Content uploaded by Charles S Henry
Author content
All content in this area was uploaded by Charles S Henry on Oct 09, 2017
Content may be subject to copyright.
The function and evolution of male and female genitalia in
Phyllophaga Harris scarab beetles (Coleoptera: Scarabaeidae)
M. P. RICHMOND*
,†,1
, J. PARK
†,‡,1
&C.S.HENRY*
*Department of Ecology and Evolutionary Biology, University of Connecticut, Storrs, CT, USA
†Division of Biological Sciences, UC San Diego, La Jolla, New York, CA, USA
‡Institute for Genomic Medicine, Columbia University, New York, NY, USA
Keywords:
genitalia evolution;
genitalia function;
genitalia mechanics;
genitalic diversity;
morphological evolution.
Abstract
Genitalia diversity in insects continues to fuel investigation of the function
and evolution of these dynamic structures. Whereas most studies have
focused on variation in male genitalia, an increasing number of studies on
female genitalia have uncovered comparable diversity among females, but
often at a much finer morphological scale. In this study, we analysed the func-
tion and evolution of male and female genitalia in Phyllophaga scarab beetles,
a group in which both sexes exhibit genitalic diversity. To document the inter-
action between male and female structures during mating, we dissected flash-
frozen mating pairs from three Phyllophaga species and investigated fine-scale
morphology using SEM. We then reconstructed ancestral character states
using a species tree inferred from mitochondrial and nuclear loci to elucidate
and compare the evolutionary history of male and female genitalia. Our dis-
sections revealed an interlocking mechanism of the female pubic process and
male parameres that appears to improve the mechanical fit of the copulatory
position. The comparative analyses, however, did not support coevolution of
male and female structures and showed more erratic evolution of the female
genitalia relative to males. By studying a group that exhibits obvious female
genitalic diversity, we were able to demonstrate the relevance of female repro-
ductive morphology in studies of male genital diversity.
Introduction
Insect genitalia are diverse morphological structures that
have various roles mediating male–female copulatory
interactions. The extensive literature base documenting
and investigating genitalia diversity has revealed a variety
of evolutionary forces acting on these structures. Cover-
ing a broad range of insect taxa, most studies support
selection on genital structures for either mate recognition
(Robertson & Paterson, 1982; Sota & Kubota, 1998;
McPeek et al., 2008; Tanabe & Sota, 2008; Masly, 2012;
Wojcieszek & Simmons, 2012, 2013; Richmond, 2014),
sexual selection (West-Eberhard, 1983; Eberhard, 1985,
1997, 2010; Hosken & Stockley, 2004) or antagonistic
coevolution (Parker, 1979; reviewed in Arnqvist & Rowe,
2005). Further, these studies reveal that patterns and
modes of selection on reproductive traits vary widely, not
only among taxa, but also among reproductive characters
of the same species (Werner & Simmons, 2008). When
such factors are considered collectively, it is clear we are
gaining a better understanding of the selective factors
operating on genital structures. However, an overarching
explanation for the widespread phenomenon of species-
specific genitalia remains elusive, and we continue to ask
why genitalia are so diverse.
Male genitalia of insects tend to be heavily sclerotized,
rigid structures that are seemingly more diverse than
female reproductive morphology. Because species-speci-
fic male traits are more apparent to human investigators,
they are usually the primary focus of genital evolution
studies. However, most (if not all) male genitalia struc-
tures interact in some way with the female reproductive
system during copulation (e.g. Jagadeeshan & Singh,
Correspondence: Maxi P. Richmond, University of California, San Diego,
9500 Gilman Dr., La Jolla, CA 92093, USA.
Tel.: 858 246 0350; fax: 858 534 7108;
e-mail: maxi.p.richmond@gmail.com
1
Present address: University of California, San Diego, 9500 Gilman Dr.,
La Jolla, CA, USA
ª2016 EUROPEAN SOCIETY FOR EVOLUTIONARY BIOLOGY. J. EVOL. BIOL.
1
JOURNAL OF EVOLUTIONARY BIOLOGY ª2016 EUROPEAN SOCIETY FOR EVOLUTIONARY BIOLOGY
doi: 10.1111/jeb.12955
2006; Wojcieszek et al., 2012), and some recent studies
provide evidence that female traits are more diverse than
previously recognized (Polihronakis, 2006; Puniamoor-
thy et al., 2010; Yassin & Orgogozo, 2013; Mattei et al.,
2015). For example, the presence of microgrooves and
pockets (Yassin & Orgogozo, 2013) and female neuronal
responses (Heifetz & Wolfner, 2004; Mattei et al., 2015)
indicate that different aspects of female reproductive
anatomy and physiology may be involved in sensory and
mechanical copulatory mechanisms. Those findings sug-
gest that an understanding of how and why genital struc-
tures evolve is best achieved from the perspective of both
sexes, and by considering structural interactions of the
genitalia during mating (McPeek et al., 2009; Eberhard,
2011; Wojcieszek et al., 2012).
Here, we investigate diversity of male and female geni-
talia in the scarab beetle genus Phyllophaga Harris by
integrating functional data from copulating pairs with
analyses of comparative morphological evolution. In this
genus, both male and female genitalia are obviously spe-
cies-specific and heavily sclerotized (Smith, 1888). In
males, there are two aspects of the genitalia that are
highly diverse. Our study focuses on the heavily sclero-
tized male parameres used for species identification
(Fig. 1). These paired structures are typically inserted
into the female pygidium during copulation, and take on
a variety of shapes. They can be fused and form a tube or
unfused; if the parameres are unfused, they can be sym-
metric or asymmetric and adorned with a variety of
hooks and internal and external processes. The aedeagus,
or internal sac, is the sperm transmission organ of the
male genitalia and also exhibits extensive diversity.
Although mostly membranous, the aedeagus can assume
different shapes and is often adorned with various types
of sclerotized spines. Previous analyses of copulatory
behaviour in Phyllophaga suggested that species-specific
structures associated with the aedeagus function to hold
and/or stimulate the female (Eberhard, 1993).
The species specificity of the female genitalia in Phyl-
lophaga largely results from the shape of the pubic pro-
cess, which can be fused or unfused with varying
degrees of bifurcation (Fig. 2). In both the male and
female, there is fine-scale microsculpturing present on
many structures that can be visualized with SEM.
Whereas the form of male parameres and female pubic
process have been scrutinized from a taxonomic per-
spective (Smith, 1888; Glasgow, 1916; B€
oving, 1942;
Luginbill & Painter, 1953; Woodruff & Beck, 1989),
there are no data describing how these structures inter-
act during copulation. Through analysis of the male–
female mating complex, our goal is to determine the
function of the parameres and pubic process, and test
the hypothesis that these structures are coevolving.
We investigated the function of male and female
genitalia by flash-freezing mating pairs of several spe-
cies caught in wild populations and dissecting the inter-
acting components. We analysed fine-scale attributes of
the male aedeagus and internal female reproductive
tract anatomy using scanning electron microscopy
(SEM). We then conducted comparative morphological
(a) (b) (c) (d) (e)
Fig. 1 Character states of male parameres shown from the caudal perspective with corresponding character state symbols shown in Fig. 6;
a) state 0 –fused, symmetric (Phyllophaga crenulata); b) state 1 –unfused, symmetric (Phyllophaga parvidens); c) state 2 –unfused,
asymmetric shape (Phyllophaga perlonga); d) state 3 –unfused, asymmetric size (Phyllophaga praetermissa); e) state 4 –unfused, asymmetric
shape and size (Phyllophaga ilicis).
