Available via license: CC BY-NC-ND 4.0
Content may be subject to copyright.
The evolutionary history of mammoth wasps (Hymenoptera: Scoliidae)
Khouri, Z.1, Gillung, J.P.2, Kimsey, L.S.1
1 Bohart Museum of Entomology, University of California, Davis, CA, U.S.A; 2 Lyman Entomological Museum, McGill
University, Montreal, Quebec, Canada.
Abstract
Scoliid wasps comprise a clade of aculeate insects whose larvae are parasitoids of scarabaeid beetle
grubs. While scoliids have been studied and used as biological control agents, research into the group's
evolution, as well as the stability of scoliid taxonomy, has been limited by a lack of reliable
phylogenies. We use ultraconserved element (UCE) data under concatenation and the multispecies
coalescent to infer a phylogeny of the Scoliidae. In order to mitigate potential issues arising from
model misspecification, we perform data filtering experiments using posterior predictive checks and
matched-pairs tests of symmetry. Our analyses confirm the position of Proscolia as sister to all other
extant scoliids. We also find strong support for a sister group relationship between the campsomerine
genus Colpa and the Scoliini, rendering the Campsomerini non-monophyletic. Campsomerini
excluding Colpa (hereafter Campsomerini sensu stricto) is inferred to be monophyletic, with the
Australasian genus Trisciloa recovered as sister to the remaining members of the group. Out of nine
genera in which more than one species was sampled, Campsomeriella, Dielis, Megascolia, and Scolia
are inferred to be non-monophyletic. Analyses incorporating fossil data indicate an Early Cretaceous
origin of the crown Scoliidae, with the split between Scoliini + Colpa and Campsomerini s.s. most
probably occurring in the Late Cretaceous. Posterior means of Scoliini + Colpa and Campsomerini s.s.
crown ages are estimated to be in the Paleogene, though age 95% HPD intervals extend slightly back
past the K-Pg boundary, and analyses including fossils of less certain placement result in more
posterior mass on older ages. Our estimates of the stem ages of Nearctic scoliid clades are consistent
with dispersal across Beringia during the Oligocene or later Eocene. Our study provides a foundation
for future research into scoliid wasp evolution and biogeography by being the first to leverage genome-
scale data and model-based methods. However, the precision of our dating analyses is constrained by
the paucity of well-preserved fossils reliably attributable to the scoliid crown group. Despite
concluding that the higher-level taxonomy of the Scoliidae is in dire need of revision, we recommend
that taxonomic changes be predicated on datasets that extend the geographic and taxonomic sampling
of the current study.
Introduction
Members of the family Scoliidae, sometimes referred to as mammoth wasps, are large fossorial
aculeates that comprise one of the most visually striking and easily identifiable hymenopteran clades.
The family has a cosmopolitan distribution and includes approximately 560 described species (Osten,
2005). Adult mammoth wasps feed primarily on nectar, with honeydew (Illingworth, 1921) and
possibly pollen (Jervis, 1998) also reported as food sources. The larvae develop as ectoparasitoids on
the larvae of scarabaeid beetles (Clausen, 1940). Some studies have highlighted interesting aspects of
mammoth wasp natural history, such as parasitism of ant inquilines (Burmeister, 1854; Jonkman 1980),
pseudocopulation with orchids (Jones & Gray, 1974; Ciotek et al., 2006), fidelity of males to patrolling
sites (Tani & Ueno, 2013), and efficient location of subterranean hosts (Inoue & Endo, 2008). Despite
this, no study has attempted to reconstruct a phylogeny of the family, which precludes the examination
of scoliid biology in an evolutionary context.
The lack of a solid phylogenetic hypothesis has also contributed to a lack of taxonomic clarity and
stability. Day et al. (1981) referred to the group as "over-burdened nomenclatorially". Subsequently
Argaman (1996), while describing the state of scoliid taxonomy as "disastrous", established a new
.CC-BY-NC-ND 4.0 International licenseavailable under a
(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprintthis version posted June 26, 2022. ; https://doi.org/10.1101/2022.01.24.474473doi: bioRxiv preprint
subfamily, 21 new tribes, and 62 new genera without conducting a phylogenetic analysis. In assembling
a checklist of all scoliid species, Osten (2005) ignored the taxonomic changes implemented by
Argaman and implicitly synonymized many of the new taxa by placing their type species in other
groups (Elliott, 2011; Kimsey & Brothers, 2016). Currently, the need for a thorough taxonomic revision
is recognized (Elliott, 2011).
A robust phylogeny is a prerequisite for studies of character evolution, diversification patterns over
time, and biogeography, as well as for a natural taxonomy. In turn, the lack of a stable natural
taxonomy hampers research by making species determination difficult and by impeding the
communication and indexing of scientific information. In the case of mammoth wasps, this is
especially apparent in the context of their use as agents for the biological control of scarabaeid pests
(Illingworth, 1921; Wilson, 1960; DeBach, 1964). Misidentification of the control agent (for an
example, see Elliott (2011) on research by the Queensland Bureau of Sugar Experiment Stations)
precludes the repeatability of research and past biological control attempts and means that valuable
information discovered in the process cannot easily be traced to the right organism (Rosen, 1986). This
is particularly unfortunate, since a large portion of what is currently known about scoliid development,
phenology, and host interaction was discovered while evaluating and using mammoth wasps for
biological control (e.g. Illingworth, 1921; Miyagi, 1960). In the process of updating the BIOCAT
database of introductions of biological control agents, Cock et al. (2016) listed Scoliidae among the
groups requiring further taxonomic work.
In the present study, we aim to establish a solid foundation for research into mammoth wasp evolution
and systematics. We use ultraconserved element (UCE) sequence data (Faircloth et al., 2012; 2015) to
infer scoliid phylogenetic trees using concatenation and under the multispecies coalescent (Rannala &
Yang, 2003; Degnan & Rosenberg, 2009). Additionally, we leverage existing fossil data to estimate a
timeline of scoliid evolution. To better understand potential biases resulting from model
misspecification, we perform data filtering experiments based on matched-pairs tests of symmetry
(Jermiin et al., 2017; Naser-Khdour et al., 2019) and assessments of model adequacy using data-based
posterior predictive checks (Bollback, 2002; Huelsenbeck et al., 2001; Doyle et al., 2015).
Methods
Taxon and locus selection
We successfully sequenced 85 specimens of Scoliidae for this study. Taxon selection was aimed at
maximizing taxonomic and biogeographic diversity within the limits imposed by the availability of
material from which DNA could be extracted. All biogeographic realms are represented, but with
weaker sampling in Australasia and the Neotropical and Palearctic regions. We also included
previously published data (Johnson et al., 2013; Faircloth et al., 2015; Branstetter et al., 2017a; Peters
et al., 2018) from six additional scoliid specimens. See Table S1 for specimen collection data, voucher
information, and resources used for taxonomic determination.
Based on an examination of morphology, we suspected that Scolia bicincta may constitute two separate
species. We therefore sequenced multiple individuals from each putative species. However, given the
focus of the current study on reconstructing the scoliid phylogeny and identifying major clades rather
than on species delimitation, we retained only two specimens following a preliminary phylogenetic
analysis (see below).
We used the bradynobaenid genus Apterogyna as the only outgroup, and mined UCE sequences (see
"Sequence quality control, assembly, and UCE identification" section below) from the partial genome
published by Johnson et al. (2013). No sequences from other bradynobaenid taxa are publicly
available, and we were unsuccessful in sequencing the specimens of Bradynobaenus chubutinus to
.CC-BY-NC-ND 4.0 International licenseavailable under a
(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprintthis version posted June 26, 2022. ; https://doi.org/10.1101/2022.01.24.474473doi: bioRxiv preprint
which we had access. Bradynobaenidae is well-supported as the sister group to Scoliidae (Johnson et
al., 2013; Branstetter et al., 2017a; Peters et al., 2018). It is a species-poor clade, making it easier to
avoid highly disproportionate taxon sampling, which would be difficult if ants or apoids were used.
Adding more distant outgroups also increases the chance that heterogeneity in the evolutionary process
across lineages results in more severe violations of homogeneous phylogenetic models.
We used the hymenoptera-v2 ant-specific probe set (Branstetter et al., 2017b) targeting 2524 UCEs and
12 nuclear genes ("legacy" markers).
Wet lab methods
We extracted DNA from pinned and ethanol-preserved specimens using QUIAGEN DNeasy Blood &
Tissue Kits. Extractions were semi-nondestructive. In the case of pinned specimens, we first removed
them from their pins. For most specimens, we made holes in the right side of the thorax using an insect
pin, then soaked the specimen in lysis buffer overnight. We used the buffer, now containing DNA, for
subsequent extraction steps. We then washed the specimens in 95% ethanol and either dried and
remounted them or returned them to ethanol. For especially large specimens (e.g. of Megascolia) we
only used a sample of thoracic muscle for extraction. For some medium-to-large specimens that are
part of longer collection series, we separated the metasoma and the head from the mesosoma, and
soaked the mesosoma in lysis buffer overnight. In some cases, quantities of extraction reagents used
had to be proportionally adjusted to accommodate specimen size. Finally, we either reassembled the
specimen for remounting, or mounted the parts on separate points on the same pin.
We prepared, enriched, and pooled libraries using the hymenoptera-v2 ant-specific probe set following
the protocols of Faircloth et al. (2015) as modified for use at the Ward Ant Lab (Ward & Branstetter,
2017). This was done in two separate batches. High-throughput sequencing was performed at the
Huntsman Cancer Institute, University of Utah on an Illumina HiSeq 2500 platform (125 cycle paired-
end) for the first batch and at the Novogene facility in Sacramento, CA on an Illumina HiSeq 4000 for
the second batch.
Sequence quality control, assembly, and UCE identification
After receiving demultiplexed reads, we used three different bioinformatics pipelines for quality
control and de novo assembly.
Pipeline A:
We performed quality-aware 3' adapter trimming with Scythe (https://github.com/vsbuffalo/scythe)
version 0.991. This was followed with 5' adapter trimming with cutadapt (Martin, 2011) version 1.14
using a minimum overlap of 3 and an error tolerance of 0.16. We subsequently trimmed the reads with
sickle (Joshi & Fass, 2011) version 1.33 using a quality threshold of 34 and a length threshold of 50.
Assembly was done with Trinity (Grabherr et al., 2011) version 2.6.6 using a kmer size of 31. We also
generated alternative assemblies with Velvet (Zerbino & Birney, 2008) version 1.2.10 and
VelvetOptimiser (https://github.com/tseemann/VelvetOptimiser) version 2.2.4. However, the Velvet
assemblies yielded significantly fewer UCE-containing contigs (data not shown, available upon
request) and were not used for subsequent steps.
Pipeline B:
We used HTStream (https://github.com/s4hts/HTStream) version 1.1.0 for adapter and quality
trimming. The HTStream pipeline consisted of the following steps: (1) calculating basic statistics on
the raw reads with hts_Stats (2) screening for phiX with hts_SeqScreener, (3) removing polyA/T
sequences with hts_PolyATTrim with minimum size set to 100, (4) screening for adapter contamination
.CC-BY-NC-ND 4.0 International licenseavailable under a
(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprintthis version posted June 26, 2022. ; https://doi.org/10.1101/2022.01.24.474473doi: bioRxiv preprint
with hts_SeqScreener using the i5 and i7 adapter sequences corresponding to each sample, with the
kmer size set to 15, and with the percentage-hits argument set to 0.01, (5) a second round of adapter
screening with hts_AdapterTrimmer, (6) quality-based 5' and 3' trimming with hts_QWindowTrim, (7)
extracting the longest subsequences without "N"s using hts_NTrimmer with the minimum length set to
50, and finally (8) calculating statistics on the processed reads with hts_Stats. In order to speed up read
processing, we wrote a python script that can run the pipeline in parallel on more than one sample if the
number of available CPU cores is at least twice the number of steps in the pipeline.
We then assembled the reads with Spades (Bankevich et al., 2012) using a wrapper script from the
phyluce package (Faircloth, 2016), version 1.6.8. Except for increasing allowed memory usage, settings
were left at phyluce defaults.
Pipeline C:
We used Illumiprocessor (Faircloth, 2013), a wrapper around Trimmomatic (Bolger et al., 2014) and
part of the phyluce package, for adapter and quality trimming. Spades was used for de novo assembly
as in Pipeline B above.
For all pipelines, we used FastQC (Andrews, 2010) to evaluate reads before and after quality-control
procedures.
We put reads from the first sequencing batch through Pipeline A and subsequently Pipeline B, while
reads from the second sequencing batch were processed with Pipeline B and (with the exception of two
samples) Pipeline C. In the case of ingroup taxa with previously published data (Colpa sexmaculata,
Colpa alcione, Proscolia sp. EX568, Scolia hirta, Scolia verticalis, and Scoliinae sp. EX577), we used
the available assemblies and did not redo quality control and assembly. In all cases, we used the
phyluce_assembly_match_contigs_to_probes, phyluce_assembly_get_match_counts,
phyluce_assembly_get_fastas_from_match_counts, and phyluce_assembly_explode_get_fastas_file
scripts to identify UCE-containing contigs and write them to fasta files for downstream analyses.
Pipelines A and B recovered similar numbers of UCEs per sample, although Pipeline B resulted in
assemblies with higher N50 as calculated in QUAST (Gurevich et al., 2013) version 5.0.2 on both
whole assemblies and assemblies filtered to UCE-containing contigs only. Pipelines B and C were
close in terms of both number of recovered UCEs and N50. See Tables S2-3 for details. However, each
pipeline recovered some UCEs that the other pipelines did not. Therefore, we combined the assemblies,
choosing the longer contig in cases where a contig containing the same UCE was recovered in both
assemblies. However, longer contigs may either represent genuine sequence or be the result of
assembly errors. We visually inspected alignments prior to most downstream analyses to identify and
remove misaligned sequences possibly originating from misassembly.
Due to low UCE yield from some samples in the second sequencing batch (likely due to failed
enrichment) and concerns over contamination, we did the following to identify problematic samples:
(1) selected loci that were represented by > 75% of taxa, (2) aligned sequences from those loci using
MAFFT (Katoh & Standley, 2013), (3) edge-trimmed the alignments using the
phyluce_align_get_trimmed_alignments_from_untrimmed script from phyluce, and (4) estimated a
phylogeny (Fig. S1) using maximum likelihood (ML) with IQTREE (Minh et al., 2020; Hoang et al.,
2018; Chernomor et al., 2016; Nguyen et al. 2015) version 2.0-rc2 while partitioning by locus and
filtering out loci using a matched-pairs test of symmetry (Jermiin et al., 2017; Naser-Khdour et al.,
2019). Thirteen taxa associated with suspected failed enrichments clustered together in two "clades"
with very long branches, corroborating the spurious nature of the obtained sequences (Fig. S1). These
taxa were not used in subsequent phylogenetic analyses and are not included in the counts under the
taxon and locus selection section above.
.CC-BY-NC-ND 4.0 International licenseavailable under a
(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprintthis version posted June 26, 2022. ; https://doi.org/10.1101/2022.01.24.474473doi: bioRxiv preprint
In the case of Apterogyna, we mined UCE sequences from the partial genome of Johnson et al. (2013).
We aligned UCE probes to the contigs using the phyluce_probe_run_multiple_lastzs_sqlite script from
the phyluce package. We then extracted matching sequences in fasta format using the
phyluce_probe_slice_sequence_from_genomes script, setting the flanking length to 700 bases.
Phylogenetic analysis
Unless otherwise indicated, we performed all multiple-sequence alignments using MAFFT version
7.407 with the E-INS-i algorithm (Altschul, 1998). Preliminary visual inspection of alignments
confirmed that they often contain multiple conserved, well-aligned regions separated by ambiguously
aligned regions. This better conforms to the assumptions behind the E-INS-i algorithm. L-INS-i
(Gotoh, 1993), on the other hand, assumes a single, contiguous alignable region. All edge-trimming
was done using the phyluce_align_get_trimmed_alignments_from_untrimmed script. See log files (in
repository listed under the data availability section below) for parameters used. All Bayesian
phylogenetic analyses were performed using RevBayes (Höhna et al., 2014; 2016) version 1.0.12
unless otherwise indicated.
Analysis 1a:
We performed a preliminary run combining all (non-spurious) data from both sequencing batches,
including all Scolia bicincta samples. This helped inform which S. bicincta samples to retain, as
discussed below. We selected loci that had no more that 20% missing data at the site level (after
including taxa without data) and estimated a phylogeny using ML with IQTREE while partitioning by
locus and filtering out loci using a matched-pairs test of symmetry (0.05 p-value cutoff).
Analysis 1b:
We performed a second ML analysis with the goal of leveraging data from as many loci and taxa as
possible while maintaining acceptable total levels of missing data. Given that analysis 1a indicated that
samples of S. bicincta fall into two distinct clades that are sister to S. dubia and S. mexicana
respectively (Fig. 1), we removed all but two S. bicincta samples (one from each putative species). In
addition to phylogenetic position, the decision on which samples to retain was based on the number of
recovered UCEs and on assembly quality statistics calculated using QUAST. We also removed Scolia
hirta and Scoliinae sp. EX577, both from previously published studies, because they had very high
fractions of missing data. After taxon removal, we redid alignment and edge-trimming. We then sorted
loci by increasing fraction of missing data at the site level and progressively selected loci until the
cumulative fraction of missing data reached 25% (1235 loci were selected at this point). After filtering
using tests of symmetry in IQTREE, we retained 727 loci. We concatenated the alignments and selected
a substitution and across-site rate variation (ASRV) model (from a pool of substitution models from the
GTR (Tavaré, 1986) family and discretized gamma (Yang, 1994) and free-rates ASRV models) for each
locus based on Bayesian Information Criterion (BIC) (Schwarz, 1978) scores. We then estimated a
phylogeny and performed 1000 ultrafast bootstrap replicates while leaving other IQTEE settings at
default.
Analysis 1c:
In order to account for potential gene-tree-gene-tree conflict due to incomplete lineage sorting, we
estimated species trees using the program ASTRAL-MP version 1.15.1 (Yin et al., 2019). Starting with
the same set of taxa used in analysis 1b, we redid alignment and edge-trimming, discarding alignments
shorter than 600 bases. Given that highly fragmentary sequences can negatively affect accuracy
(Sayyari et al., 2017), we subsequently removed taxa with more than 50% missing data and discarded
alignments that retained fewer than 66 taxa. We then inferred gene trees using IQTREE with model
selection settings similar to those in analysis 1b above while also performing matched-pairs tests of
.CC-BY-NC-ND 4.0 International licenseavailable under a
(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprintthis version posted June 26, 2022. ; https://doi.org/10.1101/2022.01.24.474473doi: bioRxiv preprint
symmetry. We based subsequent species tree inference on three sets of gene trees: The first set
contained trees corresponding to all loci, the second contained only trees from loci that failed the
maximum test of symmetry (p-value < 0.05), and the last contained only trees from loci that passed.
Additionally, we estimated posterior distributions of gene trees in a Bayesian framework under the
GTR+G model, followed by posterior predictive simulation (Bollback, 2002; Brown, 2014b; Doyle et
al., 2015; Höhna et al., 2018) and calculation of posterior predictive p-values using two test statistics:
multinomial likelihood (Goldman, 1993; Bollback, 2002) and chi-squared (Huelsenbeck et al., 2001;
Foster 2004). Similarly to the ML-based analyses above, we then used different sets of maximum clade
credibility (MCC) gene trees and gene tree posterior distribution samples (3000 trees per gene) for
species tree inference with ASTRAL-MP. Using an alpha of 0.05 and the Bonferroni correction to
account for multiple testing, we treated loci for which the posterior predictive p-value with either test
statistic was < 0.025 as loci for which the model was likely inadequate. This set included 922 of the
total 954 loci. For each test statistic, we also split the loci into two sets, each respectively representing
loci with the lowest and highest effect sizes for that statistic. Finally, we created another similar pair of
sets but based on the Pythagorean sum of the effect sizes for both statistics. When using gene tree
posterior distribution samples with ASTRAL, we performed bootstrapping using the -b option and set
the number of replicates to 1000.
In datasets used for analyses 1a-c, Apterogyna and Proscolia have disproportionately high fractions of
missing data (49% and 73% respectively) compared to other taxa. However, removing these taxa
means the loss of the only outgroup. We therefore took a two-step approach: First, we performed an
analysis (2a) only using loci with data available from both Apterogyna and Proscolia to minimize the
potential impact of missing data on the inferred position of the root as well as on the placement of
Proscolia. However, significantly cutting down the base dataset could result in loss of resolution in
some parts of the tree. To address this, we performed another set of analyses (analyses 3a and 3b; see
below) excluding Apterogyna and Proscolia as well as loci used in analysis 2a but conditioning on the
position of the root inferred in analysis 2a. This allowed use of the remaining majority of the original
data to resolve relationships within Scoliidae.
Analysis 2a:
We started with the same taxon set as for analysis 1b and selected aligned, trimmed fasta files
corresponding to the 647 loci that have sequences from both Apterogyna and Proscolia. We used the
biclustering algorithm of Uitert et al. (2008) as implemented in the R (Core R Team, 2020) package
BicBin (https://github.com/TylerBackman/BicBin) to find large, dense biclusters of taxa and loci. We
chose a set of 68 taxa and 484 loci with >99% completeness (presence or absence of sequence for a
given taxon and locus pair treated as a binary value). We then retrieved unaligned, untrimmed fasta
files corresponding to the above loci and removed the taxa that are not part of the selected set. The
sequences were then aligned and edge-trimmed. Given that the phylogenetic models we planned to use
do not directly model indels (gaps are treated as missing data) and that unique indels are unlikely to
contribute significant information, we removed all unique indels (i.e. columns where all taxa except
one are represented by a gap) from the alignments. Calculating basic alignment statistics using AMAS
(Borowiec, 2016) and visually inspecting the alignments in AliView (Larsson, 2014) revealed that
Apterogyna sequences were (1) sometimes much shorter than those of other taxa for a given locus and
(2) sometimes had poorly aligned sections. We therefore only retained alignments containing at least
500 non-ambiguous bases for both Apterogyna and Proscolia. We then manually trimmed alignment
edges that contained no Apterogyna sequence and also trimmed any parts with suspected alignment
uncertainty while discarding alignments that were poor throughout their length. Any alignments that
became shorter than 300 bases were also discarded.
