Family-Level Relationships of The Spittlebugs and

See discussions, stats, and author profiles for this publication at: https://www.researchgate.
net/publication/230029120
Family-level relationships of the spittlebugs and froghoppers (Hemiptera:

Cicadomorpha: Cercopoidea)
Article in Systematic Entomology · March 2010

DOI: 10.1111/j.1365-3113.2009.00520.x
CITATIONS READS
102 641
2 authors:
Jason R. Cryan Gavin J. Svenson

University of Utah Cleveland Museum of Natural History
90 PUBLICATIONS 2,265 CITATIONS 84 PUBLICATIONS 1,718 CITATIONS
SEE PROFILE SEE PROFILE
All content following this page was uploaded by Jason R. Cryan on 24 April 2024.
The user has requested enhancement of the downloaded file.

Systematic Entomology (2010), 35, 393–415 DOI: 10.1111/j.1365-3113.2009.00520.x
Family-level relationships of the spittlebugs and

froghoppers (Hemiptera: Cicadomorpha: Cercopoidea)
J A S O N R . C R Y A N and G A V I N J . S V E N S O N
Laboratory for Conservation and Evolutionary Genetics, New York State Museum, Albany, NY, U.S.A.
Abstract. The spittlebug superfamily Cercopoidea (Hemiptera: Cicadomorpha) com-

prises approximately 3000 phytophagous species (including some economically impor-
tant pests of grass crops) classified among the families Cercopidae, Aphrophoridae,
Epipygidae, Clastopteridae and Machaerotidae. However, the monophyly of these taxa
has never been tested and the evolutionary relationships among these major lineages
are unknown. Presented here are the results of the first ever phylogenetic investiga-
tion of the higher-level relationships within Cercopoidea, based on DNA nucleotide
sequence data from six loci (18S rDNA, 28S rDNA, histone 3, wingless, cytochrome
oxidase I and cytochrome oxidase II) generated from exemplars of 109 spittlebug
species representing all five described families, seven of eight subfamilies and 61
genera (eight additional exemplars, representing a selection of other Auchenorrhyn-
cha taxa, were included as outgroups). The resulting topologies are used to evaluate the
monophyly of each cercopoid family, and further to calculate divergence date estimates
to examine the chronological origins and historical diversification of Cercopoidea.
The results of this investigation suggest that: (i) four of the five described fami-
lies are monophyletic; Epipygidae was recovered consistently as originating within
Aphrophoridae; (ii) the exclusively Old World Machaerotidae is the most anciently
diversified family of extant spittlebugs; (iii) New World Cercopidae (i.e. Ischnorhini-
nae) constitute a derived monophyletic lineage; (iv) the genus Microsargane Fowler,
classified currently within Aphrophoridae, actually belongs within Cercopidae; and (v)
the origins of the major spittlebug lineages probably coincided with the breakup of
Pangaea and, subsequently, Gondwana, as well as major floristic diversification such
as the rise of angiosperms.
Introduction discharged from the insect’s alimentary system and supple-

mented by mucopolysaccharides and proteins produced by
The insect superfamily Cercopoidea (Hemiptera: Auchen- the specialized Malpighian tubules of the immature insects
orrhyncha: Cicadomorpha) comprises approximately 3000 (Rakitov, 2002); these spittle masses provide protection from
described species distributed among approximately 340 gen- predation, parasitism and desiccation. Family-level differences
era in five families (Cercopidae, Aphrophoridae, Clastopteri- are apparent in this life-history strategy: nymphs of Cercopidae
dae, Machaerotidae and Epipygidae) that are collectively and Aphrophoridae tend to produce copious spittle masses with
called spittlebugs, froghoppers or, less commonly, cuckoo-spit relatively large air bubbles, whereas nymphs of Clastopteridae
fabricate smaller, more viscous spittle masses with smaller air
insects. The common name ‘spittlebug’ refers to the nymphal
bubbles (Carvalho & Webb, 2005); nymphs of Machaeroti-
habit of covering themselves with a frothy, saliva-like mass
dae are generally found on their host plants immersed in fluid
composed of tiny air bubbles trapped in processed plant fluids
within calcified tubes formed from excretory products, and pro-
Correspondence: Jason R. Cryan, Laboratory for Conserva- duce small spittle masses when outside of the tubes during
tion and Evolutionary Genetics, New York State Museum, 3140 moulting (Evans, 1940; Maa, 1963; Rakitov, 2002). Although
Cultural Education Center, Albany, NY 12230, U.S.A. E-mail: nymphs of Epipygidae are unknown (and no direct observa-
[email protected] tions on their biology have been recorded), it is speculated
© 2010 The Authors