(a) (b) (c) (d)
Fig. 2 Character states of female pubic process shown from the ventral perspective with corresponding character state symbols shown in
Fig. 6; a) state 0 –platelike (P. crenulata); b) state 1 –elongate (Phyllophaga fusca), no bifurcation; c) state 2 –bifurcation at tip
(Phyllophaga hirticula); d) state 3 –bifurcating from base (Phyllophaga luctuosa).
ª2016 EUROPEAN SOCIETY FOR EVOLUTIONARY BIOLOGY. J. EVOL. BIOL. doi: 10.1111/jeb.12955
JOURNAL OF EVOLUTIONARY BIOLOGY ª2016 EUROPEAN SOCIETY FOR EVOLUTIONARY BIOLOGY
2M. P. RICHMOND ET AL.
analyses to establish the evolutionary trajectory of male
and female genitalia. We reconstructed ancestral
character states on a species tree to determine (i) the
direction of change of male and female genitalia, (ii)
the sequence of structural transitions, and (iii) whether
male and female genitalia are coevolving. Incorporation
of female morphology into our functional and compara-
tive analyses provides an unusual look into how diverse
male and female structures interact and evolve. Our
results revealed a unique interlocking mechanism
between the male and female genitalia that is mechani-
cally cooperative; however, we found little evidence of
coevolution between male and female structures. By
choosing to work on a group with obvious female geni-
talic diversity, our study demonstrates the importance
of investigating the male–female copulatory interaction,
and the necessity of including female morphology even
in taxa with cryptic diversity.
Materials and methods
Life history
Phyllophaga beetles are mostly nocturnal and are well
known for their large size (1.5–2.0 cm) and characteristic
brownish colour (Horn, 1887). The length of the life cycle
varies from 1 to 3 years and tends to correlate with lati-
tude, with longer 3-year life cycles occurring at higher lat-
itudes and shorter 1- to 2-year life cycles at lower
latitudes (Reinhard, 1941). The larvae are root pests of
many crop and grass species, and upon emergence in the
spring adults feed on tree foliage throughout their 2- to 3-
week mating season. As is true of many insects, male gen-
italia of Phyllophaga are species-specific and diverse; how-
ever, in contrast to many other insect species, Phyllophaga
female genitalia are also species-specific. Reproduction in
Phyllophaga is initiated by the female when she everts her
pheromone gland to attract males. When males approach,
the ensuing interaction is clumsy until the male positions
himself on top of the female and the male genitalia are
inserted through the open female pygidium. Once copu-
lation has been initiated, the male releases his front and
mid-legs from the female’s dorsum and hangs by his geni-
talia with minimal contact by his hind legs (Fig. 3a).
Copulation in Phyllophaga species lasts 1.0–2.5 h
(Polihronakis, 2008).
Specimen collection
This study focused on Phyllophaga species from the sub-
genus Phyllophaga that occur in the eastern United
States. Specimens for DNA work were collected in the
field using mercury vapour and ultraviolet lights and
stored on ice in 95% ethanol until storage at 20 °Cin
the laboratory. Taxonomic identifications were per-
formed using keys in Luginbill & Painter (1953) and
Woodruff & Beck (1989).
Genitalia structure and function
To determine how male and female genitalia interact
during copulation, we located mating pairs on low-
hanging tree branches during peak emergence and
froze them with ethyl chloride spray (Gebauer Com-
pany, Cleveland, OH, USA) for approximately 30 s.
Once the beetles were frozen through, indicated by
frost build-up on their exoskeleton, we placed them
in 95% ethanol while still frozen (following Eberhard,
1993). We dissected and imaged mating pairs by
slowly removing the exoskeleton and internal struc-
tures under a Leica MZ16 stereomicroscope fitted
with a camera to capture and document the interac-
tions between male and female structures.
We used SEM to visualize the internal structure of
the female reproductive tract, and to locate the sperm
emission duct on the male aedeagus. We dissected
female and male reproductive structures and cleared
them in 5% KOH followed by a series of 15-min etha-
nol dehydration steps. To prevent the membranous tis-
sues from wrinkling during drying, we conducted
additional dehydration steps consisting of three 15-min
hexamethyldisilazane (HMDS) changes.
We defined categorical character states for compara-
tive analyses of genitalic character evolution with the
intent of capturing the broad range of shapes observed
in the parameres of the male genitalia and pubic pro-
cess of the female genitalia. The five character states
(Fig. 1) for male genitalia were as follows: symmetric
parameres fused into a tubelike structure (0), symmet-
ric parameres not fused and/or not forming a tube (1),
parameres asymmetric in shape but more or less sym-
metric in size (2), parameres asymmetric in size but
more or less symmetric in shape (3), and parameres
asymmetric in size and shape (4). The form of the
female genitalia was categorized into four possible states
(Fig. 2): flat, platelike pubic process with no extension
(0), straight or linear pubic process extension (1), pubic
process extension linear or laterally expanded with
bifurcation at tip (2), and pubic process extension not
linear, bifurcating from or near base (3). In most cases,
we scored male and female character states from col-
lected specimens, but we also relied on published illus-
trations and images if both sexes were not collected.
Phylogenetic analysis
Genomic DNA was extracted from thorax or leg muscle
using DNeasy blood and tissue kits (Qiagen, Valencia,
CA, USA) or Nucleospin
Tissue Kits (Clontech, Moun-
tain View, CA, USA). When possible, we included two
specimens per species from two different collecting
localities in the phylogenetic analysis. We collected
sequence data from four loci to reconstruct phyloge-
netic relationships [mitochondrial: cytochrome oxidase
subunit I (COI); nuclear: carbamoyl-phosphate synthetase
ª2016 EUROPEAN SOCIETY FOR EVOLUTIONARY BIOLOGY. J. EVOL. BIOL. doi: 10.1111/jeb.12955
JOURNAL OF EVOLUTIONARY BIOLOGY ª2016 EUROPEAN SOCIETY FOR EVOLUTIONARY BIOLOGY
Evolution of male and female genitalia 3
2 (CAD), internal transcribed spacer region 2 (ITS2)
and an anonymous marker (AM)] (Table S1). Amplified
fragments were cycle sequenced using BigDye Termina-
tor 1.1 cycle sequencing kits and run on an ABI 3100
(Applied Biosystems, Foster City, CA, USA), or sent to
Genewiz, Inc. (San Diego, CA, USA) for sequencing.
Sequence chromatograms were edited in SEQUENCHER
v4.8 (Gene Codes Corp., Ann Arbor, MI, USA) and
aligned by eye in SEALv2.0 (Rambaut, 2007). Aligned
COI sequences were translated to amino acid sequences
to check for stop codons. We coded polymorphic sites
in the nuclear gene sequences using IUPAC ambiguity
codes. All nuclear markers were easily aligned by eye
except for the ITS2 regions amplified from the out-
groups Phyllophaga crinita (Burmeister) and Phyllophaga
tristis (Fabricius); unalignable regions of sequence for
those taxa were included but coded as missing data.
We analysed data from each locus individually in
MrBayes v3.1.2 (Huelsenbeck & Ronquist, 2001; Ron-
quist & Huelsenbeck, 2003) to identify any major topo-
logical differences between the loci and to check for
spurious amplification of gene paralogs. Models of
molecular evolution that best fit each marker were
determined using the Akaike Information Criterion cal-
culated in JMODELTEST v2.1.1 (Guindon & Gascuel, 2003;
Darriba et al., 2012) resulting in the GTR +I+G mod-
els for the COI, CAD and the anonymous marker, and
GTR +G for ITS2.