.CC-BY-NC-ND 4.0 International licenseavailable under a
(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprintthis version posted June 26, 2022. ; https://doi.org/10.1101/2022.01.24.474473doi: bioRxiv preprint
In order to assess model adequacy on the remaining 177 loci, we performed Bayesian phylogenetic
analyses under the GTR+G model followed by posterior predictive simulation on each locus
individually using the program RevBayes. We calculated the multinomial likelihood and chi-squared
(as applied to nucleotide composition across taxa) test statistics and associated posterior predictive p-
values and effect sizes on the empirical and simulated data using custom R code. For the purpose of
filtering data for which the available model is suspected of being inadequate, one must choose some
threshold. In advance of looking at the output, we decided to use an overall alpha of 0.05 and use the
Bonferroni correction to account for multiple testing. We therefore discarded loci for which the
posterior predictive p-value with either test statistic was < 0.025. We concatenated the remaining 31
alignments and used them for phylogeny estimation. Each locus was assigned a separate GTR+G
substitution model and tree length parameter (i.e. branch length multiplier), while a single vector of
branch lengths drawn from a flat Dirichlet prior was shared among partitions. See used Rev scripts for
further details. We assessed convergence for numerical parameters through visualization of posterior
samples in Tracer (Rambaut et al., 2018) version 1.7. For tree topologies, we made plots comparing
posterior probabilities of splits across both runs using the bonsai (May & Moore, 2017) version 0.9 R
package and calculated the Average Standard Deviation of Split Frequencies (ASDSF).
Analysis 2b:
Rasnitsyn (1993) identified only one fossil from Shangwang, Shandong, China as unequivocally
belonging to the scoliid crown group. This fossil was attributed by Zhang (1989) to the extant species
Scolia prismatica, currently in the genus Megacampsomeris. Yu et al. (2021) dated the Shanwang shale
to approximately 18.5 Ma, in the early Miocene. Species described in later studies (Rasnitsyn &
Martınez-Delclos, 1999; Nel et al., 2013; Zhang et al., 2015) are either connected to the crown
Scoliidae by venation characters alone, or are of uncertain placement. This limits the information
available to precisely estimate divergence times. Given this limitation and our inability to examine the
M. prismatica specimen, we chose a conservative approach and estimated a broad timeline of scoliid
evolution by calibrating the node representing the most recent common ancestor of Scoliidae and
Bradynobaenidae using the age of Protoscolia normalis, a putative stem scoliid dated to approximately
125.5 Ma (Haichun et al., 2002). We started with 177 processed alignments from analysis 2a (i.e. the
state of the dataset after removal of short alignments but prior to filtering using posterior predictive
checks). We then performed analyses on individual loci followed by posterior predictive simulation. We
used a birth-death prior on tree topologies and node ages with a scaled beta prior on the root age (125.5
Ma minimum age, 174.1 Ma maximum age, 132.5 Ma expected age, and a standard deviation of 5.5
Ma) and an uncorrelated lognormal relaxed clock model. See used Rev scripts for further details. After
filtering loci in a similar manner to what was done in analysis 2a, we concatenated and analyzed the
remaining 63 loci, adding rate multiplier parameters to allow the overall substitution rates to vary
among loci.
In addition to the conservative primary analysis, we tested the effect of calibrating additional nodes
using fossils of less certain placement. Although it is doubtful that the fossil described by Zhang (1989)
belongs to an extant species, for the first additional analysis, we used it to set an 18.5 Ma minimum age
(lognormal node age "prior" offset by 18.5, with a mean of 5.0 (mu ≈ 1.44) relative to the offset and a
sigma of 0.587405) for the Megacampsomeris clade. For the second analysis, we used both the
Megacampsomeris calibration above as well as a calibration of the scoliid crown group age based on
Araripescolia magnifica (Nel et al., 2013) (lognormal node age "prior" offset by 112.6 Ma, a mean of
10.0 relative to the offset and a sigma of 0.587405).
.CC-BY-NC-ND 4.0 International licenseavailable under a
(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprintthis version posted June 26, 2022. ; https://doi.org/10.1101/2022.01.24.474473doi: bioRxiv preprint
Analysis 2c:
In order to account for potential gene-tree-gene-tree conflict due to incomplete lineage sorting, we
performed a species tree estimation analysis under the multispecies coalescent (e.g. Rannala & Yang,
2003; Degnan & Rosenberg, 2009) using the BEAST2 (Bouckaert et al., 2019) package STACEY
(Jones, 2017). We used the same 63 loci from analysis 2b. Collapse weight was drawn from a beta prior
with an alpha of 1.0 and a beta of 19.0 (mean 0.05, to reflect the belief that most samples are likely
from distinct species). We used a lognormal prior on the popPriorScale parameter with a mean and
standard deviation (in real space) of 1.0E-6 and 2.0 respectively. We enabled estimation of the relative
death rate, which in this context corresponds to using a birth-death (as opposed to Yule) tree prior, and
used a strict clock model. The site model was set to GTR+G, unlinked among loci. We ran four
independent chains and combined and summarized the output using the logcombiner and treeannotator
tools packaged with BEAST2.
Analysis 2d:
We additionally performed species tree estimation using ASTRAL-MP. We used the same 177 starting
loci from analysis 2a, but reran Bayesian gene tree estimation and posterior predictive simulation after
removing taxa which had no data for a given locus. We then assembled sets of loci based on posterior
predictive effect sizes in a manner similar to that in analysis 1c.
Analysis 3a:
In order to leverage more data to resolve relationships within Scoliidae, we set up an analysis that
conditions on the position of the root inferred in analysis 1b while removing Proscolia and Apterogyna
from the dataset. We followed a locus and taxon selection, alignment, and trimming procedure similar
to that in analysis 2a. We chose a set of 72 taxa and 617 loci at 91% completeness from a pool of loci
that excludes those used in analysis 1b. After discarding all alignments that, after trimming, were
shorter than 300 bases or had more than 25% missing data at the site level, 469 alignments were
retained. We did not trim alignments manually at this stage as the number of loci was large and the
exclusion of Apterogyna and Proscolia improved alignment quality (assessed by visual inspection of a
subset of alignments). We then ran Bayesian phylogenetic analyses followed by posterior predictive
simulation on each individual alignment as in 2a. All alignments which passed filtering, as well as
some that did not, were visually evaluated, and in a few cases problematic regions were manually
trimmed. One locus was excluded due to very poor alignment. We reran posterior predictive tests on all
alignments that have been altered. We then performed a concatenated analysis analogous to that in 2a,
which included all loci that passed filtering and were not subsequently edited and loci which were
edited and subsequently passed filtering.
Analysis 3b:
The data processing and phylogenetic analysis procedures were analogous to those of analysis 3a,
except we used a birth-death prior on trees and node ages (with no node calibration and with the root
age arbitrarily fixed to 100 units) and an uncorrelated lognormal clock model.
Results
Sequence quality control, assembly, and UCE identification
Using pipeline A (Scythe + cutadapt + sickle + Trinity), we recovered 1941.9 UCE-containing contigs
on average across specimens from batch 1, which is almost identical to the 1943.7 UCE-containing
contigs recovered when using pipeline B (HTStream + Spades). However, the output of pipeline B had
.CC-BY-NC-ND 4.0 International licenseavailable under a
(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprintthis version posted June 26, 2022. ; https://doi.org/10.1101/2022.01.24.474473doi: bioRxiv preprint
higher average N50 (2112.6 versus 1191.4, calculated from on-target contigs only) and a higher
average number of UCE-containing contigs longer than 1000 bases (1429.0 versus 968.3).
The differences between outputs from pipelines B and C applied to specimens from batch 2 were in
some ways less pronounced. The average number of UCE-containing contigs was 1726.3 and 1785.6
for pipelines B and C respectively, while average values for N50 were 1477.8 and 1482.5 respectively.
The average number of on-target contigs longer than 1000 bases was 1036.0 for pipeline B and 1077.6
for pipeline C. When calculating these statistics, we excluded batch 2 samples for which we suspected
failed enrichment (see corresponding section under Methods for details and SI tables 1 and 2 for full
QUAST statistics).
Overall, we recovered a total of 2495 UCE loci and an average of 1883.6 UCE loci per taxon across 91
taxa (including 6 taxa from previously published studies).
Phylogenetic analysis
Analysis 1a:
A total of 176 loci were retained after all filtering steps and used to estimate a phylogeny by maximum
likelihood (Fig. 1). We recovered Proscolia as the sister group to all remaining Scoliidae, which
correspond to the subfamily Scoliinae sensu Day et al. (1981). The tribe Scoliini is monophyletic.
However, in contrast to the assumptions behind the current scoliid taxonomy (Osten, 2005), the genus
Colpa was recovered as sister to the Scoliini, rendering the Campsomerini paraphyletic.
A clade represented by the scoliine genera Megascolia, Pyrrhoscolia, and Carinoscolia is sister to all
other Scoliini, which in turn form three distinct groups. All New World members of the genus Scolia
form a clade. We recovered Scolia verticalis, an Australasian species, as sister to the morphologically
unusual Nearctic species Triscolia ardens. Given the unexpected nature of this pairing, we conducted
an additional analysis (see Supporting Information for details) using (1) the "legacy" markers enriched
from T. ardens as part of this study and from S. verticalis (from Faircloth et al. (2015), the source of S.
verticalis UCE data used in this study), (2) corresponding Sanger data from the same specimen of S.
verticals (Brady et al., 2006; Ward & Fisher, 2016), and (3) corresponding Sanger data from different
specimens of T. ardens (Pilgrim et al., 2008) and S. verticalis (Klopfstein & Ronquist, 2013).
Sequences from the specimens used in this study grouped with their corresponding sequences from
independent samples (Fig. S2), which makes contamination or data curation errors a less likely
explanation for the relationship between T. ardens and S. verticalis inferred here. All remaining
sampled Scoliini form an Old World clade that is sister to the clade consisting of New World Scolia +
(T. ardens + S. verticalis).
Samples of Scolia bicincta fall into two separate clades: one sister to Scolia mexicana and the other
sister to Scolia dubia. This suggests the two groups belong to different species.
Campsomerini minus Colpa (provisionally referred to as Campsomerini sensu stricto from here on) is
monophyletic. Trisciloa saussurei (not to be confused with members of the genus Triscolia) is inferred
to be the sister taxon to the remaining Campsomerini sensu stricto. Within the latter group, all sampled
New World taxa form a single clade. The closest relative of this New World clade is the Indomalayan
taxon Colpacampsomeris indica, followed by a clade including the Afrotropical Megameris soleata, the
Australiasian Laevicampsomeris formosa, and the Indomalayan genus Megacampsomeris.
Megacampsomeris itself is recovered as monophyletic. Taxa occurring in Madagascar, such as
Micromeriella pilosella and some Campsomeriella, have their closest affinities with Afrotropical taxa
but do not form a monophyletic group.
.CC-BY-NC-ND 4.0 International licenseavailable under a
(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprintthis version posted June 26, 2022. ; https://doi.org/10.1101/2022.01.24.474473doi: bioRxiv preprint
Analysis 1b:
We used 727 loci from 76 taxa to reconstruct the tree in Fig. 2. The results are largely congruent with
those from analysis 1a above, with the exception of the Triscolia ardens + Scolia verticalis group being
recovered as sister to the Old World Scoliini (minus Megascolia + Pyrrhoscolia + Carinoscolia) as
opposed to sister to the New World Scolia. Colpa is still recovered as sister to the Scoliini. The non-
monophyly of Dielis, due to Dielis pilipes being more closely related to Xanthocampsomeris than to
other Dielis, is likewise corroborated.
Analysis 1c:
For all analyses, the topology of the "main" ASTRAL tree (based only on ML or MCC gene trees) was
effectively the same as the consensus topology estimated using gene tree posterior distributions and
bootstrapping. Differences were limited to quadripartitions with very low support (e.g. 0.46 local
posterior probability for most probable resolution, versus 0.35 for the next most probable alternative) or
to relationships within species (e.g. Dielis plumipes).
The inferred topology based on ML trees from all loci (Fig. 3C) agrees with that from analysis 1b
above. The topology based only on loci not failing the maximum test of symmetry (Fig. 4, Fig. 3A) is
identical, but with reduced support for the quadripartition involving Megameris soleata,
Laevicampsomeris formosa + Megacampsomeris, Colpacampsomeris indica + New World
Campsomerini, and the remaining Campsomerini. The topology inferred from loci failing the symmetry
test (Fig. 3B) maintained high support for this quadripartition. On the other hand, the position of
Triscolia ardens + Scolia verticalis became more uncertain, with 0.50 local posterior probability for the
same placement as the other analyses above and 0.30 local posterior probability for Triscolia ardens +
Scolia verticalis being sister to the New World Scolia.
Results of the analysis using MCC trees (as a way of summarizing tree posterior distributions) from all
loci (Fig. 5D) agree with the ML-based results above with respect to the Campsomerini sensu stricto.
However, the placement of Triscolia ardens + Scolia verticalis is not resolved, with 0.47 and 0.46 local
posterior probability for a sister relationship with the sampled Old World Scolia and with the New
World Scolia respectively. The ASTRAL tree based on loci with the lowest combined posterior
predictive effect sizes (Fig. 5A) is similar to the tree above, with 0.46 local posterior probably in favor
of (Triscolia ardens + Scolia verticalis) + New World Scolia, but a slightly lower probability (0.36) in
favor Triscolia ardens + Scolia verticalis being sister to the Old World Scolia. The analysis of loci with
highest combined posterior predictive effect sizes (Fig. 5B) resulted in stronger (0.82 local posterior
probability) support for the (Triscolia ardens + Scolia verticalis) + New World Scolia hypothesis.
Unexpectedly, this relationship was likewise supported (0.87 local posterior probability) when using
only the 32 loci for which the model was not found to be inadequate (Fig. 5C) using posterior
predictive checks, but resolution within the Campsomerini was significantly reduced. Crucially, all
analyses agree with respect to the placement of Colpa as sister to the Scoliini.
Analysis 2a:
The tree in Fig. 6 is the Maximum A Posteriori (MAP) tree summarized from two independent runs
based on 31 loci for which the model was not found to be inadequate. The MCMC exhibited good
convergence with respect to topology (see Fig. 7A for a comparison of split frequencies between runs).
The average standard deviation of split frequencies was approximately 0.001.
This analysis places emphasis on reducing missing data in the outgroup and in Proscolia, removing
poorly aligned sites, and reducing potential model violation at the expense of dataset size. Despite this,
the tree backbone is fully resolved, with only a few shallow nodes having lower posterior probabilities.
With respect to the position of the root, the results corroborate those from analyses 1a, 1b, and 1c:
.CC-BY-NC-ND 4.0 International licenseavailable under a
(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprintthis version posted June 26, 2022. ; https://doi.org/10.1101/2022.01.24.474473doi: bioRxiv preprint
Proscolia is sister to the Scoliinae, Colpa is sister to the Scoliini, and Campsomerini sensu stricto is
sister to Scoliini + Colpa, with Campsomerini in the traditional sense being non-monophyletic. The
position of Triscolia as sister to an Old World scoliine clade is congruent with that in analysis 1b but
not analysis 1a. Scolia verticalis, which was recovered as sister to Tricolia ardens in previous analyses,
was not represented here and in subsequent analyses due to a high proportion of missing data. While
Colpacampsomeris indica was likewise excluded from this analysis for the same reason, Megameris
soleata is placed as sister to the New World Campsomerini instead of being sister to Laevicampsomeris
+ Megacampsomeris as in analyses 1a and 1b.
Analysis 2b:
A total of 63 loci were retained post-filtering and used to construct a chronogram (Fig. 8). See Fig. 7B
for a plot of split frequencies from two independent runs. While most clade posterior probabilities are
close to 1 and none are lower than 0.94, node age credible intervals are broad due to only one
calibration point being available. The crown Scoliini are inferred to have likely originated after the
Cretaceous-Paleogene (K-Pg) extinction event. The mean estimated crown ages of Campsomerini
sensu stricto and of Scoliini + Colpa are 49 million years (Ma) and 58 Ma respectively, although the
associated 95% highest posterior density (HPD) intervals extend past the K-Pg boundary. The mean
estimated age of crown Scoliinae is 84 Ma, with lower and upper bounds of the 95% HPD interval at
56 Ma and 107 Ma respectively. The crown Scoliidae as a whole (and thus the split between
Proscoliinae and Scoliinae) has a 95% HPD age interval bounded by 96 Ma and 145 Ma, placing the
likely origin of the group in the Early Cretaceous.
Results from the analyses including additional fossil calibrations (Fig. 9-10) were broadly congruent
with the results above, but with greater ages estimated for most nodes after the Scoliinae/Proscoliinae
split. When using both additional calibrations, the posterior distributions of ages for Campsomerini
sensu stricto and of Scoliini + Colpa had means of 63 Ma and 69 Ma respectively, with more posterior
mass on pre-K-Pg ages compared to the more conservative analysis above.
There are some topological differences between the results of these analyses and the tree from analysis
2a, mostly in the relationships of Old World Scolia and the position of Megameris soleata as sister to
(Laevicampsomeris + Megacampsomeris) + New World Campsomerini sensu stricto. However, both
sets of analyses agree on the placement of Colpa as sister to the Scoliini and of Triscolia ardens as
sister to the Old World Scolia clade.
Analysis 2c:
The species or minimal clusters (SMC) tree inferred under the multispecies coalescent using STACEY
(Fig. 11-12) recovered many of the same major clades as the other analyses. However, some
relationships, particularly those that had conflicting resolutions among the previous analyses, were
poorly resolved. Specifically, while Colpa is still sister to a monophyletic Scoliini and the Megascolia
+ Pyrrhoscolia + Carinoscolia clade is sister to all other Scoliini, the position of Triscolia ardens
within the latter group is uncertain. Trisciloa is still sister to all other members of Campsomerini sensu
stricto, the New World members of which form a monophyletic group. Megameris soleata,
Laevicampsomeris + Megacampsomeris, and the New World Campsomerini sensu stricto form a clade,
though the relationships among them is uncertain. Likewise, the relationships among this clade, the
Cathimeris + Micromeriella clade, and the Campsomeriella + Tristimeris clade are not resolved.
Analysis 2d:
The "main" ASTRAL topology, estimated using MCC trees only, was mostly congruent with the
consensus topology, estimated using posterior samples and bootstrapping, in the case of the dataset
with all loci (Fig. 13D) and of the dataset with loci having the highest-third combined posterior
.CC-BY-NC-ND 4.0 International licenseavailable under a
(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprintthis version posted June 26, 2022. ; https://doi.org/10.1101/2022.01.24.474473doi: bioRxiv preprint
predictive effect sizes (Fig. 13C), with a few differences in the resolution of shallow nodes with low
support. The dataset with loci having the lowest posterior predictive effect sizes showed somewhat
bigger differences between the "main" (Fig. 13A) and bootstrap consensus (Fig. 13B) topologies, the
"main" topology notably placing Proscolia as sister to the Campsomerini sensus stricto, albeit with low
support.
The topology inferred using all loci mostly agrees with the results of analysis 2a, 2b, and 2c with the
exception of Megameris soleata being inferred to be more closely related to Laevicampsomeris and
Megacampsomeris than to the New World Campsomerini clade. Additionally, Triscolia ardens is
placed as sister to the New World Scoliini, as opposed to being sister to the Old World Scolia clade as
in analyses 2a and 2b and its position being unresolved as in analysis 2c. Analysis of the subset of loci
with the highest combined posterior predictive effect sizes produced results almost identical to those
based on all loci. Conversely, as reported above, using loci with the lowest posterior predictive effect
sizes resulted in the unexpected placement of Proscolia as sister to Campsomerini sensus stricto.
Relationships were otherwise similar to those inferred using other locus sets, but with lower local
posterior probabilities associated with many quadripartitions.
Analyses 3a and 3b:
Analyses 3a and 3b are based on data from 115 and 159 loci respectively. The results (Fig. 14-15) agree
with each other and mostly agree with those from analysis 1b. Differences include Triscolia ardens
being sister to the Old World scoliine clade and Megameris soleata being sister to Laevicampsomeris
formosa.
Discussion
Phylogenetic results and taxonomic implications
This is the first study to use molecular data to reconstruct the mammoth wasp phylogeny. Our results
corroborate some long-standing phylogenetic hypotheses originally based on morphological data while
contradicting others. Scoliid taxonomy has historically been unstable and confusing (see Elliott (2011)
and Kimsey & Brothers (2016) for commentary). In the following discussion, we use Osten (2005) as
the reference for the current status of taxon names unless otherwise specified. We use Campsomerini
sensu stricto to refer to Campsomerini excluding Colpa and taxa more closely related to Colpa than to
the Scoliini.
The genus Proscolia was originally described by Rasnitsyn (1977), hypothesized to be sister to the
remaining extant Scoliidae, and placed in a new subfamily Proscoliinae, with the other extant Scoliidae
relegated to the Scoliinae. Day et al. (1981) and Osten (2005) maintained this arrangement and treated
the former subfamilies Scoliinae and Campsomerinae as the scoliine tribes Scoliini and Campsomerini
respectively (Fig. 16C). Notable exceptions to this approach include earlier works by Osten (1988,
1993), where he argued against the inclusion of Proscolia in the Scoliidae, and Argaman (1996), who
radically revised the higher-level scoliid taxonomy without conducting an explicit phylogenetic
analysis. Argaman elevated the Campsomerini (minus Colpa and its presumed close relatives) back to
subfamily rank (Fig. 16D) and placed it as sister to the remaining extant Scoliidae (including the
Proscoliinae). Pilgrim et al. (2008) included three scoliids in their study and placed Proscolia as either
sister to the other two scoliids or as sister to Bradynobaenidae + other Scoliidae. Two more recent
molecular phylogenetic studies of aculeates that included five and three scoliid species respectively
(Debevec et al., 2012; Branstetter et al., 2017a) placed Proscolia as sister to all other scoliids. All
analyses in the present study (Fig. 16E) strongly support this placement.
