta, although the infection rate for individual crickets can vary significantly per site (Table 6, Etzler et al. 2020). Male crickets of medium size had a significantly higher rate of parasitism than females in the Etzler (2019) study. Amongst the calyptrate flies, Stylogaster eggs were predominantly attached to female flies, although this could be an artifact of limited sampling as most species have very few records and many records are from different collection events (Table 2) or it may be due to a higher abundance of female flies where Stylogaster search for hosts. More surprisingly, while specimens of both (and therefore all) species of Tricyclea were impaled, none of the four species of Hemigymnochaeta Corti was impaled (Table 1), even though all six species are very similar (at least to a human observer) and were collected at the same event. Other studies have recovered specimens of Hemigymnochaeta carrying eggs of Stylogaster (  (Villeneuve 1922). The flies included in the present study were collected as they were attracted to the benzoquinone-based defence secretions of juliform millipedes (T. Pape, pers. obs.), which is behaviour known from the millipede-associated species of the Nearctic flesh fly genus Spirobolomyia Townsend and probably the pantropical scuttle fly genus Myriophora Brown (Hash et al. 2017, Hash et al. 2018), but not previously documented for blow flies. A possible explanation could be that foraging driver ants encountering millipedes will cause the latter to release their defence secretions, which will attract a variety of flies (Table 1) and which are then coming into the range of host-seeking Stylogaster females hunting for hosts. This would then be similar to the case of Neotropical Tachinidae with Stylogaster eggs, which have host-seeking behaviour associated with foraging army ants and a similar pattern of Stylogaster eggs attached on the abdomen as seen in Tricyclea, although the proportion of the Tachinidae with Stylogaster eggs, Calodexia spp. at 0.8% and Phasia ecitonis at 0.3%, is much lower than that reported here for Tricyclea (24%) ( Table 6, Rettenmeyer 1961a).
This will not, however, explain why the species of Tricyclea have Stylogaster eggs predominantly inserted at the tip of the abdomen rather than distributed randomly as for other calyptrate flies, nor will it explain why, in the material studied here, species of Tricyclea are impaled, while those of Hemigymnochaeta are not.

Phylogenetics and biogeography
Due to our limited data on Stylogaster hosts, there seems to be no phylogenetic pattern in the position of Stylogaster with confirmed hosts. Stylogaster species that parasitise crickets are found in all three major Stylogaster groups and both in the Nearctic, Neotropics and Afrotropics (Fig. 4). The pattern seems to be the same for Stylogaster parasitising cockroaches. The only rearing record is from the Afrotropical S. varifrons, but Stylogaster eggs have also been found on cockroaches from two Stylogaster species with distributions in the Nearctic and the Neotropics. The same holds true for the records of Stylogaster species with eggs on dipterans, which are also found in both the Nearctic, Neotropics and Afrotropics, although it is noteworthy that the majority of records -and all the non-tachinids -are from the Afrotropics (Table 2).

Conclusion
Tricyclea fasciata and T. sp. A appear to be likely candidates for dipteran hosts of Stylogaster, even though a rearing record is still needed to finally confirm this. The records of Stylogaster eggs on Tricyclea differ from those from other calyptrates and support the hypothesis that species of Tricyclea are hosts of Stylogaster. First, the proportion of Tricyclea with Stylogaster eggs reported here (24%) is higher than most of the other calyptrate observations. Second, the Stylogaster egg placement on the abdomen of Tricyclea is similar to that on the confirmed hosts of Stylogaster and not random as for most of the other calyptrates. Third, the morphologically very similar Hemigymnochaeta that were collected from the same site as the egg-carrying Tricyclea spp. had no Stylogaster eggs, which suggests targeted rather than indiscriminate oviposition. We are getting closer to understanding the biology of Stylogaster (Fig. 3), but there are still some major questions left. First and foremost, there is a need for hosts confirmed through rearing, which will also bring an indication of the host range. We need more data on the oviposition behaviour of Stylogaster females and, in particular, on how the larva hatches from the egg and enters its host.
Answering these questions is crucial if we want to understand the complex biological interactions that Stylogaster is a part of and the early evolution of Conopidae. For example, if host location is mainly by visual cues, looking for patterns or movement, as the observations of Stylogaster darting at moving hosts near army ants would indicate, then that would explain the association of Stylogaster with army ants and the eggs impaled in Neotropical Calodexia and, possibly, also why eggs are found predominantly in Afrotropical calyptrates with similarly-coloured abdomen.
is inside the host, extrusible sac and spines keeping the egg from falling off (2). How the larva exits the egg is unknown, the two proposed ways are illustrated (1 and 2) (1985), Kotrba (1997), Woodley and Judd (1998). Phylogenetic tree modified from Gibson et al. (2012). Most-parsimonious cladogram generated from combined molecular and morphological data, see Gibson et al. (2012) for details. Host and biogeographical information added. Stylogaster with confirmed hosts underlined. Biogeographical information from Stuke (2017) and host information from Table 3.