We generated a posterior distribution of species trees
in BEAST v1.8.2 (Drummond et al., 2012) using the
(a) (b) (c)
(d)
Fig. 3 Phyllophaga hirticula mating complex; a) copulating beetles with female grasping a branch and male hanging from female, b)
external left lateral: entire left paramere of male genitalia inserted into female pygidium c) external right lateral: right paramere not fully
inserted with female terminal sclerite positioned in valley formed by right paramere (inset: right lateral view of male genitalia), d) internal,
female dorsal perspective: bifurcating tips of female pubic process situated between and underneath male parameres. Arrow pointing at
female pubic process inserted between and locking underneath male parameres.
ª2016 EUROPEAN SOCIETY FOR EVOLUTIONARY BIOLOGY. J. EVOL. BIOL. doi: 10.1111/jeb.12955
JOURNAL OF EVOLUTIONARY BIOLOGY ª2016 EUROPEAN SOCIETY FOR EVOLUTIONARY BIOLOGY
4M. P. RICHMOND ET AL.
*BEAST option (Heled & Drummond, 2010). All genes
were unlinked and substitutions were modelled as
described for the gene tree analyses. An uncorrelated
exponential relaxed clock was used with a birth-death
process species tree prior and default priors for model
parameters and statistics. The Markov chain was run
for 60 million generations, and the mean and effective
sample size of parameters was checked using TRACER
v1.6.0 (Rambaut et al., 2014).
Genitalia evolution
Ancestral state reconstruction of male and female geni-
talic characters was performed in the program BAYESTRAITS
v2.0 (Pagel et al., 2004) using Markov chain Monte Carlo
(MCMC) methods. Due to the complexity of the model
and number of potential transitions, we used reverse-
jump MCMC with an exponential hyperprior (mean 0.0,
variance 10.0) for the rate parameters, and ran the analy-
sis for 30 million generations. We evaluated nodes with
uncertain character state reconstructions by fixing the
character state (using the ‘Fossil’ command) and compar-
ing the marginal log likelihoods from both the harmonic
mean and stepping stones sampler. Genitalia evolution
was reconstructed using 1000 species trees from the pos-
terior distribution generated from *BEAST. We used the
program BAYESTREES v1.3 (Meade & Pagel, 2012) to reduce
the posterior distribution from 60 000 trees (MCMC
chain run for 60 million generations recording trees every
1000 steps) to 1000 trees by removing the first 55 000
trees, followed by random removal of 4000 of the remain-
ing 5000 trees. Phyllophaga crinita and P. tristis were used
to root the species trees based on previous analyses of COI
sequence data (Polihronakis, 2008).
We tested for correlated evolution between the male
and female structures by comparing independent and
dependent models of character evolution in BAYESTRAITS
using Bayes factors from the marginal log likelihoods
from the harmonic mean and stepping stones sampler.
Bayes factors are similar to likelihood ratio tests and are
commonly used in Bayesian statistics to evaluate non-
nested competing models. Because the correlated evo-
lution analysis requires binary traits, the male genitalia
were scored as either symmetric or asymmetric, and
the female pubic process was scored as not bifurcating
(including species with no pubic process extension) or
bifurcating (either from base or at tip). Tests of corre-
lated evolution were run as described above for the
ancestral character state reconstructions.
Results
Specimen collection
Over 1800 specimens of Phyllophaga comprising 59 spe-
cies were collected for this study. Taxon sampling for
this study included nearly half of the Phyllophaga s. str.
endemic to the United States. Whereas denser taxon
sampling is always desired, current sampling is justified
based on inclusion of species from nearly all of the
major species groups within the subgenus (B€
oving,
1942). Most specimens were easily identified to species
using characters of the male and female genitalia. How-
ever, there was a series of approximately ten male spec-
imens that closely resembled Phyllophaga fraterna Harris
but had slight variations of paramere structure that pre-
cluded a positive identification as P. fraterna. Those
individuals are undergoing further study but are
referred to here as P. cf. fraterna.
Genitalia structure and function
A total of five pairs of mating Phyllophaga from three
species were frozen and dissected to investigate how
the male and female genitalia interact during copula-
tion. Of the three species examined, two had male
parameres that were asymmetric in size and shape
(Phyllophaga bipartita Horn and Phyllophaga hirticula
Knoch) and one had parameres asymmetric in shape
only (Phyllophaga fusca Froelich). The pubic process
form was different in each of the three species; P. bipar-
tita bifurcates from the base, P. hirticula bifurcates at
the tip and P. fusca is elongate with no bifurcation. In
both species having parameres asymmetric in size and
shape, the males inserted the entire left paramere (the
larger) into the female where it hooked on an internal
shelf of the terminal sternite (Figs 3b and 4a). The
smaller right paramere, shaped like a valley in both
species, remained partially outside the female and fit
the terminal sternite of the female abdomen resulting
in a one-sided mating position (Figs 3c and 4b). Further
dissection of these mating pairs revealed that the bifur-
cating female pubic process of P. hirticula and P. bipar-
tita was inserted between and positioned underneath
the parameres (Figs 3d and 4d). In both cases, the lock-
ing mechanism appears to become stronger as more
tension is applied to it. In other words, as the male
leans back and releases his hold on the female with his
legs, the increased tension on the genitalia stabilizes
the locking mechanism. Although the mechanism was
similar for both species, the species-specific morphology
of the pubic process bifurcation resulted in different
types of ‘fit’ between the male and female structures.
In P. bipartita, the bifurcations of the pubic process
were more pronounced resulting in increased contact
with the male parameres (Fig. 4d), whereas in P. hirtic-
ula the bifurcation is smaller (Fig. 3d). Dissection of the
mating pairs also allowed us to visualize the aedeagus
of the male, the actual sperm transmission organ,
which was positioned inside of the bursa copulatrix of
the female (Fig. S1).
In the P. fusca mating pair, a species having parameres
asymmetric in shape but not size, both parameres were
fully inserted into the female and appeared to be
ª2016 EUROPEAN SOCIETY FOR EVOLUTIONARY BIOLOGY. J. EVOL. BIOL. doi: 10.1111/jeb.12955
JOURNAL OF EVOLUTIONARY BIOLOGY ª2016 EUROPEAN SOCIETY FOR EVOLUTIONARY BIOLOGY
Evolution of male and female genitalia 5
squeezing the linear pubic process of the female (Fig. 5).
Whereas this interaction is similar to that seen in P. bi-
partita and P. hirticula with regard to direct contact
between the parameres and pubic process, P. fusca male
and female genitalia did not engage in a locking mecha-
nism as seen in the other two species.
The SEM analysis of internal female reproductive
anatomy revealed spines in the genital tract and bursa
copulatrix itself that were pointed in the same direction
the aedeagus would be inserted (Fig. S2). The SEM of
the male aedeagus revealed the location of the opening
from which sperm would presumably be emitted near
the base of the aedeagus rather than at the tip
(Fig. S3). This location corresponds to the position of
the opening to the female spermatheca closer to the
entrance of the bursa copulatrix (Fig. S1). Thus,
whereas it appears that the inflated head of the aedea-
gus (which is equipped with varying degrees of arma-
ture depending on the species) takes up the majority of
the space of the bursa copulatrix, the sperm are ejected
closer to the base of this structure.