.CC-BY-NC-ND 4.0 International licenseavailable under a
(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprintthis version posted June 26, 2022. ; https://doi.org/10.1101/2022.01.24.474473doi: bioRxiv preprint
The taxonomic treatment of the species currently comprising the genus Colpa has historically varied
significantly. To date, none of the taxonomic changes have been supported by phylogenetic analyses.
However, the following authors generally presented informal phylogenetic arguments when making
taxonomic decisions. Bradley (1950a), using the name Campsoscolia for the genus including what is
now Colpa and Dasyscolia, argued for a "basal" placement of these taxa, presumably meaning they fall
outside the clade formed by the remaining Scoliidae (Fig. 16A). Betrem (1965) erected the tribe
Trielini (emended by Betrem & Bradley (1972) to Trielidini) within the Campsomerinae (Fig. 16B) to
contain the genera Trielis (corresponding to Campsoscolia as used by Bradley (1950a) and currently
understood (Day et al., 1981) to be a junior synonym of Colpa), Crioscolia (currently treated as a
subgenus of Colpa), and Guigliana, which was formally described later by Bradley & Betrem (1967).
Following the demotion of Campsomerinae to tribe rank by Day et al. (1981), Colpa and its allies were
kept within the Campsomerini (Fig. 16C), with the implied relationships being Proscoliinae +
(Campsomerini + Scoliini). Argaman (1996) on the other hand, created a new subfamily Colpinae
(corresponding to the Trielidini of Betrem and Bradley (1972)) and placed it as sister to the Scoliini
(which he elevated to subfamily rank), concluding that the Campsomerini sensu stricto (also elevated
to subfamily rank) is sister to Proscoliinae + (Colpinae + Scoliinae) (Fig. 16D).
Debevec et al. (2012) included five scoliid species in their analyses, one of them being Colpa
sexmaculata, but the main text contains no discussion of Colpa and the relationships within the
Scoliidae. If we assume the monophyly of Campsomerini sensu stricto and of Colpa (each only
represented by one species), the phylogenies included with the supporting information place
Proscoliinae as sister to Campsomerini sensu stricto + (Colpa + Scoliini). All analyses in the current
study agree with the latter hypothesis (Fig. 16E) while using a significantly larger dataset and
attempting to mitigate the effects of non-randomly-distributed missing data and phylogenetic model
violation.
In light of these results, morphological similarities between Colpa and the Campsomerini, such as the
presence of an articulation between the basal and apical parts of the volsella and the presence of the
second recurrent vein, are likely plesiomorphies. We recommend the exclusion of Colpa from
Campsomerini when a formal taxonomic revision of Scoliidae is undertaken. However, a phylogenetic
analysis establishing the positions of Guigliana and Dasyscolia (not represented in this study) should
be considered a prerequisite of such a revision. Both genera lack the transverse impressed impunctate
band on the frons, which serves as the defining feature of Colpa, but share with Colpa and the Scoliini
some mesothoracic characters (Bradley, 1950a; Betrem & Bradley, 1972). If Guigliana and Dasyscolia
form a monophyletic group with Colpa, the establishment of a tribe Colpini may be justified.
Otherwise, if they are more closely related to or nested within the Scoliini, it may be reasonable to
transfer Colpa, Guigliana, and Dasyscolia to that tribe. More complete sampling of this group would
also allow the evaluation of its subgeneric classification. The subgenus Colpa (Crioscolia) has a
strongly disjunct distribution in both the New and Old World (Bradley, 1950a). Our results (Fig. 2)
indicate the paraphyly of Colpa (Colpa): the Nearctic Colpa (Colpa) octomaculata is more closely
related to the Nearctic Colpa (Crioscolia) alcione than it is to the Palearctic Colpa (Colpa)
sexmaculata. In addition to allowing a critical evaluation of the phylogenetic validity of Colpa
subgenera, a molecular phylogeny including more Colpa species would contribute significant
biogeographic information, as this group appears to have undergone dispersal and/or vicariance events
between the Old World and the Americas independently of the Scoliini and the Campsomerini sensu
stricto.
Campsomerini sans Colpa is inferred to be monophyletic in all our analyses, with Trisciloa always
sister to the remaining members of the group. Likewise, all sampled New World Campsomerini sensu
stricto form a clade with high support in all analyses. Colpacampsomeris indica is consistently inferred
.CC-BY-NC-ND 4.0 International licenseavailable under a
(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprintthis version posted June 26, 2022. ; https://doi.org/10.1101/2022.01.24.474473doi: bioRxiv preprint
to be the closest relative of this New World clade in all analyses in which the former was included.
However, we have not sampled any species from South America, so it remains unknown whether those
share a closer relationship with the New World taxa sampled here or with Old World scoliids. Dielis
pilipes groups with Xanthocampsomeris as opposed to with other Dielis. This is consistent with D.
pilipes lacking some prominent morphological characteristics shared by other Dielis, such as a medial
longitudinal furrow on the clypeus and a deep transverse furrow on the anterior of abdominal sternum
II. Bradley (1964) states the opinion that D. pilipes should be excluded from Dielis, but this change was
never formalized.
Megacampsomeris is always monophyletic in these analyses. Other consistently monophyletic groups
include (1) Micromeriella with Cathimeris as the sister taxon and (2) the group consisting of
Campsomeriella, Tristimeris, and some Malagasy species (undescribed or of uncertain taxonomic
placement, provisionally labeled Campsomeriella sp. in the figures). Both Tristimeris and the Malagasy
specimens are nested within Campsomeriella.
The positions of Megameris and Laevicampsomeris are uncertain, though they are likely more closely
related to the New World Campsomerini, Megacampsomeris, and Colpacampsomeris than to other
Campsomerini. In the current study, they are each represented by only one species. More thorough
taxon sampling within these two genera will likely result in less uncertainty regarding their placement.
All analyses conducted here strongly support the monophyly of Scoliini. The first split within the
Scoliini gives rise to two clades: one consisting of Megascolia, Pyrrhoscolia, and Carinoscolia and the
other consisting of Scolia and Triscolia. Megascolia is consistently non-monophyletic in our analyses.
The situation warrants a taxonomic revision, though it should ideally be informed by future
phylogenetic studies that are able to sample Megascolia, Pyrrhoscolia, and Carinoscolia more
completely. Sequencing of multiple Carinoscolia species is especially important, given that the genus
is suggested to be polyphyletic by Golfetti (2019).
Our sampling of New World species was restricted to the Nearctic, and the affinities of Neotropical
scoliines thus remain uncertain. However, all sampled Nearctic Scolia form a single clade. The
phylogenetic position of Triscolia ardens was inconsistent across our analyses. The genus Triscolia has
a complicated taxonomic history (see Betrem & Bradley, 1964) and currently includes only two
Nearctic species, T. badia and T. ardens. In all phylogenies where Scolia verticalis is included, T.
ardens and S. verticalis are sisters. This is somewhat surprising given that S. verticalis is an
Australasian species. We have mostly ruled out contamination and misidentification (see results section
above) as potential explanations. More thorough sampling of scoliines from Australasia, Southeast
Asia, and the eastern Palearctic might reveal species related to S. verticalis and fill in the gap in
distributions, making a relationship with the Nearctic fauna more plausible. It is also possible that the
two species of Triscolia are the only extant representatives of a previously more widespread lineage.
The lack of close relatives of either species in the present study also means they are both subtended by
long branches. A combination of the potential for long branch attraction and the disproportionately high
fraction of missing data from S. verticalis raises the suspicion that the pairing might be artefactual.
Regardless of its relationship to S. verticalis, T. ardens is recovered in our analyses either as closely
related to the Nearctic Scolia clade or to the Old World Scolia clade, making it likely that the genus
Scolia is paraphyletic irrespective of which placement of T. ardens is correct. One potential course of
action is to synonymize Triscolia with Scolia. However, any taxonomic decisions involving Scolia
should take into account the phylogenetic positions of two other large Scoliine genera, Liacos and
Austroscolia, both of which are not represented in the current study.
Given the proliferation of scoliid generic names attached to groups defined mainly by superficial
characters such as color and punctation, it seems likely that there are many examples of distinctive
.CC-BY-NC-ND 4.0 International licenseavailable under a
(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprintthis version posted June 26, 2022. ; https://doi.org/10.1101/2022.01.24.474473doi: bioRxiv preprint
groups within larger genera being given their own generic names, thus rendering the larger genera
paraphyletic. Further phylogenetic studies with more complete taxon sampling are needed before a
taxonomic revision of scoliid genera is attempted. In the absence of such studies, we recommend
proceeding cautiously when describing new species (such as those belonging to the Malagasy scoliid
fauna) and avoiding the establishment of new genera or groups of higher rank without first conducting
thorough phylogenetic analyses.
Divergence times and biogeography
The precision of node age estimates in the current study is limited by the small number of fossils that
can be reliably attributed to the scoliid crown. It might be possible to slightly increase precision by
conducting analyses with a broader phylogenetic scope. Including taxa from the Apoidea and
Formicoidea could allow fossil data from those clades to inform overall rates of molecular evolution.
However, apoids and formicoids being much more diverse than scoliids makes it difficult to sample
species evenly across clades, and care must be taken to accommodate for this in any attempted
analyses. The increased likelihood of heterogeneity in the evolutionary process becoming problematic
as one expands the scope of the analysis should also be considered and addressed. Ultimately, the
discovery and description of well-preserved crown fossils is likely to be a necessary prerequisite to
achieving scoliid divergence time estimates with better precision and accuracy.
Due to weak sampling from some biogeographic regions, particularly Australasia and the Neotropics,
we did not conduct a formal phylogeographic analysis. However, our phylogenetic results do indicate
some biogeographic patterns that could be further investigated in future studies.
We estimated the stem age of the Nearctic Campsomerini sensu stricto clade to be between 19 Ma and
46 Ma (95% HPD interval) when calibrating the root age only (Fig. 8). Among taxa sampled in this
study, the closest relatives of this clade are taxa from Indomalaya, Australasia, and the eastern
Palearctic. This suggests a possible exchange of fauna across Beringia during the Oligocene or later
Eocene, which is broadly consistent with patterns observed in other animal groups (Jiang et al., 2019).
The Nearctic Scolia clade has a very similar estimated stem age (19-50 Ma). Analyses using additional
(but less reliable, in terms of the phylogenetic placement of the associated fossil) calibrations extend
the age 95% credible intervals into the early Eocene. Further refinement of node age estimates, in
conjunction with more complete geographic sampling, is needed to evaluate the possibility of late (c.
65 Ma) exposures of the Thulean Route (Brikiatis, 2014) contributing to scoliid dispersal.
The phylogenetic position of Triscolia is uncertain, and has implications for the number and timing of
biotic interchanges between North America and other regions. In addition to resolving the position of
Triscolia, future phylogenetic studies need to prioritize sampling of the South American and
Australasian scoliids. It is currently unclear whether South American Scoliini and Campsomerini sensu
stricto each represent single lineages or multiple lineages with different biogeographic origins. It is
possible that South America harbors relatively young lineages originating from Africa or the Nearctic
and dispersing into South America during the Late-Early Eocene or later (Hoffmeister, 2020) and/or
more ancient lineages with possible relationships to the Australian and African fauna. Understanding
the phylogenetic and biogeographic affinities of South American scoliids, while interesting in itself, is
also essential to understanding patterns of scoliid diversification and answering questions such as why
the Campsomerini sensu stricto are significantly more diverse than the Scoliini in the New World
tropics while the opposite pattern holds in the Afrotropic and Indomalaya (Bradley, 1950b; 1959).
Madagascar is home to members of at least two campsomerine lineages, represented in the current
study by one species of Micromeriella and several samples (probably from currently undescribed
species) falling within the Campsomeriella clade. The presence of M. pilosella is probably due to a
very recent dispersal from mainland Africa, while the Malagasy Campsomeriella lineage is older but
.CC-BY-NC-ND 4.0 International licenseavailable under a
(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprintthis version posted June 26, 2022. ; https://doi.org/10.1101/2022.01.24.474473doi: bioRxiv preprint
also most closely related to African species. Given that the Malagasy scoliid fauna has received much
less study than that of mainland Africa, it is certainly possible that among the species not sampled in
this study there exist representatives of older endemic lineages that are not closely related to either
Micromeriella or Campsomeriella. Our study additionally included two (probably undescribed) species
belonging to Scolia. The scoliine genera Liacos and Autroscolia both have representatives on the
African mainland, Madagascar, Asia, and Australia (Bradley, 1950b; Osten, 2005; Elliott 2011), while
the morphologically distinctive Mutilloscolia is confined to Madagascar (Bradley, 1959). None are
included in this study and their phylogenetic relationships to other scoliines remain poorly understood.
Although it is possible that these genera could be nested within Scolia (which is mostly identified by
lacking the defining characters of other genera), the apparent lack of morphological characters uniting
them specifically with the Scolia species sampled here suggests that the current Malagasy scoliine
diversity is likely a result of multiple dispersal (and possibly vicariance) events. This is tentatively
supported by the morphology-based phylogenies of Golfetti (2019), which place Austroscolia and
Liacos outside the clade formed by all scoliini sampled in this study.
Methodological considerations
Doyle et al. (2015) demonstrated the potential utility of filtering data using posterior predictive
methods. We made use of a similar approach, albeit limiting it to data-based (Huelsenbeck et al., 2001)
as opposed to inference-based (Brown, 2014a) tests. Molloy & Warnow (2018) used a simulation-based
approach to explore the effect of data filtering using various criteria on species tree inference using
ASTRAL (among other methods). They found that excluding loci with high gene tree estimation error
can improve the accuracy of species tree inference when levels of incomplete lineage sorting (ILS)
were moderate to low. The dependence on ILS levels was explained in terms of the number of gene
trees required to accurately reconstruct the species tree increasing with higher levels of ILS. Thus, the
negative effect of using fewer genes sometimes outweighed the positive effect of more accurate gene
trees (Molloy & Warnow, 2018). In this context, we make the following observations based on our
empirical analyses:
Using posterior predictive p-values with "conventional" cutoffs (e.g. 0.05) resulted in the exclusion of
the majority of available loci. In some cases (e.g. Fig. 13A), this led to an unexpected and implausible
species tree topology resulting from ASTRAL analyses (i.e. placement of Proscolia as sister to the
Campsomerini). This could be a result of too few loci being used. Additionally, one would expect a
correlation between the amount of data and the ability to detect model inadequacy, which might lead to
the retention of less "informative" loci. This appears to be borne out in analysis 2d, where mean
pairwise Robinson-Foulds distances among posterior topology samples were on average higher (73.3
versus 53.8) for the third of loci having the lowest posterior predictive effect sizes compared to the
third having the highest. Under these circumstances, a fully-resolved point estimate of the topology
might be a worse representation of the gene tree posterior distribution, and variance among gene tree
point estimates might be higher, even if there is no ILS and the underlying posterior distributions are
unbiased. This could explain why we observed generally lower quadripartition support values resulting
from analyses of loci with lower posterior predictive effect sizes even when the number of loci per
analysis was kept constant (Fig. 5A, B; Fig. 13A, C). In contrast to Fig. 13A, a bootstrap-based
ASTRAL species tree (Fig. 13B) that used samples from the posterior distributions of gene trees (as
opposed to point estimates) recovered Proscolia in a more plausible position that is also corroborated
by our STACEY analysis. Mirarab (2019) observed that using samples from gene tree posteriors does
not have the same negative effect on species tree accuracy as does using gene tree bootstrap replicates
in a ML framework. We concur with Mirarab (2019) that further investigation is warranted. Potential
use cases for this hybrid approach could be datasets with both (1) a limited number of genes available
(e.g. from Sanger data) where the accuracy of estimates using ASTRAL with gene-tree point estimates
.CC-BY-NC-ND 4.0 International licenseavailable under a
(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprintthis version posted June 26, 2022. ; https://doi.org/10.1101/2022.01.24.474473doi: bioRxiv preprint
may be lower and (2) with a very large number of terminal taxa where a fully Bayesian approach
(model-based coestimation of gene trees and species tree) may be more computationally challenging.
Data availability
Raw sequences are available at: TBD
Other data (assemblies, multiple sequence alignments, etc.) and raw output files that support the
findings of this study are available at: TBD
Scripts used for analyses are available at: TBD
Conflicts of interest
The authors declare that there are no conflicts of interest.
Acknowledgments
We would like to thank M. Hauser (California Department of Food and Agriculture), S.L. Heydon
(Bohart Museum of Entomology, University of California, Davis), and K. Williams (California
Department of Food and Agriculture) for providing access to scoliid specimens at their respective
institutions; B.E. Boudinot, M. Hauser, C. Parker, and T. Zavortink for giving access to specimens in
their personal collections; former members of the Ward Ant Lab, University of California, Davis (UC
Davis) M. Borowiec, B.E. Boudinot, and M. Prebus and A. Abrieux from the Chiu Lab (UC Davis) for
training the lead author in wet lab techniques; B. Moore (UC Davis) for mentoring the lead author in
phylogenetic methods; G. Attardo, J. Bond, J. Chiu, B. Johnson, S. Nadler, and P.S. Ward for the use of
their respective labs and equipment at UC Davis; L. Smith (Evolutionary Genetics Lab, Museum of
Vertebrate Zoology, University of California, Berkeley) for providing access to a sonicator and training
the lead author in its use; B. Boudinot and C. Pagan (Nadler Lab, UC Davis) for assistance with library
preparation and enrichment; former FARM HPC cluster sysadmins B. Broadley and T. Thatcher for
help with using the cluster; current and former members of the Ward Ant Lab M. Borowiec, B.E.
Boudinot, Z. Griebenow, Z. Lieberman, J. Oberski, M. Prebus, and P.S. Ward and of the Moore Lab
(UC Davis) E. Espejo, J. Gao, M.R. May, and B. Moore for helpful and insightful discussion; N. Tam
for proofreading and providing comments on the manuscript.
References
Altschul, S. F. (1998). Generalized affine gap costs for protein sequence alignment. Proteins:
Structure, Function, and Bioinformatics, 32(1), 88-96.
Andrews S. (2010). FastQC: a quality control tool for high throughput sequence data. URL:
http://www.bioinformatics.babraham.ac.uk/projects/fastqc/
Argaman, Q. (1996). Generic synopsis of Scoliidae (Hymenoptera, Scolioidea). Annales Historico
Naturales-Musei Nationalis Hungarici, 88, 171-222.
Bankevich, A., Nurk, S., Antipov, D., Gurevich, A. A., Dvorkin, M., Kulikov, A. S., ... & Pevzner, P. A.
(2012). SPAdes: a new genome assembly algorithm and its applications to single-cell sequencing.
Journal of computational biology, 19(5), 455-477.
Betrem, J. G. (1965). The African Scoliids and their Affinities. In XIIth International Congress of
Entomology (p. 120).
Betrem, J. G., & Bradley, J. C. (1964). Annotations on the genera Triscolia, Megascolia and Scolia
(Hymenoptera, Scoliidae). Zoologische Mededelingen, 39(43), 433-444.
.CC-BY-NC-ND 4.0 International licenseavailable under a
(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprintthis version posted June 26, 2022. ; https://doi.org/10.1101/2022.01.24.474473doi: bioRxiv preprint
Betrem J. G. & Bradley J. C. (1972) The African Campsomerinae (Hymenoptera, Scoliidae).
Nederlandse Entomologische Vereniging, Amsterdam.
Bollback, J. P. (2002). Bayesian model adequacy and choice in phylogenetics. Molecular Biology and
Evolution, 19(7), 1171-1180.
Bolger, A. M., Lohse, M., & Usadel, B. (2014). Trimmomatic: a flexible trimmer for Illumina sequence
data. Bioinformatics, 30(15), 2114-2120.
Borowiec, M. L. (2016). AMAS: a fast tool for alignment manipulation and computing of summary
statistics. PeerJ, 4, e1660.
Bouckaert, R., Vaughan, T. G., Barido-Sottani, J., Duchêne, S., Fourment, M., Gavryushkina, A., ... &
Drummond, A. J. (2019). BEAST 2.5: An advanced software platform for Bayesian evolutionary
analysis. PLoS computational biology, 15(4), e1006650.
Bradley, J. C. (1950a). The most primitive Scoliidae. Revista Española de Entomología (tomo
extraordinario), 1, 427-438.
Bradley, J. C. (1950b). The Ethiopian Scoliidae. In Eighth International Congress of Entomology (p. 1).
Bradley, J. C. (1959). The Scoliidae of Africa. Annals of the Transvaal Museum, 23(4), 331-362.
Bradley, J. C. (1964). Further notes on the American taxa of Campsomeris (Hymenoptera: Scoliidae).
Entomological News, 25, 101-108.
Bradley, J. C., & Betrem, J. G. (1967). The types of the Scoliidae described by Frederick Smith
(Hymenoptera). Bulletin of The British Museum (Natural History) Entomology, 20(7), 289.
Brady, S. G., Schultz, T. R., Fisher, B. L., & Ward, P. S. (2006). Evaluating alternative hypotheses for
the early evolution and diversification of ants. Proceedings of the National Academy of Sciences,
103(48), 18172-18177.
Branstetter, M. G., Danforth, B. N., Pitts, J. P., Faircloth, B. C., Ward, P. S., Buffington, M. L., ... &
Brady, S. G. (2017a). Phylogenomic insights into the evolution of stinging wasps and the origins of
ants and bees. Current Biology, 27(7), 1019-1025.