Journal compilation © 2010 The Royal Entomological Society 393
394 J. R. Cryan and G. J. Svenson
that they may not be efficient spittle producers (Hamilton, and selfing rates, suggesting that cercopoid feeding affects the
2001). Spittlebug immatures can be conspicuous, although their genetic structure of host plant populations (Ivey et al., 2004).
identification is usually impossible due to a lack of diagnostic Equally serious, perhaps, is the role cercopoids play in the
keys for nymphs. Spittlebug adults are generally identifiable to transmission of plant diseases. As Cercopoidea are exclusively
genus, but are often difficult to identify to species due to insuf- xylem feeders, and no viruses replicate in or are transported
ficient taxonomic keys and species illustrations; most cannot by plant xylem tissue, spittlebugs are potential vectors of only
be associated with their nymphal stages. bacterial diseases (Nault, 1987). Spittlebugs are competent vec-
Auchenorrhyncha (a hemipteran suborder containing Ful- tors for the causal agents of such diseases as sugar cane blight,
goroidea, Membracoidea, Cicadoidea and Cercopoidea) orig- gummosis of peaches, peach yellows and lucerne dwarf dis-
inated in the mid-Permian, approximately 270 mya (Shcher- ease (Hanna, 1966; Hamilton, 1982). Several spittlebug species
bakov, 1996, 2002) and their early diversification was concur- were implicated as potential vectors of Xylella fastidiosa, the
rent with the breakup of Pangaea and the rise of angiosperms causal agent of alfalfa dwarf disease and Pierce’s disease
(estimated to have originated 130–145 mya; Grimaldi & Engel, of grapes; of those species, only Philaenus spumarius (as
2005). Cercopoidea (and specifically, the extinct family Procer- P. leucophthalmus) Linnaeus is considered a ‘moderate’ poten-
copidae) appeared first in the geological record during the early tial threat as an invasive vector (Redak et al., 2004). Indonesian
Jurassic, approximately 205 mya, with subsequent family-level species in the genus Hindola (Machaerotidae) transmit xylem-
diversification apparently taking place during the Cretaceous limited bacteria causing Sumatra disease of cloves (Eden-
and Tertiary (Bekker-Migdisova, 1962; Shcherbakov, 2002). Green et al., 1992). Spittlebugs may vector the fungus Diplodia
The extant families Aphrophoridae and Cercopidae are known pini, which can cause flagging injury/blight disease of Scots
from mid to late Cretaceous deposits, approximately 100–90 pine, as well as eastern white, jack Japanese, loblolly, mugo,
mya (Shcherbakov, 2002; Szwedo, personal communication). pitch and Virginia pines (Hanna, 1966; Hamilton, 1982).
Within Cicadomorpha, Cercopoidea is the sister group to the Whereas spittlebugs occur in most terrestrial environments
cicada superfamily Cicadoidea (Cryan, 2005). Like cicadas, around the world, the richest cercopoid species diversity
spittlebugs feed on fluid contained in xylem tissue and many is pan-tropical: approximately 70% of described spittlebug
species exhibit a strong preference for nitrogen-fixing plants species are tropical, occurring either in the New World or
(Thompson, 1994); furthermore, different cercopoid families Old World tropics. Some species are highly habitat and host
seem to favour particular nitrogen-fixing plant groups (Thomp- specific and are therefore valuable indicators of ecological
son, 1999). Worldwide, spittlebugs inflict heavy economic diversity in habitats ranging from tropical rainforests (Adis