Phylogenetic analysis
Genomic DNA was extracted from a total of 109 indi-
viduals from 59 species (Table S2). Individual gene tree
analyses were largely congruent but did reveal three
ingroup ITS2 sequences that were allied with the
(a) (b)
(c) (d)
Fig. 4 Phyllophaga bipartita mating complex; a) external –left lateral: entire left paramere of male genitalia inserted into female pygidium,
b) external –right lateral: right paramere not fully inserted with female terminal sclerite positioned in paramere groove; inset –right
lateral view of male genitalia when not mating, c) internal, female ventral perspective: aedeagus from male genitalia can be seen going
into female genitalia ventral to pubic process, d) internal, female dorsal perspective: bifurcating pubic process situated in between and
underneath male parameres.
ª2016 EUROPEAN SOCIETY FOR EVOLUTIONARY BIOLOGY. J. EVOL. BIOL. doi: 10.1111/jeb.12955
JOURNAL OF EVOLUTIONARY BIOLOGY ª2016 EUROPEAN SOCIETY FOR EVOLUTIONARY BIOLOGY
6M. P. RICHMOND ET AL.
outgroup, indicative of paralogous loci. Those sequences
were removed from the data set.
The consensus phylogeny from the species tree anal-
ysis resolved most major groupings within the Phyl-
lophaga s. str. (Fig. 6). Well-supported nodes
distinguished several major clades, some of which cor-
responded to groups suggested by B€
oving (1942) based
on larval and genitalia characters. There was strong
support for close relationship of the two outgroup spe-
cies, P. crinita and P. tristis, which have long been con-
sidered close allies and members of the newly
resurrected genus, Trichesthes Erichson (Coco Abia
2002). There was also strong support for the
Phyllophaga aemula group, and its position as sister to
the remaining Phyllophaga s. str. In the current analysis,
the subgenus Phyllophaga is monophyletic with high
node support; however, it should be noted that COI
gene trees with more inclusive taxon sampling provided
evidence for paraphyly of the subgenus due to inclu-
sion of species from the subgenera Tostegoptera and Phy-
talus (Polihronakis, 2008). Paraphyly of many
Phyllophaga subgenera has been previously discussed
(Horn, 1887; Luginbill & Painter, 1953; Woodruff &
Beck, 1989) and points to an ongoing need for a revi-
sion of the entire genus.
Genitalia evolution
The ancestral state of male parameres and the form of
the female pubic process were estimated for the root
and twelve nodes in the phylogeny using 1000 species
trees from the posterior distribution generated in
*BEAST (Fig. 6). Most phylogenetic nodes included in
the reconstruction analyses had a posterior probability
>0.90 and were thus well sampled during the MCMC
analysis. Node 7, the most recent common ancestor of
the P. fusca and P. fraterna species groups, had a slightly
lower posterior probability (0.84), which was factored
into the analysis and the resulting posterior densities of
ancestral character state reconstruction (Pagel et al.,
2004). Because some of the posterior densities from the
ancestral state reconstruction were bimodal, we
reported the proportion of iterations with a probability
>0.50 (Tables 1 and 2). We interpreted node recon-
struction as uncertain when more than 25% of itera-
tions had a probability less than 0.50. Nodes with
uncertain ancestral character states were further anal-
ysed using the ‘Fossil’ command (Table S3). In all cases,
the marginal log likelihoods from the harmonic mean
and stepping stones sampler agreed on which state was
most likely.
Female pubic process
Evolution of the pubic process transitioned frequently,
with closely related species often having very different
pubic process forms (Fig. 6; Table 1). The ancestral state
at the root of the tree was uncertain, with minimal
support for a platelike pubic process. Whereas the
ancestral nodes for the outgroup (Node 1) and the
P. aemula group (Node 3) revealed strong support for a
platelike pubic process, support for a platelike pubic
process was less certain at Node 2. There was moderate
support for a transition to a basally bifurcating pubic
process at Node 4 that persisted through Nodes 6, 7, 9
and 10; however, descendant lineages of these ances-
tors exhibited the full spectrum of genital types. The
common ancestor of the Phyllophaga congrua clade
(Node 5) was linear, although a variety of pubic process
forms were present at the tips of this clade. The ances-
tral reconstruction for the Phyllophaga balia group
(Node 12), sister to the Phyllophaga fraterna group, was
also uncertain, with some support for a pubic process
that bifurcates at the tip. In sum, our results suggest
the female pubic process transitioned from platelike to
more elongate with a bifurcation from the base. There
were three independent transitions from a pubic pro-
cess that bifurcated from the base to a pubic process
that bifurcated at the tip (Nodes 8, 11 and 12).
Male parameres
The direction of evolution of male parameres was
strongly supported as symmetric to asymmetric, with
one transition to asymmetry in paramere shape that led
to at least two transitions to parameres asymmetric in
size (but not shape), and one transition to parameres
asymmetric in size and shape (Fig. 6; Table 2). There
was some support for fused, symmetric parameres at
the root and strong support for fused, tubelike para-
meres in the ancestor of the outgroup (Node 1). The
remaining nodes at the base of the tree (Nodes 2–6)
supported an ancestor with unfused, symmetric para-
meres. A transition to asymmetric parameres occurred
at Node 7, with support for an ancestor having
Fig. 5 Phyllophaga fusca mating complex: internal, female dorsal
perspective. Both parameres inside the female squeezing the
elongate female pubic process (at tip of arrow).
ª2016 EUROPEAN SOCIETY FOR EVOLUTIONARY BIOLOGY. J. EVOL. BIOL. doi: 10.1111/jeb.12955
JOURNAL OF EVOLUTIONARY BIOLOGY ª2016 EUROPEAN SOCIETY FOR EVOLUTIONARY BIOLOGY
Evolution of male and female genitalia 7
parameres asymmetric in shape but not size. This state
persists in the ancestor of the P. fusca group (Node 8),
but transitions to parameres that are asymmetric in
both shape and size at Node 9. There is moderate sup-
port for a transition back to parameres that are asym-
metric in shape only in the ancestor to the P. balia
group, sister to the P. fraterna group (Node 12), but the
remaining well-supported nodes of the P. fraterna group
(Nodes 10 and 11) strongly support ancestors with
parameres asymmetric in size and shape. Even though
there were several species that had parameres asym-
metric in size but not shape (e.g. P. implicita Horn,
Phyllophaga praetermissa Horn and Phyllophaga schaefferi
Saylor), there were no ancestors reconstructed with
that character state.
The moderate-to-strong support for the reconstruc-
tions presented here suggests the male genitalia first
transitioned from a symmetric, fused, tubelike form to
a symmetric form with unfused parameres. Subse-
quently, at Node 4, the female pubic process transi-
tioned from a minimal, platelike form to one that
bifurcates from the base. There were then subsequent
transitions of the female genitalia as the male genitalia
evolved different asymmetric forms; however, there
was very weak evidence for correlated evolution of
male and female genitalia. The highest Bayes factor was
3.2 (positive evidence for coevolution according to the
BayesTraits manual) using the marginal likelihoods
from the harmonic mean. The highest Bayes factor
from the stepping stones sampler, considered a more
robust method of estimating the marginal likelihood,
suggested weaker support with a value of 0.5. Weak
support for correlated evolution between male and
female genitalia was not surprising given the seemingly
disorganized distribution of female pubic process forms
throughout the tree.