Branstetter, M. G., Longino, J. T., Ward, P. S., & Faircloth, B. C. (2017b). Enriching the ant tree of life:
Enhanced UCE bait set for genome‐scale phylogenetics of ants and other Hymenoptera. Methods in
Ecology and Evolution, 8(6), 768-776.
Brikiatis, L. (2014). The De Geer, Thulean and Beringia routes: key concepts for understanding early
Cenozoic biogeography. Journal of Biogeography, 41(6), 1036-1054.
Brown, J. M. (2014a). Detection of implausible phylogenetic inferences using posterior predictive
assessment of model fit. Systematic biology, 63(3), 334-348.
Brown, J. M. (2014b). Predictive approaches to assessing the fit of evolutionary models. Systematic
biology, 63(3), 289-292.
Burmeister, H. (1854). Bemerkungen über den allgemeinen Bau und die Geschlechtsunterschiede bei
den Arten der Gattung Scolia Fabr. Abhandlungen der Naturforschenden Gesellschaft zu Halle, 1(4), 1-
46.
Chernomor, O., Von Haeseler, A., & Minh, B. Q. (2016). Terrace aware data structure for
phylogenomic inference from supermatrices. Systematic biology, 65(6), 997-1008.
.CC-BY-NC-ND 4.0 International licenseavailable under a
(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprintthis version posted June 26, 2022. ; https://doi.org/10.1101/2022.01.24.474473doi: bioRxiv preprint
Ciotek, L., Giorgis, P., Benitez-Vieyra, S., & Cocucci, A. A. (2006). First confirmed case of
pseudocopulation in terrestrial orchids of South America: pollination of Geoblasta pennicillata
(Orchidaceae) by Campsomeris bistrimacula (Hymenoptera, Scoliidae). Flora-Morphology,
Distribution, Functional Ecology of Plants, 201(5), 365-369.
Clausen, C. P. (1940). Entomophagous insects. McGraw-Hill book Company, Incorporated.
Cock, M. J., Murphy, S. T., Kairo, M. T., Thompson, E., Murphy, R. J., & Francis, A. W. (2016). Trends
in the classical biological control of insect pests by insects: an update of the BIOCAT database.
BioControl, 61(4), 349-363.
Core R Team (2020). R: A language and environment for statistical computing. R Foundation for
Statistical Computing, Vienna, Austria. URL: https://www.R-project.org/
Day, M. C., Else, G. R., & Morgan, D. (1981). The most primitive scoliidae (Hymenoptera). Journal of
natural History, 15(4), 671-684.
DeBach, P. (1964). Biological control of insect pests and weeds. Reinhold Publishing Corporation,
New York.
Debevec, A. H., Cardinal, S., & Danforth, B. N. (2012). Identifying the sister group to the bees: a
molecular phylogeny of Aculeata with an emphasis on the superfamily Apoidea. Zoologica scripta,
41(5), 527-535.
Degnan, J. H., & Rosenberg, N. A. (2009). Gene tree discordance, phylogenetic inference and the
multispecies coalescent. Trends in ecology & evolution, 24(6), 332-340.
Doyle, V. P., Young, R. E., Naylor, G. J., & Brown, J. M. (2015). Can we identify genes with increased
phylogenetic reliability?. Systematic biology, 64(5), 824-837.
Elliott, M. G. (2011). Annotated catalogue of the Australian Scoliidae (Hymenoptera). Technical
Reports of the Australian Museum, Online, 22, 1-17.
Faircloth, B. C. (2013) Illumiprocessor: a trimmomatic wrapper for parallel adapter and quality
trimming. URL: http://dx.doi.org/10.6079/J9ILL
Faircloth, B. C. (2016). PHYLUCE is a software package for the analysis of conserved genomic loci.
Bioinformatics, 32(5), 786-788.
Faircloth, B. C., Branstetter, M. G., White, N. D., & Brady, S. G. (2015). Target enrichment of
ultraconserved elements from arthropods provides a genomic perspective on relationships among H
ymenoptera. Molecular ecology resources, 15(3), 489-501.
Faircloth, B. C., McCormack, J. E., Crawford, N. G., Harvey, M. G., Brumfield, R. T., & Glenn, T. C.
(2012). Ultraconserved elements anchor thousands of genetic markers spanning multiple evolutionary
timescales. Systematic biology, 61(5), 717-726.
Foster, P. G. (2004). Modeling compositional heterogeneity. Systematic biology, 53(3), 485-495.
Goldman, N. (1993). Statistical tests of models of DNA substitution. Journal of molecular evolution,
36(2), 182-198.
Golfetti, I. F. (2019). Análise filogenética de Scolia Fabricius (Hymenoptera, Scoliidae, Scoliinae).
[Unpublished masters thesis], Instituto de Biociências, Letras e Ciências Exatas. See
http://hdl.handle.net/11449/181962
.CC-BY-NC-ND 4.0 International licenseavailable under a
(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprintthis version posted June 26, 2022. ; https://doi.org/10.1101/2022.01.24.474473doi: bioRxiv preprint
Gotoh, O. (1993). Optimal alignment between groups of sequences and its application to multiple
sequence alignment. Bioinformatics, 9(3), 361-370.
Grabherr, M. G., Haas, B. J., Yassour, M., Levin, J. Z., Thompson, D. A., Amit, I., ... & Regev, A.
(2011). Full-length transcriptome assembly from RNA-Seq data without a reference genome. Nature
biotechnology, 29(7), 644.
Gurevich, A., Saveliev, V., Vyahhi, N., & Tesler, G. (2013). QUAST: quality assessment tool for
genome assemblies. Bioinformatics, 29(8), 1072-1075.
Haichun, Z., Rasnitsyn, A. P., & Junfeng, Z. (2002). The oldest known scoliid wasps (Insecta,
Hymenoptera, Scoliidae) from the Jehol biota of western Liaoning, China. Cretaceous Research, 23(1),
77-86.
Hoang, D. T., Chernomor, O., Von Haeseler, A., Minh, B. Q., & Vinh, L. S. (2018). UFBoot2:
improving the ultrafast bootstrap approximation. Molecular biology and evolution, 35(2), 518-522.
Hoffmeister, M. F. C. (2020). From Gondwana to the Great American Biotic Interchange: The Birth of
South American Fauna. In Pilauco: A Late Pleistocene Archaeo-paleontological Site (pp. 13-32).
Springer, Cham.
Höhna, S., Coghill, L. M., Mount, G. G., Thomson, R. C., & Brown, J. M. (2018). P3: Phylogenetic
posterior prediction in RevBayes. Molecular biology and evolution, 35(4), 1028-1034.
Höhna, S., Heath, T. A., Boussau, B., Landis, M. J., Ronquist, F., & Huelsenbeck, J. P. (2014).
Probabilistic graphical model representation in phylogenetics. Systematic biology, 63(5), 753-771.
Höhna, S., Landis, M. J., Heath, T. A., Boussau, B., Lartillot, N., Moore, B. R., ... & Ronquist, F.
(2016). RevBayes: Bayesian phylogenetic inference using graphical models and an interactive model-
specification language. Systematic biology, 65(4), 726-736.
Huelsenbeck, J. P., Ronquist, F., Nielsen, R., & Bollback, J. P. (2001). Bayesian inference of phylogeny
and its impact on evolutionary biology. science, 294(5550), 2310-2314.
Illingworth, J. F. (1921). Natural enemies of sugar-cane beetles in Queensland. Queensland Bureau
Sugar Experimental Station Division of Entomology Bulletin, 13, 1-47.
Inoue, M., & Endo, T. (2008). Below-ground host location by Campsomeriella annulata (Hymenoptera:
Scoliidae), a parasitoid of scarabaeid grubs. Journal of ethology, 26(1), 43.
Jermiin, L. S., Jayaswal, V., Ababneh, F. M., & Robinson, J. (2017). Identifying optimal models of
evolution. In Bioinformatics (pp. 379-420). Humana Press, New York, NY.
Jervis, M. (1998). Functional and evolutionary aspects of mouthpart structure in parasitoid wasps.
Biological Journal of the Linnean Society, 63(4), 461-493.
Jiang, D., Klaus, S., Zhang, Y. P., Hillis, D. M., & Li, J. T. (2019). Asymmetric biotic interchange
across the Bering land bridge between Eurasia and North America. National Science Review, 6(4), 739-
745.
Johnson, B. R., Borowiec, M. L., Chiu, J. C., Lee, E. K., Atallah, J., & Ward, P. S. (2013).
Phylogenomics resolves evolutionary relationships among ants, bees, and wasps. Current Biology,
23(20), 2058-2062.
Jones, D. L., & Gray, B. (1974). The pollination of Calochilus holtzei F. Muell. American Orchid
Society Bulletin, 43, 604-606.
.CC-BY-NC-ND 4.0 International licenseavailable under a
(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprintthis version posted June 26, 2022. ; https://doi.org/10.1101/2022.01.24.474473doi: bioRxiv preprint
Jones, G. (2017). Algorithmic improvements to species delimitation and phylogeny estimation under
the multispecies coalescent. Journal of mathematical biology, 74(1-2), 447-467.
Jonkman, J. C. M. (1980). The external and internal structure and growth of nests of the leaf‐cutting ant
Atta vollenweideri Forel, 1893 (Hym.: Formicidae) Part I 1. Zeitschrift für angewandte Entomologie,
89(1‐5), 158-173.
Joshi, N. A., & Fass, J. N. (2011). Sickle: A sliding-window, adaptive, quality-based trimming tool for
FastQ files (Version 1.33). URL: https://github.com/najoshi/sickle
Katoh, K., & Standley, D. M. (2013). MAFFT multiple sequence alignment software version 7:
improvements in performance and usability. Molecular biology and evolution, 30(4), 772-780.
Kimsey, L. S., & Brothers, D. J. (2016). The life, publications and new taxa of Qabir Argaman (Carol
Nagy). Journal of Hymenoptera Research, 50(50), 141-178.
Klopfstein, S., & Ronquist, F. (2013). Convergent intron gains in hymenopteran elongation factor-1α.
Molecular phylogenetics and evolution, 67(1), 266-276.
Larsson, A. (2014). AliView: a fast and lightweight alignment viewer and editor for large datasets.
Bioinformatics, 30(22), 3276-3278.
Martin, M. (2011). Cutadapt removes adapter sequences from high-throughput sequencing reads.
EMBnet. journal, 17(1), 10-12.
May, M. R. & Moore, B. R. (2017). bonsai: Automated phylogenetic MCMC diagnosis. R package
version 0.9. URL: https://github.com/mikeryanmay/bonsai
Minh, B. Q., Schmidt, H. A., Chernomor, O., Schrempf, D., Woodhams, M. D., Von Haeseler, A., &
Lanfear, R. (2020). IQ-TREE 2: New models and efficient methods for phylogenetic inference in the
genomic era. Molecular biology and evolution, 37(5), 1530-1534.
Mirarab, S. (2019). Species tree estimation using ASTRAL: practical considerations. arXiv preprint
arXiv:1904.03826.
Miyagi, I. (1960). Ecological studies of some Japanese species of Scoliidae. Transactions of the
Shikoku Entomological Society, 6, 104-120.
Molloy, E. K., & Warnow, T. (2018). To include or not to include: the impact of gene filtering on
species tree estimation methods. Systematic Biology, 67(2), 285-303.
Naser-Khdour, S., Minh, B. Q., Zhang, W., Stone, E. A., & Lanfear, R. (2019). The prevalence and
impact of model violations in phylogenetic analysis. Genome biology and evolution, 11(12), 3341-
3352.
Nel, A., Escuillie, F., & Garrouste, R. (2013). A new scoliid wasp in the Early Cretaceous Crato
Formation in Brazil (Hymenoptera: Scoliidae). Zootaxa, 3717, 395-400.
Nguyen, L. T., Schmidt, H. A., Von Haeseler, A., & Minh, B. Q. (2015). IQ-TREE: a fast and effective
stochastic algorithm for estimating maximum-likelihood phylogenies. Molecular biology and
evolution, 32(1), 268-274.
Osten, T. (1988). Die Mundwerkzeuge von Proscolia spectator Day (Hymenoptera: Aculeata). Ein
Beitrag zur Phylogenie der "Scolioidea". Stuttgarter Beiträge zur Naturkunde, 415, 1-30.
Osten, T. (1993). The enigmatic genus Proscolia, its distribution and phylogenetic relationships
(Insecta, Hymenoptera). Biologia Gallo-hellenica, 20(1), 177-182.
.CC-BY-NC-ND 4.0 International licenseavailable under a
(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprintthis version posted June 26, 2022. ; https://doi.org/10.1101/2022.01.24.474473doi: bioRxiv preprint
Osten, T. (2005). Checkliste der Dolchwespen der Welt (Insecta: Hymenoptera, Scoliidae). Bericht der
Naturforschenden Gesellschaft Augsburg, 62, 1-62.
Peters, R. S., Niehuis, O., Gunkel, S., Bläser, M., Mayer, C., Podsiadlowski, L., ... & Krogmann, L.
(2018). Transcriptome sequence-based phylogeny of chalcidoid wasps (Hymenoptera: Chalcidoidea)
reveals a history of rapid radiations, convergence, and evolutionary success. Molecular Phylogenetics
and Evolution, 120, 286-296.
Pilgrim, E. M., Von Dohlen, C. D., & Pitts, J. P. (2008). Molecular phylogenetics of Vespoidea indicate
paraphyly of the superfamily and novel relationships of its component families and subfamilies.
Zoologica Scripta, 37(5), 539-560.
Rambaut, A., Drummond, A. J., Xie, D., Baele, G., & Suchard, M. A. (2018). Posterior summarization
in Bayesian phylogenetics using Tracer 1.7. Systematic biology, 67(5), 901.
Rannala, B., & Yang, Z. (2003). Bayes estimation of species divergence times and ancestral population
sizes using DNA sequences from multiple loci. Genetics, 164(4), 1645-1656.
Rasnitsyn AP. (1977). New subfamily of scoliid wasps (Hymenoptera, Scoliidae, Proscoliinae).
Zoologicheskii zhurnal, 66, 522-599.
Rasnitsyn, A. P. (1993). Archaeoscoliinae, an extinct subfamily of scoliid wasps (Insecta: Vespida =
Hymenoptera: Scoliidae). Journal of Hymenoptera Research, 2(1), 85-96.
Rasnitsyn, A. P., & Martınez-Delclos, X. (1999). New Cretaceous Scoliidae (Vespida = Hymenoptera)
from the Lower Cretaceous of Spain and Brazil. Cretaceous Research, 20(6), 767-772.
Rosen, D. (1986). The role of taxonomy in effective biological control programs. Agriculture,
Ecosystems & Environment, 15(2-3), 121-129.
Sayyari, E., Whitfield, J. B., & Mirarab, S. (2017). Fragmentary gene sequences negatively impact
gene tree and species tree reconstruction. Molecular biology and evolution, 34(12), 3279-3291.
Schwarz, G. (1978). Estimating the dimension of a model. The annals of statistics, 6(2), 461-464.
Tani, S., & Ueno, T. (2013). Site fidelity and long-distance homing by males of solitary parasitic wasps
(Hymenoptera: Scoliidae). The Canadian Entomologist, 145(3), 333-337.
Tavaré, S. (1986). Some probabilistic and statistical problems in the analysis of DNA sequences.
Lectures on mathematics in the life sciences, 17(2), 57-86.
Uitert, M. V., Meuleman, W., & Wessels, L. (2008). Biclustering sparse binary genomic data. Journal
of Computational Biology, 15(10), 1329-1345.
Ward, P. S., & Branstetter, M. G. (2017). The acacia ants revisited: convergent evolution and
biogeographic context in an iconic ant/plant mutualism. Proceedings of the Royal Society B: Biological
Sciences, 284(1850), 20162569.
Ward, P. S., & Fisher, B. L. (2016). Tales of dracula ants: the evolutionary history of the ant subfamily
Amblyoponinae (Hymenoptera: Formicidae). Systematic Entomology, 41(3), 683-693.
Wilson, F. (1960). A review of the biological control of insects and weeds in Australia and Australian
New Guinea. Commonwealth Agricultural Bureaux.
Yang, Z. (1994). Maximum likelihood phylogenetic estimation from DNA sequences with variable
rates over sites: approximate methods. Journal of Molecular evolution, 39(3), 306-314.
.CC-BY-NC-ND 4.0 International licenseavailable under a
(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprintthis version posted June 26, 2022. ; https://doi.org/10.1101/2022.01.24.474473doi: bioRxiv preprint
Yin, J., Zhang, C., & Mirarab, S. (2019). ASTRAL-MP: scaling ASTRAL to very large datasets using
randomization and parallelization. Bioinformatics, 35(20), 3961-3969.
Yu, J., Pang, X., Fu, W., Hilton, J., Liang, M., Jiang, Z., ... & Zhang, R. (2021). A high-resolution
timescale for the Miocene Shanwang diatomaceous shale lagerstätte (China): development of Wavelet
Scale Series Analysis for cyclostratigraphy. Geosciences Journal, 25, 561-574.
Zhang, J. (1989). Fossil insects from Shanwang, Shandong, China. Jinan: Shandong Science and
Technology Publishing House. [in Chinese].
Zhang, Q., Zhang, H., Rasnitsyn, A. P., & Jarzembowski, E. A. (2015). A new genus of Scoliidae
(Insecta: Hymenoptera) from the Lower Cretaceous of northeast China. Cretaceous Research, 52, 579-
584.
Zerbino, D. R., & Birney, E. (2008). Velvet: algorithms for de novo short read assembly using de
Bruijn graphs. Genome research, 18(5), 821-829.
Supporting Information References
Bartlett, O. C. (1912). The North American Digger Wasps of the Subfamily Scoliinae. Annals of the
Entomological Society of America, 5(4), 293-340.
Betrem, J. G. (1928). Monographie der Indo-Australischen Scoliiden Mit zoogeographischen
Betrachtungen. Treubia, 9 (Suppl.), 1-388.
Bradley, J. C. (1928a). A Revision of the New World Species of Trielis a Subgenus of Campsomeris
(Hymenoptera: Scoliidae). Transactions of the American Entomological Society (1890-), 54(3), 195-
214.
Bradley, J. C. (1928b). The Species of Campsomeris (Hymenoptera-Scoliidae) of the Plumipes Group,
Inhabiting the United States, the Greater Antilles, and the Bahama Islands. Proceedings of the Academy
of Natural Sciences of Philadelphia, 80, 313-337.
Bradley, J. C. (1957). The taxa of Campsomeris (Hymenoptera: Scoliidae) occurring in the New World.
Transactions of the American Entomological Society (1890-), 83(2), 65-77.
Grissell, E. E. (2007). Scoliid Wasps of Florida, Campsomeris, Scolia and Trielis spp. (Insecta:
Hymenoptera: Scoliidae). Institute of Food and Agricultural Sciences Extension Electronic Data
Information Source, 1-9. See https://edis.ifas.ufl.edu/pdf/IN/IN74500.pdf
Kalyaanamoorthy, S., Minh, B. Q., Wong, T. K., Von Haeseler, A., & Jermiin, L. S. (2017).
ModelFinder: fast model selection for accurate phylogenetic estimates. Nature methods, 14(6), 587-
589.
Kim, J. K., & Yoon, I. B. (1993). A taxonomic study of subfamily Scoliinae from Korea (Hymenoptera:
Scoliidae). Entomological Research Bulletin, 19, 1-6.
Krombein, K. V. (1978). Biosystematic studies of Ceylonese wasps, II: A monograph of the Scoliidae
(Hymenoptera: Scolioidea). Smithsonian Contributions to Zoology.
Osten, T. (2000). Die Scoliiden des Mittelmeer-Gebietes und angrenzender Regionen (Hymenoptera).
Ein Bestimmungsschlüssel. Linzer biologische Beiträge, 32(2), 537-593.
Rohwer, S. A. (1927). Some scoliid wasps from tropical America. Journal of the Washington Academy
of Sciences, 17(6), 150-155.
.CC-BY-NC-ND 4.0 International licenseavailable under a
(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprintthis version posted June 26, 2022. ; https://doi.org/10.1101/2022.01.24.474473doi: bioRxiv preprint
Figure 1. Maximum likelihood phylogeny including all samples (analysis 1a). Two distinct clades of Scolia
bicincta are highlighted in yellow. Node support values are based on Ultrafast Bootstrapping in IQTREE; darker
nodes reflect higher support.