damage to grass crops (Heinrichs & Barrion, 2004), including & Schubert, 1984) to temperate grasslands (Panzer et al.,
rice (Pacheco et al., 1984; Wilson & Claridge, 1991), sugar- 1995; Arenz & Joern, 1996). Relatively little is known about
cane (Rodman & Miller, 1992), corn (Peck et al., 2001) and spittlebug biodiversity; Kosztarab et al. (1990) estimated that
improved pasture grasses such as Brachiaria sp. and Axonopus less than 20% of the world’s cercopoid fauna is described.
sp. (Clark et al., 1980; Pires et al., 2000; Holmann & Peck, Of the described species of Cercopoidea, less than 30% have
2002); for example, spittlebugs in the genera Aeneolamia, been described adequately or illustrated from a taxonomic
Mahanarva, Zulia, Deois, Notozulia and Prosapia are consid- perspective (Hamilton in Carvalho & Webb, 2005).
ered serious pests of Central and South American grass crops, Perhaps unsurprisingly for such a systematically neglected
causing up to a 70% reduction in agricultural yields within (yet biologically important) lineage, even the family-level clas-
infested areas (Sanz, 1997; Thompson, 2004; Leite et al., 2005; sification of Cercopoidea remains controversial: although five
Thompson & León González, 2005). cercopoid families have been described, various workers rec-
Cercopoid damage to host plants is often less obvious than ognize three (Hamilton, 2001), four (Dietrich, 2002; Holzinger
damage resulting from infestations of other agricultural pests et al., 2003) or five families (Dietrich, 2005). Whereas some
(i.e. grasshoppers, beetles, leafhoppers) and may be under- of this incongruity is due to disagreement about the taxonomic
estimated (Kosztarab et al., 1990). Physical damage to host rank of certain lineages (i.e. a given lineage might be recog-
plants is usually shown by interference of plant growth due to nized as either a family or a subfamily by different workers),
excessive liquid and nutrient removal and/or deformation of some is also due to fundamental differences in the placement
leaves, seeds and fruits due to the puncturing of plant tissues. of certain lineages; for example, Hamilton (2001) proposed
This injury results in increased plant respiration and inhibi- a reclassification of Cercopoidea that included Machaerotidae
tion of translocation and, ultimately, a general disruption of (as Machaerotinae) within Clastopteridae, citing as evidence
equilibrium (Clark et al., 1980) that reduces plant fitness and, the deep antennal pits observed in both groups and reasoning
consequently, plant value. Other research has demonstrated that that a single hind wing vein differentiates Machaerotidae and
spittlebugs can reduce seed set, photosynthetic rates, biomass, Clastopteridae taxonomically. Hamilton (2001) also suggested,
development and flower number, and can affect pollinator visi- based on features such as wing folding patterns and foreleg
tation (Parman & Wilson, 1982; Zajac & Wilson, 1984; Meyer articulation, that Aphrophoridae is polyphyletic and, therefore,
& Whitlow, 1992; Meyer, 1993; Meyer & Root, 1993; Cronin should not be recognized as a distinct group (see Discussion).
& Abrahamson, 1999; Carson & Root, 1999; Hamback, 2001; Even less is understood about spittlebug evolution, as cer-
Carr & Eubanks, 2002). There is evidence that spittlebugs alter copoid phylogeny remains largely unexplored; to date, the
plant community structure by changing inbreeding depression only quantitative analysis relevant to higher-level spittlebug
© 2010 The Authors