0.01
**
**
*
*
*
**
*
*
*
*
*
11
*
*
10
9
12
*
*
*
*
*
*
*
7
8
**
**
6
*
*
*
**
**
*
4
5
2
3
1crinita
tristis
Female pubic process
No extension
Linear/fused
Linear or laterally expanded with bifurcation at tip
Bifurcating from base
Male parameres
Symmetric - fused into a tube-like shape
Symmetric - not fused
Shape asymmetry
Size asymmetry
Shape and size asymmetry
Female Male
ilicis
cf. fraterna
knochii
subtonsa
lodingi
cf. fraterna
pearliae
cf. fraterna
foxii
delata
profunda
fraterna
hirticula
forsteri
tecta
paternoi
nebulosa
hornii
inversa
luctuosa
prunina
bipartita
calceata
diffinis
marginalis
spreta
implicita
balia
hirsuta
vilifrons
anxia
fusca
drakii
fervida
perlonga
postrema
schaefferi
arkansana
praetermissa
cupuliformis
micans
crassissima
errans
clypeata
latifrons
longitarsa
glaberrima
congrua
hirtiventris
futilis
aemula
crenulata
parvidens
glabricula
submucida
torta
pseudofloridana
aemula
group
fusca
group
balia
group
fraterna
group
marginali
s
group
congrua
group
Fig. 6 Species tree inferred in *BEAST showing character states of each tip species, and ancestral character states indicated for the root and
all reconstructed nodes, regardless of level of certainty. *indicates posterior probability clade support >0.70.
ª2016 EUROPEAN SOCIETY FOR EVOLUTIONARY BIOLOGY. J. EVOL. BIOL. doi: 10.1111/jeb.12955
JOURNAL OF EVOLUTIONARY BIOLOGY ª2016 EUROPEAN SOCIETY FOR EVOLUTIONARY BIOLOGY
8M. P. RICHMOND ET AL.
We observed unusual patterns of morphological simi-
larity and dissimilarity when considering phylogenetic
relatedness. For example, there were several instances
where two unrelated species had very similar pubic
process forms, but the male parameres would be very
different. Specifically, P. luctuosa (Horn) (sister to the
P. fraterna group) and P. crassissima (Blanchard) (in the
P. fusca group) have similar pubic process forms that
both bifurcate from the base; however, the male geni-
talia of these two species are very different. In
P. luctuosa, the male genitalia are asymmetric in size
and shape, whereas in P. crassissima they are unfused,
but symmetric. These examples illustrate the quick and
sporadic nature of female genitalic change and help
explain why there was minimal evidence for
coevolution.
Discussion
Our results revealed a unique interlocking mechanism
between male and female structures in some Phyl-
lophaga species, where the female genitalia bifurcate
and hook underneath the male parameres. The male
and female genital morphology in Phyllophaga thus
appears to function to enhance the mechanical fit
between the sexes as the male hangs from the female.
Comparative phylogenetic analyses show contrasting
patterns of male and female genital evolution, with lit-
tle to no support for coevolution. In males, we found
support for symmetric genitalia as the ancestral state,
with more closely related species tending to share simi-
lar character states. In contrast, female genitalia have
evolved more erratically, making it difficult to infer
ancestral states with confidence.
The copulatory interaction
The interlocking mechanism of the male parameres and
bifurcating female pubic process in two of the three
species of Phyllophaga examined here is an important
finding for several reasons. First, there is no evidence
that the interacting structures are directly involved in
sperm transfer or receipt, suggesting their function is
primarily mechanical. Second, these structures exhibit
high intra- and interspecific diversity in Phyllophaga
(Smith, 1888; Polihronakis, 2006), which leads to
ample variation on which selection can act to increase
the interlocking capability of mating individuals. Third,
there are two types of bifurcating female genitalia that
likely increase locking capability and both are spread
throughout the tree. These findings raise questions
about how and why these structures have evolved, and
why they are species-specific.
Because the male parameres and female pubic pro-
cess are not involved in sperm transfer or receipt, there
is a broader range of functional roles to consider in
explaining the species specificity of these traits. Previ-
ous studies have shown that many genital structures
are involved with grasping or clasping due to sexual
conflict (Arnqvist & Rowe, 2005), female stimulation
(Eberhard, 1985) and mate recognition between con-
specifics (McPeek et al., 2008; Richmond, 2014). The
genitalic interaction in Phyllophaga is not concordant
with antagonistic coevolution for several reasons. First,
we do not find evidence of coevolution between male
and female genital structures, which would be expected
if an arms race due to sexual conflict were taking place.
Table 1 Ancestral character states for female genitalia reported as
the proportion of iterations with probability >0.50; shaded boxes
indicate nodes with high uncertainty (proportion of iterations with
<0.50 probability is >25%); *indicates support using the ‘Fossil’
command based on comparison of the marginal log likelihoods.
0 1 2 3
Proportion of
iterations with
<0.50 probability
Root 0.19*0.05 0.02 0.04 70%
Node 1 0.99 0 0 0 1%
Node 2 0.18*0.20 0.10 0.09 43%
Node 3 0.99 0 0 0 1%
Node 4 0 0.03 0.09 0.48*40%
Node 5 0 0.26*0.10 0.25 39%
Node 6 0 0.03 0.10 0.75 12%
Node 7 0 0 0.02 0.73 25%
Node 8 0 0 0.29*0.57 14%
Node 9 0 0 0.01 0.92 7%
Node 10 0 0 0 0.86 14%
Node 11 0 0 0.90 0 10%
Node 12 0 0 0.12*0.43 45%
Table 2 Ancestral character states for male genitalia reported as
the proportion of iterations with probability >0.50; shaded boxes
indicate nodes with high uncertainty (proportion of iterations with
<0.50 probability is >25%); *Indicates support using the ‘Fossil’
command based on comparison of the marginal log likelihoods.
0 1 2 3 4
Proportion of
iterations with
<0.50 probability
Root 0.26*0.21 0 0 0 53%
Node 1 0.99 0 0 0 0 1%
Node 2 0.03 0.91 0 0 0 6%
Node 3 0.08 0.87 0 0 0 5%
Node 4 0 0.92 0.04 0.01 0 3%
Node 5 0 0.94 0 0 0 6%
Node 6 0 0.81 0.10 0.01 0.01 7%
Node 7 0 0.15 0.34*0.02 0.09 40%
Node 8 0 0.01 0.94 0.03 0 2%
Node 9 0 0.01 0.09 0.03 0.78 9%
Node 10 0 0 0.03 0.06 0.84 7%
Node 11 0 0 0 0 0.89 11%
Node 12 0 0 0.26*0.01 0.39 44%
ª2016 EUROPEAN SOCIETY FOR EVOLUTIONARY BIOLOGY. J. EVOL. BIOL. doi: 10.1111/jeb.12955
JOURNAL OF EVOLUTIONARY BIOLOGY ª2016 EUROPEAN SOCIETY FOR EVOLUTIONARY BIOLOGY
Evolution of male and female genitalia 9
Second, there is no evidence that the male structures
have offensive capabilities and can grasp the female
against her will, or that the female structures are defen-
sive and have evolved to fend off the male. In fact,
either sex could theoretically disengage by twisting
their genitalia 90°. And third, the microstructures of
the female genital tract appear cooperative rather than
antagonistic.
Female stimulation seems unlikely given that both
male and female structures are highly sclerotized; how-
ever, it is possible that membranous regions connected
to the sclerotized regions are involved in a sensory
response. The remaining explanation is that the interac-
tion is cooperative and results from selection for mate
recognition and/or a robust mechanical fit. The SEM
analysis of fine-scale structuring provides further sup-
port for this hypothesis by revealing spines within the
female reproductive tract that point in the same direc-
tion the aedeagus is inserted, suggesting additional
mechanisms for maintaining stability during copulation.