0.04
Pyrrhoscolia_cf_usambaraensis_SPK2I_Af
Scolia_apherema_SPK2X_In
Scolia_sp_SPK3W_In
Scolia_bicincta_SPK6E
Tristimeris_cf_javana_SPK5O_In
Colpacampsomeris_indica_SPK5N_In
Campsomeriella_sp_SPK1W_Ma
Campsomeriella_sp_SPK2V_Ma
Scolia_bicincta_P1_SPK5Y
Scolia_bicincta_SPK6G
Scolia_bicincta_P3_SPK6C
Scolia_cf_lebongensis_SPK5P_In
Scolia_sp_SPK2T_Af
Micromeriella_marginella_SPK6K_In
Campsomeriella_cf_madonensis_SPK1B_Af
Campsomeriella_sp_SPK2Q_Ma
Megacampsomeris_pulchrivestita_SPK3O_In
Pygodasis_quadrimaculata_SPK3C_Nr
Scolia_bicincta_P3_SPK6B
Micromeriella_cf_pilosella_SPK1M_Af
Colpa_octomaculata_EZ3_Nr
Scolia_sauteri_SPK5R_In
Scolia_4-pustulata_SPK5I_In/Pa
Scolia_hirta_1KITE
Megascolia_capitata_SPK1A_In
Megacampsomeris_schulthessi_SPK5T_In/Pa
Micromeriella_pilosella_SPK1U_Ma
Scolia_cf_affinis_SPK5A_In
Micromeriella_cf_atropus_SPK2M_Af
Campsomeriella_cf_annulata_SPK5Q_In
Megascolia_cf_hageni_SPK3V_In
Scolia_bicincta_P1_SPK5X
Trisciloa_saussurei_SPK3M_As
Scolia_mexicana_SPK3R_Nr
Scolia_nobilitata_SPK2F_Nr
Scolia_sp_SPK2O_Ma
Scolia_cf_affinis_SPK5L_In/Pa
Dielis_plumipes_confluenta_SPK3F_Nr
Campsomeriella_collaris_SPK5J_In/Pa
Triscolia_ardens_SPK2A_Nr
Scolia_bicincta_P2_SPK5Z
Dielis_plumipes_plumipes_SPK3D_Nr
Xanthocampsomeris_limosa_SPK3L_Nr
Scolia_bicincta_P1_SPK5V
Megascolia_cf_procer_SPK1Q_In
Megascolia_cf_hageni_SPK2H_In
Scolia_dubia_dubia_SPK3S_Nr
Carinoscolia_vittifrons_SPK5H_Pa
Scolia_cf_affinis_SPK2D_In
Scoliinae_sp_EX577
Laevicampsomeris_formosa_SPK2N_As
Pygodasis_ephippium_SPK2Z_Nr/Nt
Dielis_trifasciata_SPK3I_Nr
Scolia_bicincta_SPK6H
Scolia_carmichaeli_SPK6N_In
Dielis_pilipes_SPK1Y_Nr
Scolia_apherema_SPK6L_In
Scolia_decorata_SPK5K_In/Pa
Scolia_bicincta_SPK3U_Nr
Megascolia_cf_azurea_SPK6J_In
Scolia_bicincta_P2_SPK6A
Megacamsomeris_cf_lindenii_SPK5S_In
Cathimeris_hymenaea_SPK2L_Af
Campsomeriella_cf_caelebs_SPK1E_Af
Apterogyna_za01_Genome
Scolia_sp_SPK2R_Ma
Scolia_bicincta_P1_SPK5W_Nr
Scolia_cf_flaviceps_SPK1R_Pa
Megameris_soleata_SPK1F_Af
Scolia_bicincta_SPK6F
Micromeriella_cf_aureola_SPK1D_Af
Colpa_alcione_SPK1N_Nr
Megacampsomeris_pulchrivestita_SPK2Y_In
Dielis_dorsata_SPK3K_Nt
Scolia_bicincta_SPK6D
Scolia_sp_SPK5C_Af
Scolia_sp_SPK5F_Af
Colpa_sexmaculata_1KITE_Pa
Scolia_cf_affinis_SPK6M_In
Scolia_bicincta_SPK6I
Proscolia_sp_EX568_Pa
Scolia_picteti_SPK4Z_In
Megascolia_sp_SPK5M_In
Scolia_verticalis_HPG11_As
Cathimeris_hymenaea_SPK2J_Af
Scolia_sp_SPK5D_Af
Dielis_tolteca_SPK2E_Nr/Nt
Micromeriella_hyalina_SPK2K_Af
Campsomeriella_sp_SPK2P_Ma
Dielis_plumipes_fossulana_OE10_Nr
Scolia_cf_nobilis_SPK5G_Pa
.CC-BY-NC-ND 4.0 International licenseavailable under a
(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprintthis version posted June 26, 2022. ; https://doi.org/10.1101/2022.01.24.474473doi: bioRxiv preprint
Figure 2. Maximum likelihood phylogeny based on 727 UCE loci (analysis 1b). Node support values based on
Ultrafast Bootstrapping in IQTREE. All unlabeled internal nodes have 100% bootstrap support.
0.09
Dielis_pilipes_SPK1Y_Nr
Scolia_sp_SPK5F_Af
Scolia_apherema_SPK6L_In
Scolia_sauteri_SPK5R_In
Scolia_sp_SPK5C_Af
Dielis_trifasciata_SPK3I_Nr
Campsomeriella_cf_annulata_SPK5Q_In
Scolia_picteti_SPK4Z_In
Megascolia_capitata_SPK1A_In
Scolia_apherema_SPK2X_In
Scolia_cf_flaviceps_SPK1R_Pa
Cathimeris_hymenaea_SPK2J_Af
Scolia_cf_affinis_SPK6M_In
Colpa_sexmaculata_1KITE_Pa
Dielis_plumipes_plumipes_SPK3D_Nr
Megascolia_cf_azurea_SPK6J_In
Triscolia_ardens_SPK2A_Nr
Colpa_alcione_SPK1N_Nr
Scolia_carmichaeli_SPK6N_In
Megascolia_cf_procer_SPK1Q_In
Campsomeriella_sp_SPK2Q_Ma
Tristimeris_cf_javana_SPK5O_In
Pyrrhoscolia_cf_usambaraensis_SPK2I_Af
Megascolia_sp_SPK5M_In
Scolia_verticalis_HPG11_As
Pygodasis_quadrimaculata_SPK3C_Nr
Megascolia_cf_hageni_SPK2H_In
Trisciloa_saussurei_SPK3M_As
Scolia_sp_SPK2R_Ma
Dielis_dorsata_SPK3K_Nt
Megacampsomeris_schulthessi_SPK5T_In/Pa
Carinoscolia_vittifrons_SPK5H_Pa
Scolia_cf_lebongensis_SPK5P_In
Apterogyna_za01_Genome
Colpa_octomaculata_EZ3_Nr
Scolia_sp_SPK2T_Af
Scolia_cf_affinis_SPK5A_In
Xanthocampsomeris_limosa_SPK3L_Nr
Scolia_nobilitata_SPK2F_Nr
Dielis_tolteca_SPK2E_Nr/Nt
Micromeriella_cf_aureola_SPK1D_Af
Scolia_sp_SPK3W_In
Dielis_plumipes_confluenta_SPK3F_Nr
Scolia_4-pustulata_SPK5I_In/Pa
Micromeriella_marginella_SPK6K_In
Pygodasis_ephippium_SPK2Z_Nr/Nt
Micromeriella_cf_atropus_SPK2M_Af
Proscolia_sp_EX568_Pa
Scolia_cf_nobilis_SPK5G_Pa
Laevicampsomeris_formosa_SPK2N_As
Colpacampsomeris_indica_SPK5N_In
Megameris_soleata_SPK1F_Af
Scolia_cf_affinis_SPK2D_In
Micromeriella_hyalina_SPK2K_Af
Campsomeriella_collaris_SPK5J_In/Pa
Scolia_dubia_dubia_SPK3S_Nr
Scolia_decorata_SPK5K_In/Pa
Scolia_cf_affinis_SPK5L_In/Pa
Megacamsomeris_cf_lindenii_SPK5S_In
Megacampsomeris_pulchrivestita_SPK2Y_In
Campsomeriella_sp_SPK2P_Ma
Scolia_sp_SPK2O_Ma
Campsomeriella_cf_madonensis_SPK1B_Af
Scolia_bicincta_P1_SPK5W_Nr
Campsomeriella_cf_caelebs_SPK1E_Af
Campsomeriella_sp_SPK2V_Ma
Micromeriella_cf_pilosella_SPK1M_Af
Scolia_bicincta_SPK3U_Nr
Scolia_sp_SPK5D_Af
Megacampsomeris_pulchrivestita_SPK3O_In
Megascolia_cf_hageni_SPK3V_In
Cathimeris_hymenaea_SPK2L_Af
Dielis_plumipes_fossulana_OE10_Nr
Micromeriella_pilosella_SPK1U_Ma
Campsomeriella_sp_SPK1W_Ma
Scolia_mexicana_SPK3R_Nr
89
48
.CC-BY-NC-ND 4.0 International licenseavailable under a
(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprintthis version posted June 26, 2022. ; https://doi.org/10.1101/2022.01.24.474473doi: bioRxiv preprint
Laevicampsomeris_formosa_SPK2N_As
Scolia_cf_nobilis_SPK5G_Pa
Scolia_apherema_SPK2X_In
Scolia_bicincta_P1_SPK5W_Nr
Scolia_bicincta_SPK3U_Nr
Colpa_octomaculata_EZ3_Nr
Pygodasis_quadrimaculata_SPK3C_Nr
Apterogyna_za01_Genome
Cathimeris_hymenaea_SPK2L_Af
Campsomeriella_cf_madonensis_SPK1B_Af
Tristimeris_cf_javana_SPK5O_In
Megascolia_cf_azurea_SPK6J_In
Scolia_sp_SPK5F_Af
Campsomeriella_sp_SPK2Q_Ma
Scolia_mexicana_SPK3R_Nr
Scolia_sauteri_SPK5R_In
Dielis_tolteca_SPK2E_Nr/Nt
Campsomeriella_cf_caelebs_SPK1E_Af
Micromeriella_cf_atropus_SPK2M_Af
Micromeriella_cf_pilosella_SPK1M_Af
Scolia_sp_SPK3W_In
Campsomeriella_sp_SPK2V_Ma
Cathimeris_hymenaea_SPK2J_Af
Scolia_sp_SPK5C_Af
Scolia_picteti_SPK4Z_In
Pyrrhoscolia_cf_usambaraensis_SPK2I_Af
Scolia_cf_affinis_SPK6M_In
Pygodasis_ephippium_SPK2Z_Nr/Nt
Megascolia_sp_SPK5M_In
Micromeriella_hyalina_SPK2K_Af
Megacampsomeris_pulchrivestita_SPK2Y_In
Colpacampsomeris_indica_SPK5N_In
Triscolia_ardens_SPK2A_Nr
Micromeriella_cf_aureola_SPK1D_Af
Scolia_dubia_dubia_SPK3S_Nr
Carinoscolia_vittifrons_SPK5H_Pa
Megascolia_cf_procer_SPK1Q_In
Colpa_alcione_SPK1N_Nr
Campsomeriella_cf_annulata_SPK5Q_In
Colpa_sexmaculata_1KITE_Pa
Megacampsomeris_schulthessi_SPK5T_In/Pa
Scolia_cf_flaviceps_SPK1R_Pa
Scolia_cf_affinis_SPK2D_In
Scolia_cf_lebongensis_SPK5P_In
Dielis_plumipes_fossulana_OE10_Nr
Dielis_plumipes_confluenta_SPK3F_Nr
Scolia_decorata_SPK5K_In/Pa
Micromeriella_pilosella_SPK1U_Ma
Megascolia_capitata_SPK1A_In
Dielis_pilipes_SPK1Y_Nr
Campsomeriella_collaris_SPK5J_In/Pa
Scolia_sp_SPK2R_Ma
Megascolia_cf_hageni_SPK3V_In
Scolia_apherema_SPK6L_In
Micromeriella_marginella_SPK6K_In
Proscolia_sp_EX568_Pa
Scolia_carmichaeli_SPK6N_In
Scolia_cf_affinis_SPK5L_In/Pa
Scolia_4-pustulata_SPK5I_In/Pa
Megameris_soleata_SPK1F_Af
Megascolia_cf_hageni_SPK2H_In
Scolia_cf_affinis_SPK5A_In
Scolia_verticalis_HPG11_As
Megacampsomeris_pulchrivestita_SPK3O_In
Campsomeriella_sp_SPK2P_Ma
Scolia_nobilitata_SPK2F_Nr
Dielis_dorsata_SPK3K_Nt
Dielis_trifasciata_SPK3I_Nr
Scolia_sp_SPK2T_Af
Xanthocampsomeris_limosa_SPK3L_Nr
Trisciloa_saussurei_SPK3M_As
Megacamsomeris_cf_lindenii_SPK5S_In
Campsomeriella_sp_SPK1W_Ma
Scolia_sp_SPK2O_Ma
Scolia_sp_SPK5D_Af
Dielis_plumipes_plumipes_SPK3D_Nr
Apterogyna_za01_Genome
Colpa_alcione_SPK1N_Nr
Scolia_apherema_SPK6L_In
Campsomeriella_cf_caelebs_SPK1E_Af
Scolia_verticalis_HPG11_As
Megascolia_cf_hageni_SPK3V_In
Laevicampsomeris_formosa_SPK2N_As
Dielis_plumipes_confluenta_SPK3F_Nr
Scolia_sp_SPK2R_Ma
Scolia_sp_SPK5F_Af
Scolia_cf_affinis_SPK2D_In
Tristimeris_cf_javana_SPK5O_In
Scolia_mexicana_SPK3R_Nr
Megascolia_sp_SPK5M_In
Campsomeriella_sp_SPK2P_Ma
Campsomeriella_sp_SPK2Q_Ma
Megacampsomeris_pulchrivestita_SPK2Y_In
Scolia_cf_lebongensis_SPK5P_In
Scolia_sp_SPK3W_In
Micromeriella_pilosella_SPK1U_Ma
Campsomeriella_collaris_SPK5J_In/Pa
Megacampsomeris_schulthessi_SPK5T_In/Pa
Scolia_cf_flaviceps_SPK1R_Pa
Scolia_picteti_SPK4Z_In
Scolia_cf_nobilis_SPK5G_Pa
Scolia_sp_SPK5D_Af
Megascolia_cf_azurea_SPK6J_In
Megameris_soleata_SPK1F_Af
Scolia_cf_affinis_SPK6M_In
Scolia_sp_SPK2O_Ma
Dielis_pilipes_SPK1Y_Nr
Scolia_carmichaeli_SPK6N_In
Xanthocampsomeris_limosa_SPK3L_Nr
Dielis_tolteca_SPK2E_Nr/Nt
Micromeriella_cf_aureola_SPK1D_Af
Carinoscolia_vittifrons_SPK5H_Pa
Megacamsomeris_cf_lindenii_SPK5S_In
Cathimeris_hymenaea_SPK2J_Af
Scolia_cf_affinis_SPK5L_In/Pa
Micromeriella_marginella_SPK6K_In
Scolia_dubia_dubia_SPK3S_Nr
Scolia_bicincta_P1_SPK5W_Nr
Campsomeriella_cf_madonensis_SPK1B_Af
Dielis_plumipes_fossulana_OE10_Nr
Megascolia_cf_hageni_SPK2H_In
Scolia_sp_SPK5C_Af
Campsomeriella_sp_SPK1W_Ma
Scolia_sp_SPK2T_Af
Micromeriella_cf_pilosella_SPK1M_Af
Campsomeriella_sp_SPK2V_Ma
Micromeriella_hyalina_SPK2K_Af
Scolia_4-pustulata_SPK5I_In/Pa
Megacampsomeris_pulchrivestita_SPK3O_In
Cathimeris_hymenaea_SPK2L_Af
Scolia_decorata_SPK5K_In/Pa
Colpacampsomeris_indica_SPK5N_In
Pygodasis_quadrimaculata_SPK3C_Nr
Pyrrhoscolia_cf_usambaraensis_SPK2I_Af
Dielis_dorsata_SPK3K_Nt
Micromeriella_cf_atropus_SPK2M_Af
Scolia_bicincta_SPK3U_Nr
Scolia_cf_affinis_SPK5A_In
Scolia_apherema_SPK2X_In
Proscolia_sp_EX568_Pa
Megascolia_capitata_SPK1A_In
Triscolia_ardens_SPK2A_Nr
Scolia_sauteri_SPK5R_In
Campsomeriella_cf_annulata_SPK5Q_In
Trisciloa_saussurei_SPK3M_As
Dielis_trifasciata_SPK3I_Nr
Dielis_plumipes_plumipes_SPK3D_Nr
Megascolia_cf_procer_SPK1Q_In
Pygodasis_ephippium_SPK2Z_Nr/Nt
Scolia_nobilitata_SPK2F_Nr
Colpa_octomaculata_EZ3_Nr
Colpa_sexmaculata_1KITE_Pa
Scolia_bicincta_SPK3U_Nr
Megascolia_cf_hageni_SPK3V_In
Scolia_cf_affinis_SPK5A_In
Scolia_picteti_SPK4Z_In
Dielis_plumipes_fossulana_OE10_Nr
Dielis_dorsata_SPK3K_Nt
Scolia_sauteri_SPK5R_In
Megacamsomeris_cf_lindenii_SPK5S_In
Colpa_sexmaculata_1KITE_Pa
Proscolia_sp_EX568_Pa
Cathimeris_hymenaea_SPK2L_Af
Micromeriella_cf_aureola_SPK1D_Af
Megascolia_capitata_SPK1A_In
Scolia_mexicana_SPK3R_Nr
Dielis_tolteca_SPK2E_Nr/Nt
Xanthocampsomeris_limosa_SPK3L_Nr
Campsomeriella_collaris_SPK5J_In/Pa
Carinoscolia_vittifrons_SPK5H_Pa
Scolia_bicincta_P1_SPK5W_Nr
Scolia_apherema_SPK2X_In
Scolia_sp_SPK2T_Af
Dielis_plumipes_plumipes_SPK3D_Nr
Campsomeriella_sp_SPK2P_Ma
Scolia_cf_affinis_SPK5L_In/Pa
Triscolia_ardens_SPK2A_Nr
Scolia_cf_lebongensis_SPK5P_In
Megascolia_cf_hageni_SPK2H_In
Campsomeriella_cf_caelebs_SPK1E_Af
Micromeriella_cf_atropus_SPK2M_Af
Megascolia_cf_azurea_SPK6J_In
Scolia_carmichaeli_SPK6N_In
Tristimeris_cf_javana_SPK5O_In
Campsomeriella_cf_annulata_SPK5Q_In
Scolia_sp_SPK2R_Ma
Pygodasis_ephippium_SPK2Z_Nr/Nt
Colpa_alcione_SPK1N_Nr
Apterogyna_za01_Genome
Scolia_4-pustulata_SPK5I_In/Pa
Campsomeriella_sp_SPK1W_Ma
Scolia_sp_SPK3W_In
Micromeriella_cf_pilosella_SPK1M_Af
Campsomeriella_sp_SPK2V_Ma
Campsomeriella_cf_madonensis_SPK1B_Af
Cathimeris_hymenaea_SPK2J_Af
Scolia_dubia_dubia_SPK3S_Nr
Scolia_nobilitata_SPK2F_Nr
Scolia_sp_SPK5C_Af
Scolia_cf_nobilis_SPK5G_Pa
Micromeriella_pilosella_SPK1U_Ma
Megascolia_sp_SPK5M_In
Megacampsomeris_pulchrivestita_SPK2Y_In
Megacampsomeris_pulchrivestita_SPK3O_In
Micromeriella_hyalina_SPK2K_Af
Colpacampsomeris_indica_SPK5N_In
Pygodasis_quadrimaculata_SPK3C_Nr
Colpa_octomaculata_EZ3_Nr
Scolia_cf_affinis_SPK2D_In
Micromeriella_marginella_SPK6K_In
Scolia_sp_SPK2O_Ma
Megascolia_cf_procer_SPK1Q_In
Pyrrhoscolia_cf_usambaraensis_SPK2I_Af
Laevicampsomeris_formosa_SPK2N_As
Scolia_sp_SPK5F_Af
Scolia_cf_affinis_SPK6M_In
Dielis_trifasciata_SPK3I_Nr
Scolia_verticalis_HPG11_As
Megacampsomeris_schulthessi_SPK5T_In/Pa
Scolia_sp_SPK5D_Af
Scolia_apherema_SPK6L_In
Scolia_decorata_SPK5K_In/Pa
Campsomeriella_sp_SPK2Q_Ma
Scolia_cf_flaviceps_SPK1R_Pa
Megameris_soleata_SPK1F_Af
Dielis_plumipes_confluenta_SPK3F_Nr
Dielis_pilipes_SPK1Y_Nr
Trisciloa_saussurei_SPK3M_As
0.69
0.98
0.88
0.96
0.93
0.93
0.50
0.97
0.87
Figure 3. ASTRAL species trees (analysis 1c) based
on ML trees of (A) loci not failing the maximum test of
symmetry, (B) loci failing the the maximum test of
symmetry, and (C) all loci. Branch labels represent the
local posterior probability of the associated
quadripartition. All unlabeled quadripartitions have a
local posterior probability of 1.0.
A B
C
.CC-BY-NC-ND 4.0 International licenseavailable under a
(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprintthis version posted June 26, 2022. ; https://doi.org/10.1101/2022.01.24.474473doi: bioRxiv preprint
3.0
Laevicampsomeris_formosa_SPK2N_As
Scolia_cf_affinis_SPK6M_In
Scolia_cf_affinis_SPK2D_In
Megascolia_cf_hageni_SPK3V_In
Campsomeriella_sp_SPK2P_Ma
Micromeriella_cf_aureola_SPK1D_Af
Xanthocampsomeris_limosa_SPK3L_Nr
Campsomeriella_cf_caelebs_SPK1E_Af
Megascolia_cf_hageni_SPK2H_In
Scolia_sp_SPK2T_Af
Micromeriella_marginella_SPK6K_In
Micromeriella_cf_pilosella_SPK1M_Af
Campsomeriella_cf_annulata_SPK5Q_In
Triscolia_ardens_SPK2A_Nr
Scolia_sp_SPK3W_In
Scolia_sp_SPK5D_Af
Colpacampsomeris_indica_SPK5N_In
Megascolia_sp_SPK5M_In
Micromeriella_pilosella_SPK1U_Ma
Megascolia_capitata_SPK1A_In
Scolia_sp_SPK2O_Ma
Cathimeris_hymenaea_SPK2J_Af
Dielis_plumipes_confluenta_SPK3F_Nr
Colpa_octomaculata_EZ3_Nr
Scolia_decorata_SPK5K_In/Pa
Scolia_sp_SPK2R_Ma
Campsomeriella_sp_SPK2V_Ma
Carinoscolia_vittifrons_SPK5H_Pa
Scolia_cf_affinis_SPK5L_In/Pa
Micromeriella_hyalina_SPK2K_Af
Scolia_picteti_SPK4Z_In
Scolia_4-pustulata_SPK5I_In/Pa
Campsomeriella_sp_SPK1W_Ma
Dielis_plumipes_plumipes_SPK3D_Nr
Micromeriella_cf_atropus_SPK2M_Af
Scolia_bicincta_P1_SPK5W_Nr
Dielis_plumipes_fossulana_OE10_Nr
Scolia_verticalis_HPG11_As
Megascolia_cf_procer_SPK1Q_In
Scolia_sp_SPK5C_Af
Megascolia_cf_azurea_SPK6J_In
Dielis_dorsata_SPK3K_Nt
Scolia_cf_flaviceps_SPK1R_Pa
Pygodasis_ephippium_SPK2Z_Nr/Nt
Scolia_cf_affinis_SPK5A_In
Proscolia_sp_EX568_Pa
Colpa_alcione_SPK1N_Nr
Scolia_sp_SPK5F_Af
Scolia_apherema_SPK6L_In
Campsomeriella_collaris_SPK5J_In/Pa
Dielis_tolteca_SPK2E_Nr/Nt
Cathimeris_hymenaea_SPK2L_Af
Scolia_cf_lebongensis_SPK5P_In
Trisciloa_saussurei_SPK3M_As
Megacampsomeris_pulchrivestita_SPK2Y_In
Scolia_sauteri_SPK5R_In
Scolia_dubia_dubia_SPK3S_Nr
Scolia_mexicana_SPK3R_Nr
Scolia_nobilitata_SPK2F_Nr
Pyrrhoscolia_cf_usambaraensis_SPK2I_Af
Scolia_cf_nobilis_SPK5G_Pa
Scolia_bicincta_SPK3U_Nr
Scolia_carmichaeli_SPK6N_In
Megacampsomeris_pulchrivestita_SPK3O_In
Campsomeriella_sp_SPK2Q_Ma
Megacampsomeris_schulthessi_SPK5T_In/Pa
Dielis_pilipes_SPK1Y_Nr
Apterogyna_za01_Genome
Megameris_soleata_SPK1F_Af
Megacamsomeris_cf_lindenii_SPK5S_In
Dielis_trifasciata_SPK3I_Nr
Colpa_sexmaculata_1KITE_Pa
Scolia_apherema_SPK2X_In
Pygodasis_quadrimaculata_SPK3C_Nr
Tristimeris_cf_javana_SPK5O_In
Campsomeriella_cf_madonensis_SPK1B_Af
Figure 4. ASTRAL species tree (analysis 1c) based on ML trees of loci not failing the maximum test of
symmetry. Branch labels represent the local posterior probability of the associated quadripartition. All
unlabeled quadripartitions have a local posterior probability of 1.0.
0.88
0.69
0.98
.CC-BY-NC-ND 4.0 International licenseavailable under a
(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprintthis version posted June 26, 2022. ; https://doi.org/10.1101/2022.01.24.474473doi: bioRxiv preprint
Megascolia_capitata_SPK1A_In
Scolia_cf_affinis_SPK5L_In/Pa
Scolia_sp_SPK2T_Af
Megameris_soleata_SPK1F_Af
Megacampsomeris_pulchrivestita_SPK3O_In
Scolia_cf_affinis_SPK5A_In
Megascolia_cf_hageni_SPK2H_In
Dielis_dorsata_SPK3K_Nt
Scolia_apherema_SPK2X_In
Dielis_tolteca_SPK2E_Nr/Nt
Dielis_pilipes_SPK1Y_Nr
Xanthocampsomeris_limosa_SPK3L_Nr
Megascolia_cf_azurea_SPK6J_In
Megacampsomeris_schulthessi_SPK5T_In/Pa
Scolia_verticalis_HPG11_As
Cathimeris_hymenaea_SPK2L_Af
Dielis_plumipes_confluenta_SPK3F_Nr
Campsomeriella_sp_SPK1W_Ma
Scolia_bicincta_P1_SPK5W_Nr
Carinoscolia_vittifrons_SPK5H_Pa
Apterogyna_za01_Genome
Scolia_nobilitata_SPK2F_Nr
Campsomeriella_cf_madonensis_SPK1B_Af
Dielis_plumipes_plumipes_SPK3D_Nr
Micromeriella_cf_aureola_SPK1D_Af
Colpa_sexmaculata_1KITE_Pa
Micromeriella_marginella_SPK6K_In
Campsomeriella_sp_SPK2V_Ma
Megascolia_sp_SPK5M_In
Scolia_decorata_SPK5K_In/Pa
Tristimeris_cf_javana_SPK5O_In
Trisciloa_saussurei_SPK3M_As
Colpa_octomaculata_EZ3_Nr
Scolia_sp_SPK2R_Ma
Campsomeriella_cf_annulata_SPK5Q_In
Dielis_trifasciata_SPK3I_Nr
Micromeriella_hyalina_SPK2K_Af
Scolia_sp_SPK5C_Af
Scolia_cf_flaviceps_SPK1R_Pa
Micromeriella_pilosella_SPK1U_Ma
Scolia_cf_nobilis_SPK5G_Pa
Campsomeriella_cf_caelebs_SPK1E_Af
Laevicampsomeris_formosa_SPK2N_As
Micromeriella_cf_pilosella_SPK1M_Af
Scolia_sp_SPK2O_Ma
Scolia_mexicana_SPK3R_Nr
Scolia_cf_lebongensis_SPK5P_In
Scolia_cf_affinis_SPK6M_In
Megascolia_cf_procer_SPK1Q_In
Triscolia_ardens_SPK2A_Nr
Dielis_plumipes_fossulana_OE10_Nr
Campsomeriella_sp_SPK2Q_Ma
Pyrrhoscolia_cf_usambaraensis_SPK2I_Af
Cathimeris_hymenaea_SPK2J_Af
Pygodasis_ephippium_SPK2Z_Nr/Nt
Colpacampsomeris_indica_SPK5N_In
Scolia_carmichaeli_SPK6N_In
Pygodasis_quadrimaculata_SPK3C_Nr
Proscolia_sp_EX568_Pa
Megascolia_cf_hageni_SPK3V_In
Scolia_dubia_dubia_SPK3S_Nr
Megacampsomeris_pulchrivestita_SPK2Y_In
Scolia_bicincta_SPK3U_Nr
Scolia_sp_SPK5F_Af
Campsomeriella_collaris_SPK5J_In/Pa
Scolia_cf_affinis_SPK2D_In
Scolia_picteti_SPK4Z_In
Colpa_alcione_SPK1N_Nr
Scolia_apherema_SPK6L_In
Micromeriella_cf_atropus_SPK2M_Af
Scolia_sp_SPK5D_Af
Campsomeriella_sp_SPK2P_Ma
Scolia_sp_SPK3W_In
Scolia_4-pustulata_SPK5I_In/Pa
Megacamsomeris_cf_lindenii_SPK5S_In
Scolia_sauteri_SPK5R_In
Scolia_bicincta_SPK3U_Nr
Megascolia_cf_hageni_SPK2H_In
Cathimeris_hymenaea_SPK2L_Af
Colpa_sexmaculata_1KITE_Pa
Campsomeriella_cf_caelebs_SPK1E_Af
Micromeriella_marginella_SPK6K_In
Micromeriella_pilosella_SPK1U_Ma
Scolia_picteti_SPK4Z_In
Megacampsomeris_pulchrivestita_SPK2Y_In
Scolia_nobilitata_SPK2F_Nr
Colpa_alcione_SPK1N_Nr
Scolia_cf_affinis_SPK5L_In/Pa
Campsomeriella_sp_SPK2P_Ma
Apterogyna_za01_Genome
Scolia_sp_SPK5C_Af
Scolia_sp_SPK5F_Af
Colpa_octomaculata_EZ3_Nr
Cathimeris_hymenaea_SPK2J_Af
Pygodasis_ephippium_SPK2Z_Nr/Nt
Scolia_cf_affinis_SPK2D_In
Dielis_plumipes_plumipes_SPK3D_Nr
Scolia_sp_SPK2O_Ma
Xanthocampsomeris_limosa_SPK3L_Nr
Dielis_tolteca_SPK2E_Nr/Nt
Campsomeriella_collaris_SPK5J_In/Pa
Megacampsomeris_pulchrivestita_SPK3O_In
Dielis_dorsata_SPK3K_Nt
Megascolia_capitata_SPK1A_In
Scolia_mexicana_SPK3R_Nr
Megacampsomeris_schulthessi_SPK5T_In/Pa
Megascolia_sp_SPK5M_In
Scolia_4-pustulata_SPK5I_In/Pa
Scolia_cf_affinis_SPK5A_In
Scolia_sp_SPK3W_In
Colpacampsomeris_indica_SPK5N_In
Scolia_cf_nobilis_SPK5G_Pa
Dielis_plumipes_confluenta_SPK3F_Nr
Micromeriella_cf_pilosella_SPK1M_Af
Megascolia_cf_procer_SPK1Q_In
Pygodasis_quadrimaculata_SPK3C_Nr
Micromeriella_hyalina_SPK2K_Af
Scolia_cf_lebongensis_SPK5P_In
Campsomeriella_sp_SPK1W_Ma
Campsomeriella_sp_SPK2Q_Ma
Megameris_soleata_SPK1F_Af
Dielis_plumipes_fossulana_OE10_Nr
Micromeriella_cf_atropus_SPK2M_Af
Trisciloa_saussurei_SPK3M_As
Scolia_apherema_SPK6L_In
Triscolia_ardens_SPK2A_Nr
Scolia_apherema_SPK2X_In
Carinoscolia_vittifrons_SPK5H_Pa
Megascolia_cf_azurea_SPK6J_In
Dielis_pilipes_SPK1Y_Nr
Scolia_sp_SPK2R_Ma
Scolia_sp_SPK5D_Af
Campsomeriella_sp_SPK2V_Ma
Scolia_bicincta_P1_SPK5W_Nr
Pyrrhoscolia_cf_usambaraensis_SPK2I_Af
Megascolia_cf_hageni_SPK3V_In
Dielis_trifasciata_SPK3I_Nr
Campsomeriella_cf_madonensis_SPK1B_Af
Laevicampsomeris_formosa_SPK2N_As
Scolia_decorata_SPK5K_In/Pa
Scolia_cf_affinis_SPK6M_In
Scolia_carmichaeli_SPK6N_In
Micromeriella_cf_aureola_SPK1D_Af
Tristimeris_cf_javana_SPK5O_In
Campsomeriella_cf_annulata_SPK5Q_In
Scolia_cf_flaviceps_SPK1R_Pa
Scolia_dubia_dubia_SPK3S_Nr
Proscolia_sp_EX568_Pa
Scolia_sp_SPK2T_Af
Megacamsomeris_cf_lindenii_SPK5S_In
Scolia_verticalis_HPG11_As
Scolia_sauteri_SPK5R_In
Dielis_dorsata_SPK3K_Nt
Scolia_bicincta_P1_SPK5W_Nr
Triscolia_ardens_SPK2A_Nr
Megacamsomeris_cf_lindenii_SPK5S_In
Dielis_plumipes_fossulana_OE10_Nr
Pygodasis_quadrimaculata_SPK3C_Nr
Micromeriella_hyalina_SPK2K_Af
Laevicampsomeris_formosa_SPK2N_As
Pyrrhoscolia_cf_usambaraensis_SPK2I_Af
Proscolia_sp_EX568_Pa
Pygodasis_ephippium_SPK2Z_Nr/Nt
Scolia_mexicana_SPK3R_Nr
Scolia_picteti_SPK4Z_In
Dielis_plumipes_confluenta_SPK3F_Nr
Micromeriella_cf_aureola_SPK1D_Af
Megacampsomeris_pulchrivestita_SPK3O_In
Dielis_pilipes_SPK1Y_Nr
Megacampsomeris_pulchrivestita_SPK2Y_In
Scolia_apherema_SPK6L_In
Apterogyna_za01_Genome
Xanthocampsomeris_limosa_SPK3L_Nr
Scolia_sp_SPK2T_Af
Campsomeriella_collaris_SPK5J_In/Pa
Megascolia_cf_hageni_SPK3V_In
Tristimeris_cf_javana_SPK5O_In
Scolia_cf_affinis_SPK6M_In
Scolia_cf_flaviceps_SPK1R_Pa
Colpa_octomaculata_EZ3_Nr
Campsomeriella_cf_caelebs_SPK1E_Af
Scolia_nobilitata_SPK2F_Nr
Scolia_cf_affinis_SPK2D_In
Scolia_sp_SPK5C_Af
Dielis_trifasciata_SPK3I_Nr
Micromeriella_marginella_SPK6K_In
Campsomeriella_sp_SPK2Q_Ma
Scolia_cf_nobilis_SPK5G_Pa
Colpacampsomeris_indica_SPK5N_In
Campsomeriella_cf_annulata_SPK5Q_In
Colpa_alcione_SPK1N_Nr
Megameris_soleata_SPK1F_Af
Megascolia_cf_procer_SPK1Q_In
Scolia_sp_SPK3W_In
Megascolia_sp_SPK5M_In
Carinoscolia_vittifrons_SPK5H_Pa
Scolia_apherema_SPK2X_In
Cathimeris_hymenaea_SPK2L_Af
Trisciloa_saussurei_SPK3M_As
Scolia_verticalis_HPG11_As
Scolia_4-pustulata_SPK5I_In/Pa
Campsomeriella_sp_SPK2P_Ma
Dielis_plumipes_plumipes_SPK3D_Nr
Scolia_cf_affinis_SPK5L_In/Pa
Scolia_decorata_SPK5K_In/Pa
Scolia_sp_SPK2R_Ma
Scolia_sp_SPK5D_Af
Scolia_carmichaeli_SPK6N_In
Scolia_bicincta_SPK3U_Nr
Scolia_cf_lebongensis_SPK5P_In
Megascolia_cf_azurea_SPK6J_In
Scolia_sauteri_SPK5R_In
Scolia_sp_SPK2O_Ma
Megascolia_cf_hageni_SPK2H_In
Campsomeriella_cf_madonensis_SPK1B_Af
Scolia_dubia_dubia_SPK3S_Nr
Cathimeris_hymenaea_SPK2J_Af
Campsomeriella_sp_SPK1W_Ma
Micromeriella_cf_atropus_SPK2M_Af
Megascolia_capitata_SPK1A_In
Colpa_sexmaculata_1KITE_Pa
Micromeriella_pilosella_SPK1U_Ma
Dielis_tolteca_SPK2E_Nr/Nt
Campsomeriella_sp_SPK2V_Ma
Scolia_sp_SPK5F_Af
Megacampsomeris_schulthessi_SPK5T_In/Pa
Micromeriella_cf_pilosella_SPK1M_Af
Scolia_cf_affinis_SPK5A_In
0.91
0.99
0.99
0.93
0.46
0.98
0.91
0.97
0.82
A B
D
0.47
Figure 5. ASTRAL species trees (analysis 1c) based on MCC trees of (A) loci having the lowest (1/3) combined
posterior predictive effect sizes, (B) loci having the highest (1/3) combined posterior predictive effect sizes, (C)
loci for which the model was not found to be inadequate (alpha = 0.05), and (D) all loci. Branch labels
represent the local posterior probability of the associated quadripartition. All unlabeled quadripartitions have a
local posterior probability of 1.0.
Scolia_cf_affinis_SPK5A_In
Tristimeris_cf_javana_SPK5O_In
Scolia_apherema_SPK6L_In
Dielis_tolteca_SPK2E_Nr/Nt
Scolia_cf_affinis_SPK5L_In/Pa
Proscolia_sp_EX568_Pa
Scolia_dubia_dubia_SPK3S_Nr
Micromeriella_cf_atropus_SPK2M_Af
Megacamsomeris_cf_lindenii_SPK5S_In
Carinoscolia_vittifrons_SPK5H_Pa
Scolia_apherema_SPK2X_In
Colpa_alcione_SPK1N_Nr
Scolia_4-pustulata_SPK5I_In/Pa
Megascolia_capitata_SPK1A_In
Scolia_sp_SPK5D_Af
Scolia_sp_SPK2R_Ma
Megameris_soleata_SPK1F_Af
Dielis_trifasciata_SPK3I_Nr
Campsomeriella_sp_SPK2V_Ma
Campsomeriella_sp_SPK1W_Ma
Megacampsomeris_schulthessi_SPK5T_In/Pa
Micromeriella_cf_pilosella_SPK1M_Af
Campsomeriella_cf_madonensis_SPK1B_Af
Colpa_sexmaculata_1KITE_Pa
Scolia_bicincta_P1_SPK5W_Nr
Megascolia_cf_procer_SPK1Q_In
Triscolia_ardens_SPK2A_Nr
Megacampsomeris_pulchrivestita_SPK3O_In
Xanthocampsomeris_limosa_SPK3L_Nr
Scolia_sp_SPK2O_Ma
Dielis_pilipes_SPK1Y_Nr
Megascolia_cf_hageni_SPK2H_In
Cathimeris_hymenaea_SPK2J_Af
Micromeriella_pilosella_SPK1U_Ma
Micromeriella_cf_aureola_SPK1D_Af
Campsomeriella_collaris_SPK5J_In/Pa
Apterogyna_za01_Genome
Campsomeriella_sp_SPK2Q_Ma
Campsomeriella_cf_annulata_SPK5Q_In
Scolia_sp_SPK3W_In
Pyrrhoscolia_cf_usambaraensis_SPK2I_Af
Scolia_cf_affinis_SPK6M_In
Megascolia_cf_azurea_SPK6J_In
Scolia_bicincta_SPK3U_Nr
Scolia_cf_nobilis_SPK5G_Pa
Scolia_picteti_SPK4Z_In
Colpacampsomeris_indica_SPK5N_In
Scolia_decorata_SPK5K_In/Pa
Colpa_octomaculata_EZ3_Nr
Megascolia_sp_SPK5M_In
Scolia_mexicana_SPK3R_Nr
Scolia_verticalis_HPG11_As
Campsomeriella_cf_caelebs_SPK1E_Af
Scolia_cf_flaviceps_SPK1R_Pa
Megascolia_cf_hageni_SPK3V_In
Scolia_sp_SPK5F_Af
Scolia_cf_lebongensis_SPK5P_In
Dielis_plumipes_plumipes_SPK3D_Nr
Scolia_sp_SPK2T_Af
Scolia_nobilitata_SPK2F_Nr
Cathimeris_hymenaea_SPK2L_Af
Trisciloa_saussurei_SPK3M_As
Scolia_carmichaeli_SPK6N_In
Laevicampsomeris_formosa_SPK2N_As
Dielis_plumipes_confluenta_SPK3F_Nr
Campsomeriella_sp_SPK2P_Ma
Pygodasis_ephippium_SPK2Z_Nr/Nt
Dielis_dorsata_SPK3K_Nt
Pygodasis_quadrimaculata_SPK3C_Nr
Dielis_plumipes_fossulana_OE10_Nr
Scolia_cf_affinis_SPK2D_In
Megacampsomeris_pulchrivestita_SPK2Y_In
Micromeriella_marginella_SPK6K_In
Scolia_sp_SPK5C_Af
Scolia_sauteri_SPK5R_In
Micromeriella_hyalina_SPK2K_Af
C
0.87
0.99
0.43
0.98
0.98
0.59
0.81
0.58
0.78
.CC-BY-NC-ND 4.0 International licenseavailable under a
(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprintthis version posted June 26, 2022. ; https://doi.org/10.1101/2022.01.24.474473doi: bioRxiv preprint
0.04
Scolia_sp_SPK2O_Ma
Campsomeriella_cf_madonensis_SPK1B_Af
Megacamsomeris_cf_lindenii_SPK5S_In
Campsomeriella_cf_caelebs_SPK1E_Af
Megacampsomeris_pulchrivestita_SPK3O_In
Scolia_4-pustulata_SPK5I_In/Pa
Cathimeris_hymenaea_SPK2J_Af
Scolia_sp_SPK5F_Af
Scolia_sp_SPK2R_Ma
Carinoscolia_vittifrons_SPK5H_Pa
Campsomeriella_sp_SPK1W_Ma
Campsomeriella_sp_SPK2Q_Ma
Triscolia_ardens_SPK2A_Nr
Scolia_cf_lebongensis_SPK5P_In
Campsomeriella_collaris_SPK5J_In/Pa
Scolia_cf_nobilis_SPK5G_Pa
Campsomeriella_sp_SPK2P_Ma
Dielis_tolteca_SPK2E_Nr
Scolia_cf_affinis_SPK5L_In/Pa
Pygodasis_quadrimaculata_SPK3C_Nr
Dielis_plumipes_confluenta_SPK3F_Nr
Micromeriella_cf_aureola_SPK1D_Af
Dielis_plumipes_plumipes_SPK3D_Nr
Trisciloa_saussurei_SPK3M_As
Apterogyna_za01_Genome
Pyrrhoscolia_cf_usambaraensis_SPK2I_Af
Scolia_mexicana_SPK3R_Nr
Scolia_sp_SPK2T_Af
Dielis_pilipes_SPK1Y_Nr
Campsomeriella_cf_annulata_SPK5Q_In
Micromeriella_cf_atropus_SPK2M_Af
Scolia_apherema_SPK2X_In
Scolia_carmichaeli_SPK6N_In
Campsomeriella_sp_SPK2V_Ma
Scolia_sp_SPK3W_In
Scolia_cf_flaviceps_SPK1R_Pa
Micromeriella_pilosella_SPK1U_Ma
Megascolia_cf_procer_SPK1Q_In
Megascolia_sp_SPK5M_In
Megameris_soleata_SPK1F_Af
Dielis_plumipes_fossulana_OE10_Nr
Scolia_decorata_SPK5K_In/Pa
Megascolia_capitata_SPK1A_In
Megascolia_cf_hageni_SPK3V_In
Micromeriella_hyalina_SPK2K_Af
Colpa_alcione_SPK1N_Nr
Scolia_dubia_dubia_SPK3S_Nr
Scolia_sp_SPK5D_Af
Cathimeris_hymenaea_SPK2L_Af
Xanthocampsomeris_limosa_SPK3L_Nr
Scolia_nobilitata_SPK2F_Nr
Scolia_cf_affinis_SPK2D_In
Scolia_sp_SPK5C_Af
Megacampsomeris_schulthessi_SPK5T_In/Pa
Dielis_dorsata_SPK3K_Nr
Scolia_apherema_SPK6L_In
Dielis_trifasciata_SPK3I_Nr
Scolia_cf_affinis_SPK6M_In
Tristimeris_cf_javana_SPK5O_In
Scolia_bicincta_P1_SPK5W_Nr
Megascolia_cf_hageni_SPK2H_In
Scolia_bicincta_SPK3U_Nr
Pygodasis_ephippium_SPK2Z_Nr/Nt
Micromeriella_cf_pilosella_SPK1M_Af
Laevicampsomeris_formosa_SPK2N_As
Proscolia_sp_EX568_Pa
Scolia_sauteri_SPK5R_In
Colpa_octomaculata_EZ3_Nr
Figure 6. Bayesian MAP tree based on 31 UCE loci after data filtering using posterior predictive checks
(analysis 2a). All unlabeled internal nodes have posterior probabilities of 1.0. Paraphyletic Campsomerini
highlighted in blue; Scoliini highlighted in orange.