One aspect of the mating interaction that may influ-
ence the evolution of a mechanical fit of genitalia in
Phyllophaga is that in most species, males hang from the
female during mating (forming a 90°–180°angle with
the female). Interestingly, this happens in P. fusca,
where the male grasps the pubic process but it does not
lock underneath the parameres, and also in P. hirticula
and P. bipartita, where the pubic process does lock
underneath the parameres. In a previous study of copu-
latory behaviour in Phyllophaga, Eberhard (1993)
described the genital interaction in seven species that
do not have a pubic process extension (all having state
0). Whereas Eberhard (1993) found some examples of
the male parameres hooking into the female pygidium
(e.g. P. obsoleta Blanchard), there was no indication of
an interlocking mechanism between the male and
female genitalia. On the other hand, Eberhard (1993)
did observe interactions where structures of the aedea-
gus appeared to enhance the mechanical fit, or stability,
of the copulatory position. Thus, it would be worth-
while in future studies to investigate the evolutionary
transitions of holdfast structures associated with the
parameres vs. aedeagus with respect to female pubic
process morphology.
If selection is underlying the changes described here,
we would expect there to be an advantage for a tight,
mechanical fit between the male and female genitalia
during mating. A similar locking interaction has been
studied in detail in Enallagma Charpentier damselflies
where species-specific male and female structures were
shown to function in mate recognition, even among
historically allopatric populations (Robertson & Pater-
son, 1982; McPeek et al., 2008). The evolution of
diverse genitalic traits in Enallagma damselflies, which
are otherwise phenotypically identical, was hypothe-
sized to result from fluctuating population margins dur-
ing Pleistocene glacial events (McPeek et al., 2008).
Thus, the evolution of structures involved in mechani-
cal coupling does not necessarily require that selection
act on traits to prevent interspecific matings; however,
mate recognition could nonetheless result as a by-pro-
duct of selection for goodness-of-fit during copulation.
For example, if the genitalia of both sexes respond dif-
ferently to selective pressures on the mating interaction
in allopatric populations, the resulting genital diver-
gence could then play an important role in mate recog-
nition during secondary contact.
Genitalic evolution
Ancestral reconstruction of male genitalic characters
suggests that the direction of evolution was symmetric
to asymmetric, with moderate support for an ancestral
form characterized by fused, tubelike parameres and a
subsequent transition to unfused parameres (Fig. 6).
Once the parameres were unfused, the transition to
asymmetry began with the evolution of asymmetrically
shaped parameres, with further transitions to parameres
asymmetric in both size and shape.
In contrast to males, female genitalic forms vary
throughout the tree and do not appear to cluster phyloge-
netically. For example, female genitalia in all groups but
the P. aemula group include at least two pubic process
character states, and most have three. There are also
examples of species pairs, such as P. anxia (LeConte) and
P. fusca, where the male genitalia are similar but the
female pubic process is different. These differences indi-
cate that male and female structures have the potential to
evolve independently. A similar pattern was seen in
intraspecific analyses of male and female genitalia in
P. hirticula that exhibited higher diversity of the female
pubic process relative to the male parameres (Polihron-
akis, 2006), with little to no correlation between morpho-
logical variation and geographic or genetic distance
(Polihronakis, 2009). Thus, interspecific patterns of diver-
sity of male and female genitalia in Phyllophaga reflect
patterns of variation observed within species.
To date, few studies have reconstructed ancestral
character states of female genitalia in arthropods using
aphylogenetic framework. In an investigation of female
reproductive morphology in sepsid flies, Puniamoorthy
et al. (2010) found the female ventral receptacle, a
structure hypothesized to function in sperm storage and
egg fertilization, was very diverse with multiple origins
of similar ventral receptacle types. McPeek et al. (2009)
found evidence for punctuated change of female repro-
ductive structures in Enallagma damselflies, meaning
that morphological change was associated with specia-
tion events. Whereas these studies are admittedly few,
the results suggest that the evolution of female genitalia
is not stationary, as is often assumed. Further study of
female genital evolution in an evolutionary context
thus has the potential to provide much needed insight
into analyses of male genital evolution.
ª2016 EUROPEAN SOCIETY FOR EVOLUTIONARY BIOLOGY. J. EVOL. BIOL. doi: 10.1111/jeb.12955
JOURNAL OF EVOLUTIONARY BIOLOGY ª2016 EUROPEAN SOCIETY FOR EVOLUTIONARY BIOLOGY
10 M. P. RICHMOND ET AL.
Despite the direct interaction between male para-
meres and female pubic process, we did not find con-
vincing evidence of coevolution between these
structures in the Phyllophaga subgenus. This result is
contrary to other studies across a variety of arthropod
groups that report positive evidence for coevolution of
male and female structures when tested in a phyloge-
netic framework (Ramos et al., 2005; Kuntner et al.,
2009; McPeek et al., 2009; Tanabe & Sota, 2010, 2014).
One reason for this could be that our analysis of Phyl-
lophaga covered a broad phylogenetic scale that necessi-
tated the use of categorical character coding, which
may not adequately capture continuous, three-dimen-
sional trait variation. Nevertheless, based on observed
patterns of variation of the female genitalia relative to
males, a lack of evidence for coevolution was not sur-
prising. An analysis of male and female genital evolu-
tion within a smaller species group of Phyllophaga,
using continuous character information that better cap-
tures variation, would help clarify whether these struc-
tures are coevolving.
Genitalic asymmetry
Genital asymmetry in insects is relatively widespread,
with several examples occurring in each of the major
orders (Huber et al., 2007). In Phyllophaga, asymmetric
paramere shape appears to have evolved once in the
subgenus Phyllophaga, which then led to different types
of asymmetry. In the copulatory interaction of the
three species investigated here, the one with size-sym-
metric parameres (but not necessarily shape) inserted
both parameres into the female during copulation. In
the two species where males have asymmetric-sized
parameres, the larger left paramere was fully inserted
into the female whereas the smaller right paramere
remained partially outside, resulting in a tilted mating
position. A similar situation has been found in Droso-
phila pachea Patterson and Wheeler, a species with
asymmetric external lobes. In this species, the mating
position is one-sided with the male positioned at an
angle relative to the vertical axis of the female, which
does not happen in other Drosophila species with sym-
metric genitalia (Lang & Orgogozo, 2012). In such
cases, it can be difficult to elucidate which came first,
the one-sided mating position or the asymmetric struc-
tures. Huber et al. (2007) hypothesize that asymmetric
genitalia evolve after transitions to one-sided mating
positions. In Phyllophaga, asymmetric parameres
evolved after the transition from fused to unfused para-
meres, allowing opposite sides to assume separate func-
tions (Huber, 1999), which could have coincided with
an advantage for one-sided mating positions.
Reversals from paramere shape asymmetry to sym-
metry were inferred in the P. fusca group but nowhere
else in the tree. For example, in the P. fraterna group,
once genitalia transitioned to parameres asymmetric in
shape and size, there were no reversals. The same has
been shown in other groups of insects with asymmetric
genitalia; once asymmetry has evolved, there are few
documented cases of reversals to symmetry (Huber
et al., 2007). Whereas this finding has led some to spec-
ulate that asymmetry is adaptive and might lead to
higher diversification rates, a recent study of the scarab
beetle genus Schizonycha Dejean by Eberle et al. (2015)
did not find evidence of higher diversification rates in
lineages with asymmetric genitalia.