0.97
0.99
0.85
0.87
0.99
0.72
0.71
.CC-BY-NC-ND 4.0 International licenseavailable under a
(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprintthis version posted June 26, 2022. ; https://doi.org/10.1101/2022.01.24.474473doi: bioRxiv preprint
Figure 7. Comparison of split posterior probabilities between two independent MCMC runs: (A) analysis 2a; (B)
analysis 2b, root calibration only; (C) analysis 2b, root + Megacampsomeris calibration; (D) analysis 2b, root +
Megacampsomeris + Scoliidae calibration. Numbers in the top left corners represent R2.
A B
C D
.CC-BY-NC-ND 4.0 International licenseavailable under a
(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprintthis version posted June 26, 2022. ; https://doi.org/10.1101/2022.01.24.474473doi: bioRxiv preprint
Figure 8. Bayesian MAP chronogram based on 63 loci after data filtering using posterior predictive checks
(analysis 2b). Node bars represent age 95% HPD intervals. All unlabeled internal nodes have posterior
probability of 0.97 or greater. Taxonomic labels indicated with .
Present
time (Ma)
-150 -140 -130 -120 -110 -100 -90 -80 -70 -60 -50 -40 -30 -20 -10
Dielis_plumipes_fossulana_OE10_Nr
Micromeriella_cf_pilosella_SPK1M_Af
Colpa_alcione_SPK1N_Nr
Campsomeriella_collaris_SPK5J_In/Pa
Proscolia_sp_EX568_Pa
Tristimeris_cf_javana_SPK5O_In
Scolia_cf_nobilis_SPK5G_Pa
Campsomeriella_cf_caelebs_SPK1E_Af
Scolia_sp_SPK5C_Af
Scolia_mexicana_SPK3R_Nr
Dielis_dorsata_SPK3K_Nt
Megacamsomeris_cf_lindenii_SPK5S_In
Scolia_bicincta_P1_SPK5W_Nr
Xanthocampsomeris_limosa_SPK3L_Nr
Pygodasis_quadrimaculata_SPK3C_Nr
Trisciloa_saussurei_SPK3M_As
Scolia_bicincta_SPK3U_Nr
Carinoscolia_vittifrons_SPK5H_Pa
Campsomeriella_cf_annulata_SPK5Q_In
Scolia_cf_affinis_SPK2D_In
Scolia_carmichaeli_SPK6N_In
Megascolia_cf_hageni_SPK2H_In
Megameris_soleata_SPK1F_Af
Pyrrhoscolia_cf_usambaraensis_SPK2I_Af
Megacampsomeris_schulthessi_SPK5T_In/Pa
Dielis_plumipes_confluenta_SPK3F_Nr
Megascolia_capitata_SPK1A_In
Scolia_sp_SPK5F_Af
Dielis_tolteca_SPK2E_Nr/Nt
Scolia_dubia_dubia_SPK3S_Nr
Megascolia_sp_SPK5M_In
Megascolia_cf_procer_SPK1Q_In
Triscolia_ardens_SPK2A_Nr
Micromeriella_cf_aureola_SPK1D_Af
Campsomeriella_sp_SPK2V_Ma
Campsomeriella_sp_SPK2P_Ma
Scolia_sp_SPK2T_Af
Scolia_cf_lebongensis_SPK5P_In
Micromeriella_cf_atropus_SPK2M_Af
Scolia_nobilitata_SPK2F_Nr
Campsomeriella_sp_SPK2Q_Ma
Scolia_cf_flaviceps_SPK1R_Pa
Scolia_apherema_SPK6L_In
Micromeriella_pilosella_SPK1U_Ma
Scolia_sp_SPK5D_Af
Scolia_decorata_SPK5K_In/Pa
Campsomeriella_cf_madonensis_SPK1B_Af
Dielis_trifasciata_SPK3I_Nr
Laevicampsomeris_formosa_SPK2N_As
Scolia_apherema_SPK2X_In
Scolia_sauteri_SPK5R_In
Scolia_cf_affinis_SPK5L_In/Pa
Apterogyna_za01_Genome
Micromeriella_hyalina_SPK2K_Af
Pygodasis_ephippium_SPK2Z_Nr/Nt
Dielis_plumipes_plumipes_SPK3D_Nr
Colpa_octomaculata_EZ3_Nr
Megacampsomeris_pulchrivestita_SPK3O_In
Scolia_sp_SPK2R_Ma
Cathimeris_hymenaea_SPK2J_Af
Scolia_sp_SPK3W_In
Scolia_4-pustulata_SPK5I_In/Pa
Scolia_cf_affinis_SPK6M_In
Scolia_sp_SPK2O_Ma
Dielis_pilipes_SPK1Y_Nr
Campsomeriella_sp_SPK1W_Ma
Cathimeris_hymenaea_SPK2L_Af
Megascolia_cf_hageni_SPK3V_In
PaleogeneLate CretaceousEarly Cretaceous
0.94
Neogene
KEY:
Afrotropic (Af)
Madagascar (Ma)
Australasia (As)
Indomalaya (In)
Nearctic (Nr)
Neotropic (Nt)
Palearctic (Pa)
Scoliini
Campsomerini s.s.
Scoliinae
.CC-BY-NC-ND 4.0 International licenseavailable under a
(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprintthis version posted June 26, 2022. ; https://doi.org/10.1101/2022.01.24.474473doi: bioRxiv preprint
Figure 9. Bayesian MAP chronogram using additional calibration on crown Megacampsomeris (analysis 2b).
All unlabeled internal nodes have posterior probability of 0.96 or greater. Taxonomic labels indicated with .
-150 -140 -130 -120 -110 -100 -90 -80 -70 -60 -50 -40 -30 -20 -10
Dielis_plumipes_confluenta_SPK3F_Nr
Colpa_octomaculata_EZ3_Nr
Laevicampsomeris_formosa_SPK2N_As
Scolia_cf_affinis_SPK2D_In
Dielis_plumipes_plumipes_SPK3D_Nr
Pyrrhoscolia_cf_usambaraensis_SPK2I_Af
Scolia_sp_SPK5C_Af
Scolia_bicincta_P1_SPK5W_Nr
Scolia_cf_flaviceps_SPK1R_Pa
Apterogyna_za01_Genome
Megacampsomeris_schulthessi_SPK5T_In/Pa
Scolia_cf_nobilis_SPK5G_Pa
Micromeriella_cf_pilosella_SPK1M_Af
Megascolia_cf_hageni_SPK3V_In
Campsomeriella_sp_SPK2Q_Ma
Dielis_plumipes_fossulana_OE10_Nr
Micromeriella_cf_aureola_SPK1D_Af
Scolia_sp_SPK2T_Af
Cathimeris_hymenaea_SPK2L_Af
Megascolia_sp_SPK5M_In
Megacampsomeris_pulchrivestita_SPK3O_In
Micromeriella_pilosella_SPK1U_Ma
Tristimeris_cf_javana_SPK5O_In
Triscolia_ardens_SPK2A_Nr
Dielis_trifasciata_SPK3I_Nr
Scolia_bicincta_SPK3U_Nr
Scolia_sp_SPK3W_In
Scolia_sp_SPK2O_Ma
Scolia_carmichaeli_SPK6N_In
Scolia_sp_SPK2R_Ma
Carinoscolia_vittifrons_SPK5H_Pa
Scolia_cf_lebongensis_SPK5P_In
Xanthocampsomeris_limosa_SPK3L_Nr
Dielis_tolteca_SPK2E_Nr/Nt
Colpa_alcione_SPK1N_Nr
Pygodasis_quadrimaculata_SPK3C_Nr
Scolia_4-pustulata_SPK5I_In/Pa
Campsomeriella_cf_madonensis_SPK1B_Af
Scolia_decorata_SPK5K_In/Pa
Scolia_mexicana_SPK3R_Nr
Scolia_cf_affinis_SPK6M_In
Megascolia_cf_hageni_SPK2H_In
Campsomeriella_collaris_SPK5J_In/Pa
Pygodasis_ephippium_SPK2Z_Nr/Nt
Scolia_cf_affinis_SPK5L_In/Pa
Campsomeriella_cf_annulata_SPK5Q_In
Megacamsomeris_cf_lindenii_SPK5S_In
Trisciloa_saussurei_SPK3M_As
Megascolia_capitata_SPK1A_In
Scolia_dubia_dubia_SPK3S_Nr
Scolia_sauteri_SPK5R_In
Micromeriella_hyalina_SPK2K_Af
Campsomeriella_sp_SPK2P_Ma
Dielis_pilipes_SPK1Y_Nr
Dielis_dorsata_SPK3K_Nt
Micromeriella_cf_atropus_SPK2M_Af
Scolia_apherema_SPK6L_In
Scolia_sp_SPK5D_Af
Megameris_soleata_SPK1F_Af
Campsomeriella_sp_SPK1W_Ma
Cathimeris_hymenaea_SPK2J_Af
Megascolia_cf_procer_SPK1Q_In
Scolia_nobilitata_SPK2F_Nr
Proscolia_sp_EX568_Pa
Campsomeriella_sp_SPK2V_Ma
Scolia_sp_SPK5F_Af
Scolia_apherema_SPK2X_In
Campsomeriella_cf_caelebs_SPK1E_Af
PaleogeneLate CretaceousEarly Cretaceous Neogene
Present
0.93
time (Ma)
Scoliinae
Campsomerini s.s.
Scoliini
.CC-BY-NC-ND 4.0 International licenseavailable under a
(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprintthis version posted June 26, 2022. ; https://doi.org/10.1101/2022.01.24.474473doi: bioRxiv preprint
Figure 10. Bayesian MAP chronogram using additional calibrations on crown Megacampsomeris and crown
Scoliidae (analysis 2b). All unlabeled internal nodes have posterior probability of 0.97 or greater. Taxonomic
labels indicated with .
PaleogeneLate CretaceousEarly Cretaceous Neogene
Present
0.93
-150 -140 -130 -120 -110 -100 -90 -80 -70 -60 -50 -40 -30 -20 -10
Apterogyna_za01_Genome
Dielis_plumipes_fossulana_OE10_Nr
Micromeriella_cf_pilosella_SPK1M_Af
Megacampsomeris_schulthessi_SPK5T_In/Pa
Pygodasis_quadrimaculata_SPK3C_Nr
Campsomeriella_sp_SPK1W_Ma
Scolia_apherema_SPK2X_In
Scolia_bicincta_P1_SPK5W_Nr
Scolia_sp_SPK5C_Af
Dielis_dorsata_SPK3K_Nt
Carinoscolia_vittifrons_SPK5H_Pa
Scolia_sp_SPK2T_Af
Dielis_plumipes_confluenta_SPK3F_Nr
Triscolia_ardens_SPK2A_Nr
Megascolia_cf_procer_SPK1Q_In
Micromeriella_cf_atropus_SPK2M_Af
Cathimeris_hymenaea_SPK2L_Af
Megacamsomeris_cf_lindenii_SPK5S_In
Dielis_trifasciata_SPK3I_Nr
Campsomeriella_collaris_SPK5J_In/Pa
Megascolia_sp_SPK5M_In
Pygodasis_ephippium_SPK2Z_Nr/Nt
Scolia_4-pustulata_SPK5I_In/Pa
Scolia_carmichaeli_SPK6N_In
Dielis_plumipes_plumipes_SPK3D_Nr
Laevicampsomeris_formosa_SPK2N_As
Megacampsomeris_pulchrivestita_SPK3O_In
Pyrrhoscolia_cf_usambaraensis_SPK2I_Af
Scolia_bicincta_SPK3U_Nr
Scolia_cf_lebongensis_SPK5P_In
Campsomeriella_cf_caelebs_SPK1E_Af
Micromeriella_pilosella_SPK1U_Ma
Scolia_nobilitata_SPK2F_Nr
Trisciloa_saussurei_SPK3M_As
Micromeriella_cf_aureola_SPK1D_Af
Scolia_cf_affinis_SPK2D_In
Scolia_sp_SPK3W_In
Tristimeris_cf_javana_SPK5O_In
Scolia_cf_nobilis_SPK5G_Pa
Micromeriella_hyalina_SPK2K_Af
Megascolia_cf_hageni_SPK2H_In
Scolia_sp_SPK2R_Ma
Scolia_sauteri_SPK5R_In
Scolia_cf_flaviceps_SPK1R_Pa
Colpa_octomaculata_EZ3_Nr
Scolia_decorata_SPK5K_In/Pa
Campsomeriella_cf_annulata_SPK5Q_In
Campsomeriella_sp_SPK2V_Ma
Campsomeriella_cf_madonensis_SPK1B_Af
Campsomeriella_sp_SPK2Q_Ma
Scolia_sp_SPK2O_Ma
Campsomeriella_sp_SPK2P_Ma
Dielis_tolteca_SPK2E_Nr/Nt
Megascolia_cf_hageni_SPK3V_In
Proscolia_sp_EX568_Pa
Dielis_pilipes_SPK1Y_Nr
Xanthocampsomeris_limosa_SPK3L_Nr
Scolia_sp_SPK5F_Af
Scolia_dubia_dubia_SPK3S_Nr
Scolia_mexicana_SPK3R_Nr
Scolia_cf_affinis_SPK5L_In/Pa
Colpa_alcione_SPK1N_Nr
Megameris_soleata_SPK1F_Af
Cathimeris_hymenaea_SPK2J_Af
Megascolia_capitata_SPK1A_In
Scolia_sp_SPK5D_Af
Scolia_cf_affinis_SPK6M_In
Scolia_apherema_SPK6L_In
time (Ma)
Scoliinae
Campsomerini s.s.
Scoliini
.CC-BY-NC-ND 4.0 International licenseavailable under a
(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprintthis version posted June 26, 2022. ; https://doi.org/10.1101/2022.01.24.474473doi: bioRxiv preprint
Figure 11. MCC species or minimal clusters tree based on 4 independent MCMC chains run using STACEY
(analysis 2c). All unlabeled internal nodes have posterior probability of 1.0.
0.006
Dielis_dorsata_SPK3K_Nt
Scolia_carmichaeli_SPK6N_In
Campsomeriella_cf_caelebs_SPK1E_Af
Scolia_sp_SPK5F_Af
Scolia_sp_SPK2R_Ma
Campsomeriella_cf_madonensis_SPK1B_Af
Micromeriella_hyalina_SPK2K_Af
Laevicampsomeris_formosa_SPK2N_As
Dielis_pilipes_SPK1Y_Nr
Campsomeriella_sp_SPK2V_Ma
Scolia_sp_SPK5D_Af
Megascolia_capitata_SPK1A_In
Scolia_4-pustulata_SPK5I_In/Pa
Scolia_bicincta_P1_SPK5W_Nr
Scolia_decorata_SPK5K_In/Pa
Scolia_sp_SPK5C_Af
Scolia_sp_SPK3W_In
Scolia_sp_SPK2T_Af
Cathimeris_hymenaea_SPK2J_Af
Micromeriella_cf_aureola_SPK1D_Af
Scolia_apherema_SPK6L_In
Scolia_apherema_SPK2X_In
Triscolia_ardens_SPK2A_Nr
Scolia_cf_affinis_SPK5L_In/Pa
Campsomeriella_sp_SPK1W_Ma
Campsomeriella_sp_SPK2Q_Ma
Scolia_nobilitata_SPK2F_Nr
Scolia_cf_affinis_SPK6M_In
Megameris_soleata_SPK1F_Af
Campsomeriella_cf_annulata_SPK5Q_In
Scolia_cf_flaviceps_SPK1R_Pa
Xanthocampsomeris_limosa_SPK3L_Nr
Pyrrhoscolia_cf_usambaraensis_SPK2I_Af
Dielis_trifasciata_SPK3I_Nr
Scolia_sp_SPK2O_Ma
Pygodasis_ephippium_SPK2Z_Nr/Nt
Megacampsomeris_schulthessi_SPK5T_In/Pa
Trisciloa_saussurei_SPK3M_As
Megacamsomeris_cf_lindenii_SPK5S_In
Megascolia_cf_hageni_SPK2H_In
Scolia_mexicana_SPK3R_Nr
Scolia_sauteri_SPK5R_In
Scolia_cf_affinis_SPK2D_In
Micromeriella_pilosella_SPK1U_Ma
Dielis_tolteca_SPK2E_Nr/Nt
Micromeriella_cf_atropus_SPK2M_Af
Dielis_plumipes_confluenta_SPK3F_Nr
Campsomeriella_collaris_SPK5J_In/Pa
Scolia_cf_nobilis_SPK5G_Pa
Carinoscolia_vittifrons_SPK5H_Pa
Megacampsomeris_pulchrivestita_SPK3O_In
Micromeriella_cf_pilosella_SPK1M_Af
Colpa_alcione_SPK1N_Nr
Dielis_plumipes_fossulana_OE10_Nr
Cathimeris_hymenaea_SPK2L_Af
Megascolia_sp_SPK5M_In
Megascolia_cf_procer_SPK1Q_In
Scolia_dubia_dubia_SPK3S_Nr
Scolia_cf_lebongensis_SPK5P_In
Proscolia_sp_EX568_Pa
Megascolia_cf_hageni_SPK3V_In
Campsomeriella_sp_SPK2P_Ma
Apterogyna_za01_Genome
Colpa_octomaculata_EZ3_Nr
Pygodasis_quadrimaculata_SPK3C_Nr
Tristimeris_cf_javana_SPK5O_In
Scolia_bicincta_SPK3U_Nr
Dielis_plumipes_plumipes_SPK3D_Nr
0.36
0.99
0.45
0.82
0.73
0.68
0.99
0.99
0.83
0.49
0.85
0.92
0.99
0.76
0.83
.CC-BY-NC-ND 4.0 International licenseavailable under a
(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprintthis version posted June 26, 2022. ; https://doi.org/10.1101/2022.01.24.474473doi: bioRxiv preprint
Figure 12. Comparison of split posterior probabilities between four independent MCMC runs using STACEY
(analysis 2c). Numbers in the top left corners represent R2.