Asymmetry of female genitalia in Phyllophaga, and
arthropods in general, is relatively uncommon (Huber
et al., 2007). However, one instance of asymmetric
female genitalia was discovered in a newly described
species of Phyllophaga,P. nebulosa Polihronakis, which
also has asymmetric male genitalia (Polihronakis,
2007). Asymmetry of female genitalia has also been
documented in a genus of theridiid spiders; the males
have symmetric genitalia but may only use one palp
during copulation (Agnarsson, 2006).
Conclusions
Our data suggest that the function and evolution of
genitalia in the subgenus Phyllophaga is consistent with
a mechanical mechanism that stabilizes the engagement
of male and female genitalia during copulation. In sys-
tems with interlocking male–female interactions, evolu-
tion of mechanical fit may not necessarily be driven by
selection against interspecific mating, but could occur
in allopatric populations and subsequently enhance
mate recognition. Possible explanations for the advan-
tage of a highly stable male–female interaction include
more efficient sperm transfer, preventing dislodging by
predators, and/or as an attachment mechanism. Inter-
estingly, our data suggest that the evolution of species-
specific locking mechanisms may not always lead to
concerted male and female genitalic evolution. By
focusing on both sexes and the ways in which their
genitalia interact during mating, the evolution of
diverse genitalia can be more thoroughly explored.
Whereas further work is needed to identify a general
explanation for the diversity of insect genitalia, incor-
porating information on female genitalic morphology
has proven to be useful by providing insight into the
evolution of previously unknown functional interac-
tions.
Acknowledgments
I would like to thank J.Q. Richmond for providing the
image for Fig. 3a and J.Q. Richmond, R. Singh and B.
Eberhard for comments and discussion of earlier ver-
sions of this manuscript. I would also like to thank T.
Cruickshank, K.B.R. Hill, J. King, P.K. Lago, C. Legault,
D.C. Marshall, S. Noh and J.Q. Richmond for assistance
in collecting specimens in the field and R. H. McPeak
ª2016 EUROPEAN SOCIETY FOR EVOLUTIONARY BIOLOGY. J. EVOL. BIOL. doi: 10.1111/jeb.12955
JOURNAL OF EVOLUTIONARY BIOLOGY ª2016 EUROPEAN SOCIETY FOR EVOLUTIONARY BIOLOGY
Evolution of male and female genitalia 11
for providing specimens of P. errans from Oregon. I
would like to thank the State Park systems of Florida
(Donna Watkins), North Carolina (Ed Corey) and Geor-
gia (Cindy Reittinger) for their assistance with collecting
permits. Support for this research was provided to M.P.R.
by the United States National Science Foundation (DDIG
DEB-0608348), the American Museum of Natural His-
tory Theodore Roosevelt Memorial Fund, the Society of
Systematic Biologists and the University of Connecticut
Natural History Museum. In addition, laboratory space
and materials were provided by C. Simon supported by
the United States National Science Foundation (DEB 04-
22386; PEET DEB 05-29679; REU DEB-0619012). Any
opinions, findings and conclusions or recommendations
expressed in this material are those of the authors and do
not necessarily reflect the views of the NSF.
References
Agnarsson, I. 2006. Asymmetric female genitalia and other
remarkable morphology in a new genus of cobweb spiders
(Theridiidae, Araneae) from Madagascar. Biol. J. Linn. Soc.
87: 211–232.
Arnqvist, G. & Rowe, L. 2005. Sexual Conflict. Princeton Univer-
sity Press, Princeton.
B€
oving, A.G. 1942. A classification of larvae and adults of the
genus Phyllophaga (Coleoptera: Scarabaeidae). Mem. Ent. Soc.
Wash. 2:1–95.
Darriba, D., Taboada, G.L., Doallo, R. & Posada, D. 2012. jMOD-
ELTEST 2: more models, new heuristics and parallel comput-
ing. Nat. Methods 9: 772.
Drummond, A.J., Suchard, M.A., Xie, D. & Rambaut, A. 2012.
Bayesian phylogenetics with BEAUTi and the BEAST 1.7.
Mol. Biol. Evol. 29: 1969–1973.
Eberhard, W.G. 1985. Sexual Selection and Animal Genitalia. Har-
vard University Press, Cambridge.
Eberhard, W.G. 1993. Copulatory courtship and morphology
of genitalic coupling in seven Phyllophaga species (Coleop-
tera: Melolonthidae). J. Nat. Hist. 27: 683–7171.
Eberhard, W.G. 1997. Sexual selection by cryptic female choice
in insects and arachnids. In: The Evolution of Mating Systems
in Insects and Arachnids (J.C. Choe& B.J. Crespi, eds), pp. 32–
57. Cambridge University Press, New York.
Eberhard, W.G. 2010. Evolution of genitalia: theories, evi-
dence, and new directions. Genetica 138:5–18.
Eberhard, W.G. 2011. Experiments with genitalia: a commen-
tary. Trends Ecol. Evol. 26:17–21.
Eberle, J., Walbaum, W., Warnock, R.C.M., Fabrizi, S. & Ahrens,
D. 2015. Asymmetry in genitalia does not increase the rate of
their evolution. Mol. Phylogenet. Evol. 93: 180–187.
Glasgow, R.D. 1916. Phyllophaga Harris (Lachnosterna Hope): a
revision of synonymy, and one new name. Bull. Illinois State
Lab. Nat. Hist. 11: 365–379.
Guindon, S. & Gascuel, O. 2003. A simple, fast, and accurate
algorithm to estimate large phylogenies by maximum likeli-
hood. Syst. Biol. 52: 696–704.
Heifetz, Y. & Wolfner, M.F. 2004. Mating, seminal fluid com-
ponents, and sperm cause changes in vesicle release in the
Drosophila female reproductive tract. Proc. Natl Acad. Sci. USA
101: 6261–6266.
Heled, J. & Drummond, A.J. 2010. Bayesian inference of spe-
cies trees from multilocus data. Mol. Biol. Evol. 27:
570–580.
Horn, G.H. 1887. Revision of the species of Lachnosterna of
America North of Mexico. Trans. Am. Entomol. Soc. 19: 209–
297.
Hosken, D.J. & Stockley, P. 2004. Sexual selection and genital
evolution. Trends Ecol. Evol. 19:8
7–93.
Huber, B.A. 1999. Evidence for functional segregation in the
directionally asymmetric male genitalia of the spider Metago-
nia mariguitarensis (Gonzlez-Sponga) (Pholcidae: Araneae). J.
Zool. 262: 317–326.
Huber, B.A., Sinclair, B.J. & Schmitt, M. 2007. The evolution
of asymmetric genitalia in spiders and insects. Biol. Rev. 82:
647–698.
Huelsenbeck, J.P. & Ronquist, F. 2001. MrBayes: Bayesian
inference of phylogenetic trees. Bioinformatics 17: 754–755.
Jagadeeshan, S. & Singh, R.S. 2006. A time-sequence func-
tional analysis of mating behaviour and genital coupling in
Drosophila: role of cryptic female choice and male sex-drive
in the evolution of male genitalia. J. Evol. Biol. 19: 1058–
1070.
Kuntner, M., Coddington, J.A. & Schneider, J.M. 2009. Inter-
sexual arms race? Genital coevolution in nephilid spiders
(Araneae, Nephilidae). Evolution 63: 1451–1463.
Lang, M. & Orgogozo, V. 2012. Distinct copulation positions in
Drosophila pachea males with symmetric or asymmetric exter-
nal genitalia. Contrib. Zool. 81:87–94.
Luginbill, P. & Painter, H.R. 1953. May beetles of the United
States and Canada. USDA Tech. Bull. 1060:1–102.
Masly, J.P. 2012. 170 Years of “Lock-and-Key”: genital
morphology and reproductive isolation. J. Evol. Biol. 2012:
1–10.
Mattei, A.L., Riccio, M.L., Avila, F.W. & Wolfner, M.F. 2015.
Integrated 3D view of postmating responses by the Drosophila
melanogaster female reproductive tract, obtained by micro-
computed tomography scanning. Proc. Natl Acad. Sci. USA
112: 8475–8480.
McPeek, M.A., Shen, L., Torrey, J.Z. & Farid, H. 2008. The tempo
and mode of three-dimensional morphological evolution in
male reproductive structures. Am. Nat. 171: E158–E178.
McPeek, M.A., Shen, L. & Farid, H. 2009. The correlated evo-
lution of three-dimensional reproductive structures between
male and female damselflies. Evolution 63:73–83.
Meade, A. & Pagel, M. 2012. BayesTrees 1.3. http://www.evo-
lution.rdg.ac.uk/BayesTrees.html. Accessed 28 Jan 2016.
Pagel, M., Meade, A. & Barker, D. 2004. Bayesian estimation
of ancestral character states on phylogenies. Syst. Biol. 53:
673–684.
Parker, G.A. 1979. Sexual selection and sexual conflict. In: Sex-
ual Selection and Reproductive Competition in Insects (M.S.
Blum& N.A. Blum, eds), pp. 123–166. Academic Press,
London.
Polihronakis, M. 2006. Morphometric analysis of intraspecific
shape variation in male and female genitalia of Phyllophaga
hirticula (Coleoptera: Scarabaeidae: Melolonthinae). Ann.
Entomol. Soc. Am. 99: 144–150.
Polihronakis, M. 2007. New species of Phyllophaga Harris
(Coleoptera: Scarabaeidae) from the North Carolina Cape
Fear River Basin. Coleopts. Bull. 61: 429–433.
Polihronakis, M. 2008. Factors promoting the evolution of sexually
dimorphic traits and their consequences on lineage diversification in
ª2016 EUROPEAN SOCIETY FOR EVOLUTIONARY BIOLOGY. J. EVOL. BIOL. doi: 10.1111/jeb.12955
JOURNAL OF EVOLUTIONARY BIOLOGY ª2016 EUROPEAN SOCIETY FOR EVOLUTIONARY BIOLOGY
12 M. P. RICHMOND ET AL.
Phyllophaga beetles (Coleoptera: Scarabaeidae). Dissertation,
University of Connecticut, Storrs, CT.
Polihronakis, M. 2009. Hierarchical comparative analysis of
genetic and genitalic geographical structure: testing patterns
of male and female genital evolution in the scarab beetle
Phyllophaga hirticula (Coleoptera: Scarabaeidae). Biol. J. Linn.
Soc. 96: 135–149.
Puniamoorthy, N., Kotrba, M. & Meier, R. 2010. Unlocking
the “Black box”: internal female genitalia in Sepsidae (Dip-
tera) evolve fast and are species-specific. BMC Evol. Biol. 10:
275.
Rambaut, A. 2007. Se-Al. http//tree.bio.ed.ac.uk/software/seal/
. Accessed 28 January 2016.
Rambaut, A., Suchard, M.A., Xie, D. & Drummond, A.J. 2014.
Tracer v1.6. http://beast.bio.ed.ac.uk/Tracer. Accessed 28
January 2016.
Ramos, M., Coddingtion, J.A., Christenson, T.E. & Irschick,
D.J. 2005. Have male and female genitalia coevolved? A
phylogenetic analysis of genitalic morphology and sexual
size dimorphism in web-building spiders (Araneae: Arane-
oidea). Evolution 59: 1989–1999.
Reinhard, H.J. 1941. The life history of Phyllophaga tristis and
allied forms. J. Econ. Entomol. 34: 526–532.
Richmond, M.P. 2014. The role of aedeagus size and shape in
failed mating interactions among recently diverged taxa in
the Drosophila mojavensis species cluster. BMC Evol. Biol. 14:
255.
Robertson, H.M. & Paterson, H.E.H. 1982. Mate recognition
and mechanical isolation in Enallagma damselflies (Odonata:
Coenagrionidae). Evolution 36: 243.
Ronquist, F. & Huelsenbeck, J.P. 2003. MrBayes 3: Bayesian
phylogenetic inference under mixed models. Bioinformatics
19: 1572–1574.
Smith, J.B. 1888. Notes on the species of Lachnosterna of tem-
perate North America, with descriptions of new species. Proc.
US Natl. Mus. 11: 481–525.
Sota, T. & Kubota, K. 1998. Genital lock-and-key as a
selective agent against hybridization. Evolution 52:
1507–1513.
Tanabe, T. & Sota, T. 2008. Copulatory behavior and the prox-
imate effect of genital and body size differences on mechani-
cal reproductive isolation in the millipede genus Parafontaria.
Am. Nat. 171: 692–699.
Tanabe, T. & Sota, T. 2010. Multiple speciation events in an
arthropod with divergent evolution in sexual morphology.
Proc. R. Soc. B 277: 689–696.
Tanabe, T. & Sota, T. 2014. Both male and female novel traits
promote the correlated evolution of genitalia between the
sexes in an arthropod. Evolution 68:441–452.
Werner, M. & Simmons, L.W. 2008. The evolution of male gen-
italia: functional integration of genital sclerites in the dung
beetle Onthophagus taurus.Biol. J. Linn. Soc. 93: 257–266.
West-Eberhard, M.J. 1983. Sexual selection, social competi-
tion, and speciation. Q. Rev. Biol. 58: 155–183.
Wojcieszek, J.M. & Simmons, L.W. 2012. Evidence for stabiliz-
ing selection and slow divergent evolution of male genitalia
in a millipede (Antichiropus variabilis). Evolution 66: 1138–
1153.
Wojcieszek, J.M. & Simmons, L.W. 2013. Divergence in genital
morphology may contribute to mechanical reproductive iso-
lation in a millipede. Ecol. Evol. 3: 334–343.
Wojcieszek, J.M., Austin, P., Harvey, M.S. & Simmons, L.W.
2012. Micro-CT scanning provides insight into the functional
morphology of millipede genitalia. J. Zool. 287:91–95.
Woodruff, R.E. & Beck, B.M. 1989. The Scarab Beetles of Florida
Part II. Florida Department of Agriculture and Consumer
Services, Gainesville.
Yassin, A. & Orgogozo, V. 2013. Coevolution between male
and female genitalia in the Drosophila melanogaster species
subgroup. PLoS One 8: e57158.
Supporting information
Additional Supporting Information may be found
online in the supporting information tab for this article:
Table S1 Marker information and PCR conditions.
Table S2 Specimen Table.
Table S3 Log marginal likelihoods (calculated from the
harmonic mean and stepping stones sampler) of charac-
ter states at nodes with uncertain ancestral state recon-
structions.
Figure S1 Ventral view from female perspective.
Figure S2 Spines on the inner surface of the female
bursa copulatrix pointing in the direction the male
aedeagus is inserted.
Figure S3 Stem of male aedeagus. Hole where sperm
are potentially emitted denoted with *.
Received 2 February 2016; revised 23 June 2016; accepted 13 July
2016
ª2016 EUROPEAN SOCIETY FOR EVOLUTIONARY BIOLOGY. J. EVOL. BIOL. doi: 10.1111/jeb.12955
JOURNAL OF EVOLUTIONARY BIOLOGY ª2016 EUROPEAN SOCIETY FOR EVOLUTIONARY BIOLOGY
Evolution of male and female genitalia 13