.CC-BY-NC-ND 4.0 International licenseavailable under a
(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprintthis version posted June 26, 2022. ; https://doi.org/10.1101/2022.01.24.474473doi: bioRxiv preprint
Scolia_sp_SPK5C_Af
Micromeriella_pilosella_SPK1U_Ma
Megascolia_cf_hageni_SPK3V_In
Campsomeriella_collaris_SPK5J_In/Pa
Megacampsomeris_pulchrivestita_SPK3O_In
Cathimeris_hymenaea_SPK2L_Af
Scolia_dubia_dubia_SPK3S_Nr
Scolia_sp_SPK2O_Ma
Trisciloa_saussurei_SPK3M_As
Dielis_tolteca_SPK2E_Nr/Nt
Scolia_sp_SPK5D_Af
Colpa_alcione_SPK1N_Nr
Micromeriella_cf_pilosella_SPK1M_Af
Campsomeriella_cf_madonensis_SPK1B_Af
Megascolia_sp_SPK5M_In
Scolia_sp_SPK2R_Ma
Scolia_bicincta_P1_SPK5W_Nr
Scolia_cf_nobilis_SPK5G_Pa
Scolia_apherema_SPK2X_In
Scolia_decorata_SPK5K_In/Pa
Scolia_apherema_SPK6L_In
Pyrrhoscolia_cf_usambaraensis_SPK2I_Af
Megascolia_cf_procer_SPK1Q_In
Apterogyna_za01_Genome
Campsomeriella_sp_SPK2P_Ma
Scolia_sp_SPK3W_In
Megascolia_capitata_SPK1A_In
Megameris_soleata_SPK1F_Af
Campsomeriella_cf_annulata_SPK5Q_In
Micromeriella_cf_atropus_SPK2M_Af
Scolia_4-pustulata_SPK5I_In/Pa
Laevicampsomeris_formosa_SPK2N_As
Megacamsomeris_cf_lindenii_SPK5S_In
Scolia_cf_affinis_SPK2D_In
Dielis_plumipes_fossulana_OE10_Nr
Colpa_octomaculata_EZ3_Nr
Pygodasis_quadrimaculata_SPK3C_Nr
Campsomeriella_sp_SPK2Q_Ma
Scolia_sauteri_SPK5R_In
Dielis_pilipes_SPK1Y_Nr
Proscolia_sp_EX568_Pa
Scolia_nobilitata_SPK2F_Nr
Scolia_cf_flaviceps_SPK1R_Pa
Micromeriella_cf_aureola_SPK1D_Af
Scolia_sp_SPK5F_Af
Tristimeris_cf_javana_SPK5O_In
Triscolia_ardens_SPK2A_Nr
Megascolia_cf_hageni_SPK2H_In
Xanthocampsomeris_limosa_SPK3L_Nr
Scolia_cf_affinis_SPK6M_In
Campsomeriella_sp_SPK2V_Ma
Cathimeris_hymenaea_SPK2J_Af
Dielis_plumipes_plumipes_SPK3D_Nr
Dielis_plumipes_confluenta_SPK3F_Nr
Scolia_cf_affinis_SPK5L_In/Pa
Scolia_cf_lebongensis_SPK5P_In
Scolia_bicincta_SPK3U_Nr
Micromeriella_hyalina_SPK2K_Af
Scolia_sp_SPK2T_Af
Dielis_dorsata_SPK3K_Nt
Dielis_trifasciata_SPK3I_Nr
Megacampsomeris_schulthessi_SPK5T_In/Pa
Pygodasis_ephippium_SPK2Z_Nr/Nt
Carinoscolia_vittifrons_SPK5H_Pa
Campsomeriella_sp_SPK1W_Ma
Scolia_mexicana_SPK3R_Nr
Scolia_carmichaeli_SPK6N_In
Campsomeriella_cf_caelebs_SPK1E_Af
Campsomeriella_sp_SPK2P_Ma
Scolia_apherema_SPK6L_In
Scolia_cf_affinis_SPK2D_In
Scolia_sp_SPK2O_Ma
Scolia_sp_SPK2R_Ma
Trisciloa_saussurei_SPK3M_As
Campsomeriella_cf_annulata_SPK5Q_In
Dielis_plumipes_plumipes_SPK3D_Nr
Megacampsomeris_pulchrivestita_SPK3O_In
Scolia_sp_SPK5D_Af
Scolia_cf_flaviceps_SPK1R_Pa
Pyrrhoscolia_cf_usambaraensis_SPK2I_Af
Micromeriella_cf_aureola_SPK1D_Af
Dielis_dorsata_SPK3K_Nt
Dielis_trifasciata_SPK3I_Nr
Scolia_apherema_SPK2X_In
Apterogyna_za01_Genome
Dielis_plumipes_confluenta_SPK3F_Nr
Megascolia_cf_hageni_SPK3V_In
Proscolia_sp_EX568_Pa
Megascolia_sp_SPK5M_In
Scolia_bicincta_SPK3U_Nr
Tristimeris_cf_javana_SPK5O_In
Scolia_bicincta_P1_SPK5W_Nr
Pygodasis_quadrimaculata_SPK3C_Nr
Xanthocampsomeris_limosa_SPK3L_Nr
Cathimeris_hymenaea_SPK2J_Af
Scolia_cf_affinis_SPK6M_In
Dielis_tolteca_SPK2E_Nr/Nt
Scolia_sp_SPK2T_Af
Scolia_mexicana_SPK3R_Nr
Scolia_dubia_dubia_SPK3S_Nr
Colpa_octomaculata_EZ3_Nr
Megacampsomeris_schulthessi_SPK5T_In/Pa
Colpa_alcione_SPK1N_Nr
Campsomeriella_sp_SPK2V_Ma
Megascolia_cf_procer_SPK1Q_In
Scolia_cf_nobilis_SPK5G_Pa
Micromeriella_pilosella_SPK1U_Ma
Micromeriella_cf_pilosella_SPK1M_Af
Scolia_decorata_SPK5K_In/Pa
Dielis_plumipes_fossulana_OE10_Nr
Scolia_sp_SPK5F_Af
Campsomeriella_cf_caelebs_SPK1E_Af
Megameris_soleata_SPK1F_Af
Megascolia_capitata_SPK1A_In
Scolia_nobilitata_SPK2F_Nr
Scolia_cf_affinis_SPK5L_In/Pa
Pygodasis_ephippium_SPK2Z_Nr/Nt
Scolia_sp_SPK3W_In
Scolia_carmichaeli_SPK6N_In
Scolia_4-pustulata_SPK5I_In/Pa
Campsomeriella_sp_SPK1W_Ma
Triscolia_ardens_SPK2A_Nr
Scolia_cf_lebongensis_SPK5P_In
Scolia_sauteri_SPK5R_In
Dielis_pilipes_SPK1Y_Nr
Laevicampsomeris_formosa_SPK2N_As
Scolia_sp_SPK5C_Af
Micromeriella_hyalina_SPK2K_Af
Micromeriella_cf_atropus_SPK2M_Af
Carinoscolia_vittifrons_SPK5H_Pa
Campsomeriella_cf_madonensis_SPK1B_Af
Cathimeris_hymenaea_SPK2L_Af
Campsomeriella_sp_SPK2Q_Ma
Megacamsomeris_cf_lindenii_SPK5S_In
Megascolia_cf_hageni_SPK2H_In
Campsomeriella_collaris_SPK5J_In/Pa
Scolia_sp_SPK2R_Ma
Laevicampsomeris_formosa_SPK2N_As
Dielis_plumipes_confluenta_SPK3F_Nr
Megascolia_cf_hageni_SPK3V_In
Scolia_cf_affinis_SPK6M_In
Megascolia_cf_procer_SPK1Q_In
Scolia_sp_SPK5C_Af
Micromeriella_cf_atropus_SPK2M_Af
Megascolia_capitata_SPK1A_In
Scolia_sauteri_SPK5R_In
Proscolia_sp_EX568_Pa
Megameris_soleata_SPK1F_Af
Dielis_tolteca_SPK2E_Nr/Nt
Scolia_cf_affinis_SPK5L_In/Pa
Dielis_trifasciata_SPK3I_Nr
Megascolia_cf_hageni_SPK2H_In
Megacampsomeris_pulchrivestita_SPK3O_In
Pygodasis_ephippium_SPK2Z_Nr/Nt
Dielis_plumipes_fossulana_OE10_Nr
Scolia_sp_SPK2T_Af
Apterogyna_za01_Genome
Cathimeris_hymenaea_SPK2L_Af
Scolia_cf_nobilis_SPK5G_Pa
Dielis_plumipes_plumipes_SPK3D_Nr
Micromeriella_hyalina_SPK2K_Af
Carinoscolia_vittifrons_SPK5H_Pa
Scolia_cf_flaviceps_SPK1R_Pa
Scolia_sp_SPK5F_Af
Micromeriella_cf_pilosella_SPK1M_Af
Campsomeriella_sp_SPK2Q_Ma
Megacamsomeris_cf_lindenii_SPK5S_In
Cathimeris_hymenaea_SPK2J_Af
Scolia_sp_SPK2O_Ma
Triscolia_ardens_SPK2A_Nr
Trisciloa_saussurei_SPK3M_As
Scolia_carmichaeli_SPK6N_In
Scolia_bicincta_P1_SPK5W_Nr
Scolia_cf_affinis_SPK2D_In
Campsomeriella_cf_caelebs_SPK1E_Af
Scolia_apherema_SPK2X_In
Scolia_nobilitata_SPK2F_Nr
Campsomeriella_sp_SPK2V_Ma
Scolia_sp_SPK3W_In
Scolia_decorata_SPK5K_In/Pa
Tristimeris_cf_javana_SPK5O_In
Megascolia_sp_SPK5M_In
Scolia_mexicana_SPK3R_Nr
Campsomeriella_sp_SPK1W_Ma
Scolia_sp_SPK5D_Af
Campsomeriella_cf_madonensis_SPK1B_Af
Campsomeriella_collaris_SPK5J_In/Pa
Scolia_cf_lebongensis_SPK5P_In
Dielis_dorsata_SPK3K_Nt
Colpa_octomaculata_EZ3_Nr
Micromeriella_pilosella_SPK1U_Ma
Megacampsomeris_schulthessi_SPK5T_In/Pa
Pygodasis_quadrimaculata_SPK3C_Nr
Scolia_dubia_dubia_SPK3S_Nr
Xanthocampsomeris_limosa_SPK3L_Nr
Campsomeriella_cf_annulata_SPK5Q_In
Colpa_alcione_SPK1N_Nr
Scolia_bicincta_SPK3U_Nr
Scolia_apherema_SPK6L_In
Campsomeriella_sp_SPK2P_Ma
Pyrrhoscolia_cf_usambaraensis_SPK2I_Af
Micromeriella_cf_aureola_SPK1D_Af
Scolia_4-pustulata_SPK5I_In/Pa
Dielis_pilipes_SPK1Y_Nr
Scolia_sp_SPK2R_Ma
Laevicampsomeris_formosa_SPK2N_As
Dielis_plumipes_confluenta_SPK3F_Nr
Megascolia_cf_hageni_SPK3V_In
Scolia_cf_affinis_SPK6M_In
Megascolia_cf_procer_SPK1Q_In
Scolia_sp_SPK5C_Af
Micromeriella_cf_atropus_SPK2M_Af
Megascolia_capitata_SPK1A_In
Scolia_sauteri_SPK5R_In
Proscolia_sp_EX568_Pa
Megameris_soleata_SPK1F_Af
Dielis_tolteca_SPK2E_Nr/Nt
Scolia_cf_affinis_SPK5L_In/Pa
Dielis_trifasciata_SPK3I_Nr
Megascolia_cf_hageni_SPK2H_In
Megacampsomeris_pulchrivestita_SPK3O_In
Pygodasis_ephippium_SPK2Z_Nr/Nt
Dielis_plumipes_fossulana_OE10_Nr
Scolia_sp_SPK2T_Af
Apterogyna_za01_Genome
Cathimeris_hymenaea_SPK2L_Af
Scolia_cf_nobilis_SPK5G_Pa
Dielis_plumipes_plumipes_SPK3D_Nr
Micromeriella_hyalina_SPK2K_Af
Carinoscolia_vittifrons_SPK5H_Pa
Scolia_cf_flaviceps_SPK1R_Pa
Scolia_sp_SPK5F_Af
Micromeriella_cf_pilosella_SPK1M_Af
Campsomeriella_sp_SPK2Q_Ma
Megacamsomeris_cf_lindenii_SPK5S_In
Cathimeris_hymenaea_SPK2J_Af
Scolia_sp_SPK2O_Ma
Triscolia_ardens_SPK2A_Nr
Trisciloa_saussurei_SPK3M_As
Scolia_carmichaeli_SPK6N_In
Scolia_bicincta_P1_SPK5W_Nr
Scolia_cf_affinis_SPK2D_In
Campsomeriella_cf_caelebs_SPK1E_Af
Scolia_apherema_SPK2X_In
Scolia_nobilitata_SPK2F_Nr
Campsomeriella_sp_SPK2V_Ma
Scolia_sp_SPK3W_In
Scolia_decorata_SPK5K_In/Pa
Tristimeris_cf_javana_SPK5O_In
Megascolia_sp_SPK5M_In
Scolia_mexicana_SPK3R_Nr
Campsomeriella_sp_SPK1W_Ma
Scolia_sp_SPK5D_Af
Campsomeriella_cf_madonensis_SPK1B_Af
Campsomeriella_collaris_SPK5J_In/Pa
Scolia_cf_lebongensis_SPK5P_In
Dielis_dorsata_SPK3K_Nt
Colpa_octomaculata_EZ3_Nr
Micromeriella_pilosella_SPK1U_Ma
Megacampsomeris_schulthessi_SPK5T_In/Pa
Pygodasis_quadrimaculata_SPK3C_Nr
Scolia_dubia_dubia_SPK3S_Nr
Xanthocampsomeris_limosa_SPK3L_Nr
Campsomeriella_cf_annulata_SPK5Q_In
Colpa_alcione_SPK1N_Nr
Scolia_bicincta_SPK3U_Nr
Scolia_apherema_SPK6L_In
Campsomeriella_sp_SPK2P_Ma
Pyrrhoscolia_cf_usambaraensis_SPK2I_Af
Micromeriella_cf_aureola_SPK1D_Af
Scolia_4-pustulata_SPK5I_In/Pa
Dielis_pilipes_SPK1Y_Nr
A B
DC
0.78
0.65
0.99
0.58
0.77
0.82
0.75
94
73
99
95
98
74
54.1
0.72
0.86
0.52
0.98
0.54
0.97
Figure 13. ASTRAL species trees (analysis 2d) based on MCC trees of (A) loci having the lowest (1/3)
combined posterior predictive effect sizes, (C) loci having the highest (1/3) combined posterior predictive effect
sizes, and (D) all loci. (B) is an ASTRAL bootstrap consensus tree using posterior samples of gene trees from
loci having the lowest (1/3) combined posterior predictive effect sizes. Branch labels represent the local
posterior probability or bootstrap support of/for the associated quadripartition.
.CC-BY-NC-ND 4.0 International licenseavailable under a
(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprintthis version posted June 26, 2022. ; https://doi.org/10.1101/2022.01.24.474473doi: bioRxiv preprint
0.02
Campsomeriella_sp_SPK2Q_Ma
Scolia_picteti_SPK4Z_In
Scolia_carmichaeli_SPK6N_In
Megacampsomeris_pulchrivestita_SPK2Y_In
Scolia_dubia_dubia_SPK3S_Nr
Campsomeriella_sp_SPK1W_Ma
Trisciloa_saussurei_SPK3M_As
Micromeriella_pilosella_SPK1U_Ma
Dielis_trifasciata_SPK3I_Nr
Scolia_cf_nobilis_SPK5G_Pa
Dielis_plumipes_plumipes_SPK3D_Nr
Cathimeris_hymenaea_SPK2L_Af
Scolia_cf_flaviceps_SPK1R_Pa
Scolia_sp_SPK5C_Af
Tristimeris_cf_javana_SPK5O_In
Scolia_cf_affinis_SPK5A_In
Scolia_cf_affinis_SPK5L_In/Pa
Pygodasis_quadrimaculata_SPK3C_Nr
Cathimeris_hymenaea_SPK2J_Af
Pyrrhoscolia_cf_usambaraensis_SPK2I_Af
Scolia_cf_lebongensis_SPK5P_In
Scolia_sp_SPK2R_Ma
Xanthocampsomeris_limosa_SPK3L_Nr
Dielis_dorsata_SPK3K_Nt
Scolia_sp_SPK5F_Af
Megacamsomeris_cf_lindenii_SPK5S_In
Scolia_decorata_SPK5K_In/Pa
Micromeriella_cf_aureola_SPK1D_Af
Micromeriella_cf_pilosella_SPK1M_Af
Campsomeriella_cf_annulata_SPK5Q_In
Triscolia_ardens_SPK2A_Nr
Megascolia_sp_SPK5M_In
Scolia_mexicana_SPK3R_Nr
Micromeriella_marginella_SPK6K_In
Scolia_4-pustulata_SPK5I_In/Pa
Campsomeriella_collaris_SPK5J_In/Pa
Dielis_tolteca_SPK2E_Nr/Nt
Megascolia_cf_azurea_SPK6J_In
Campsomeriella_cf_madonensis_SPK1B_Af
Megascolia_cf_hageni_SPK3V_In
Micromeriella_hyalina_SPK2K_Af
Megascolia_cf_procer_SPK1Q_In
Pygodasis_ephippium_SPK2Z_Nr/Nt
Scolia_apherema_SPK2X_In
Colpacampsomeris_indica_SPK5N_In
Campsomeriella_sp_SPK2P_Ma
Scolia_nobilitata_SPK2F_Nr
Scolia_sauteri_SPK5R_In
Scolia_bicincta_P1_SPK5W_Nr
Dielis_pilipes_SPK1Y_Nr
Scolia_sp_SPK2O_Ma
Megacampsomeris_pulchrivestita_SPK3O_In
Scolia_apherema_SPK6L_In
Scolia_cf_affinis_SPK2D_In
Megameris_soleata_SPK1F_Af
Scolia_sp_SPK5D_Af
Scolia_sp_SPK2T_Af
Carinoscolia_vittifrons_SPK5H_Pa
Megascolia_cf_hageni_SPK2H_In
Dielis_plumipes_confluenta_SPK3F_Nr
Micromeriella_cf_atropus_SPK2M_Af
Scolia_cf_affinis_SPK6M_In
Scolia_bicincta_SPK3U_Nr
Laevicampsomeris_formosa_SPK2N_As
Campsomeriella_sp_SPK2V_Ma
Scolia_sp_SPK3W_In
Campsomeriella_cf_caelebs_SPK1E_Af
Dielis_plumipes_fossulana_OE10_Nr
Colpa_alcione_SPK1N_Nr
Colpa_octomaculata_EZ3_Nr
Megascolia_capitata_SPK1A_In
Megacampsomeris_schulthessi_SPK5T_In/Pa
Figure 14. Bayesian MAP tree based on 115 UCE loci after data filtering using posterior predictive checks
(analysis 3a). All unlabeled internal nodes have posterior probabilities of 1.0. Comparison of split posterior
probabilities between two independent MCMC runs on lower left.
0.98
0.98
.CC-BY-NC-ND 4.0 International licenseavailable under a
(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprintthis version posted June 26, 2022. ; https://doi.org/10.1101/2022.01.24.474473doi: bioRxiv preprint
20.0
Megacampsomeris_pulchrivestita_SPK2Y_In
Campsomeriella_sp_SPK2Q_Ma
Micromeriella_marginella_SPK6K_In
Megascolia_cf_hageni_SPK3V_In
Campsomeriella_cf_annulata_SPK5Q_In
Trisciloa_saussurei_SPK3M_As
Pyrrhoscolia_cf_usambaraensis_SPK2I_Af
Megascolia_sp_SPK5M_In
Megacamsomeris_cf_lindenii_SPK5S_In
Dielis_plumipes_fossulana_OE10_Nr
Campsomeriella_collaris_SPK5J_In/Pa
Campsomeriella_cf_caelebs_SPK1E_Af
Campsomeriella_sp_SPK2P_Ma
Micromeriella_pilosella_SPK1U_Ma
Megascolia_cf_procer_SPK1Q_In
Scolia_cf_flaviceps_SPK1R_Pa
Scolia_sp_SPK2R_Ma
Micromeriella_cf_atropus_SPK2M_Af
Scolia_nobilitata_SPK2F_Nr
Scolia_4-pustulata_SPK5I_In/Pa
Dielis_plumipes_plumipes_SPK3D_Nr
Scolia_cf_nobilis_SPK5G_Pa
Scolia_sp_SPK5D_Af
Scolia_picteti_SPK4Z_In
Colpa_alcione_SPK1N_Nr
Scolia_sauteri_SPK5R_In
Micromeriella_cf_pilosella_SPK1M_Af
Megascolia_capitata_SPK1A_In
Scolia_apherema_SPK2X_In
Campsomeriella_sp_SPK1W_Ma
Dielis_pilipes_SPK1Y_Nr
Cathimeris_hymenaea_SPK2L_Af
Scolia_sp_SPK3W_In
Scolia_sp_SPK5C_Af
Dielis_dorsata_SPK3K_Nt
Colpa_octomaculata_EZ3_Nr
Megacampsomeris_pulchrivestita_SPK3O_In
Scolia_sp_SPK5F_Af
Dielis_tolteca_SPK2E_Nr/Nt
Scolia_mexicana_SPK3R_Nr
Scolia_bicincta_P1_SPK5W_Nr
Pygodasis_quadrimaculata_SPK3C_Nr
Scolia_carmichaeli_SPK6N_In
Scolia_bicincta_SPK3U_Nr
Scolia_apherema_SPK6L_In
Scolia_sp_SPK2O_Ma
Campsomeriella_cf_madonensis_SPK1B_Af
Campsomeriella_sp_SPK2V_Ma
Cathimeris_hymenaea_SPK2J_Af
Colpacampsomeris_indica_SPK5N_In
Micromeriella_hyalina_SPK2K_Af
Scolia_dubia_dubia_SPK3S_Nr
Scolia_decorata_SPK5K_In/Pa
Scolia_cf_affinis_SPK5A_In
Scolia_sp_SPK2T_Af
Xanthocampsomeris_limosa_SPK3L_Nr
Scolia_cf_affinis_SPK6M_In
Pygodasis_ephippium_SPK2Z_Nr/Nt
Megacampsomeris_schulthessi_SPK5T_In/Pa
Micromeriella_cf_aureola_SPK1D_Af
Scolia_cf_lebongensis_SPK5P_In
Scolia_cf_affinis_SPK5L_In/Pa
Scolia_cf_affinis_SPK2D_In
Laevicampsomeris_formosa_SPK2N_As
Megameris_soleata_SPK1F_Af
Megascolia_cf_hageni_SPK2H_In
Megascolia_cf_azurea_SPK6J_In
Carinoscolia_vittifrons_SPK5H_Pa
Triscolia_ardens_SPK2A_Nr
Tristimeris_cf_javana_SPK5O_In
Dielis_trifasciata_SPK3I_Nr
Dielis_plumipes_confluenta_SPK3F_Nr
Figure 15. Bayesian MAP relative-time chronogram based on 159 UCE loci after data filtering using posterior
predictive checks (analysis 3b). All unlabeled internal nodes have posterior probabilities of 1.0. Comparison of
split posterior probabilities between two independent MCMC runs on lower left.
0.67
.CC-BY-NC-ND 4.0 International licenseavailable under a
(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprintthis version posted June 26, 2022. ; https://doi.org/10.1101/2022.01.24.474473doi: bioRxiv preprint
Campsoscolia
Campsomeris
Trisciloa
Scolia
Scoliinae
Trielidini
Campsomerini
Campsomerinae
Proscoliinae
Scoliini
Campsomerini
Scoliinae
Campsomerinae
Proscoliinae
Colpinae
Scoliinae
Proscoliinae
Campsomerini s.s.
Colpa
Scoliini
Figure 16. Hypotheses regarding the
relationships among major scoliid lineages: (A)
Bradley (1950a); (B) Betrem (1965), Betrem &
Bradley (1972); (C) Rasnitsyn (1977), Day et al.
(1981), Osten (2005); (D) Argaman (1996); (E)
current study. Taxa containing species currently
(Osten, 2005) in Colpa, Dasyscolia, and/or
Guigliana marked with .
A
C
B
D
E
.CC-BY-NC-ND 4.0 International licenseavailable under a
(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprintthis version posted June 26, 2022. ; https://doi.org/10.1101/2022.01.24.474473doi: bioRxiv preprint