Journal compilation © 2010 The Royal Entomological Society, Systematic Entomology, 35, 393–415
Higher phylogeny of Cercopoidea 395
phylogeny is a study of the hemipteran infraorder Cicadomor- All amplicons were visualized with 1–2% agarose gel elec-
pha, based on DNA nucleotide sequence data from three genes trophoresis with ethidium bromide staining to verify amplifi-
(Cryan, 2005). That analysis examined relationships among cation, then purified using GeneClean (BIO 101, Vista, CA),
the superfamilies Membracoidea, Cicadoidea and Cercopoidea, QIAquick (Qiagen, Valencia, CA) or ExoSAP-IT (GE Health-
and therefore included only a small taxonomic sample (22 care Biosciences, Piscataway, NY) protocols. Gene regions
species) of spittlebugs. Within a strongly supported, mono- were sequenced with complements and sufficient overlap with
phyletic Cercopoidea, results indicated support for the mono- adjacent regions to ensure accuracy of sequence data. All
phyly of the families Cercopidae, Clastopteridae and Epipy- sequences were obtained using d-rhodamine terminator cycle
gidae, although the latter was included within a paraphyletic sequencing on an ABI 3730XL sequencer at the High Through-
Aphrophoridae. Only a single exemplar of Machaerotidae was put Genomics Unit at the University of Washington.
included; although the monophyly of that family was not
tested, the exemplar species was placed as a sister group to
the Clastopteridae. Alignment
Here we investigate the phylogeny of the major spittlebug
lineages in the context of a broader taxonomic sample, and Sequence fragments were imported into sequencher® 4.8
specifically: (i) test the hypothesis of monophyly for each of (Genecodes, 2006) and automatically trimmed to remove
the five described cercopoid families; (ii) reconstruct evolu- primer regions. Contigs were assembled after sequence inspec-
tionary relationships among the major spittlebug lineages; and tion and editing based on chromatograms to ensure the quality
(iii) test various taxonomic and biogeographical hypotheses of base calls. In addition, indels were confirmed for accuracy
articulated in existing literature on Cercopoidea. The results by using a reference sequence for comparison. All sequence
are based on six gene regions: the three loci (18S rDNA, 28S data have been submitted to GenBank (see Table 1 for a com-
rDNA and histone 3) sampled originally by Cryan (2005), but plete list of accession numbers).
now generated from a larger taxonomic sample, analysed in Contigs for protein-coding genes were imported into the
combination with DNA nucleotide sequence data from three program mega version 4 (Tamura et al., 2007) for alignment
additional loci [cytochrome oxidase I (COI), cytochrome oxi- based on the amino acid sequences. After the proper reading
dase II (COII) and wingless]. frame was determined, sequences were back translated with
codon position information and exported for concatenation
with the total combined dataset. The two ribosomal genes were
Materials and methods aligned manually in sequencher according to conserved and
nonconserved regions in order to trim the flanking sequences
Taxon sampling of the contigs to ensure only corresponding gene regions were
included. All gene contigs were exported in fasta format and
Insect specimens (Table 1) were collected into 95–100% aligned using the programs muscle version 3.6 (Edgar, 2004)
ethanol and are stored at −80◦ C in the New York State with default settings and mafft version 5.8 (Katoh et al., 2002,
Museum’s Genome Bank (Albany, NY). The 109 spittlebug 2005) with the L-INS-i algorithm and ‘maxiterate’ set to 1000.
species included in this study represent all five described Resulting ribosomal sequence data alignments from mus-
spittlebug families, seven (of eight) subfamilies, 18 (of ∼32) cle and mafft were independently concatenated with the
tribes and 61 (of ∼320) genera. Additionally, eight non- resulting protein-coding gene sequence data from mega using
Cercopoidea species were included as outgroups (Table 1), macclade version 4.06 (Maddison & Maddison, 2003), thus
representing the other major lineages of Auchenorrhyncha: yielding two separate character matrices (hereafter referred to
Cicadoidea (cicadas, the putative sister group of Cercopoidea; as the muscle and mafft matrices). The resulting aligned
Cryan, 2005), Membracoidea (leaf- and treehoppers) and dataset from muscle was 5834 bp in length; treating gaps
Fulgoroidea (planthoppers). as missing under the parsimony criterion, there were 1831
informative characters, 478 autapomorphic characters and
Molecular methods 3525 invariable characters. The resulting aligned dataset from
mafft was 5681 bp in length; treating gaps as missing
Molecular protocols for DNA extraction, amplification and under the parsimony criterion, there were 1844 informative
sequencing followed those of Cryan et al. (2000, 2001) characters, 417 autapomorphic characters and 3420 invari-
and Urban & Cryan (2007, 2009). In total, six loci (two able characters. Both data matrices are publicly accessible at
mitochondrial and four nuclear) were sequenced: 18S rRNA http://www.nysm.nysed.gov/lceg/datasets.html.
(∼1200 bp), 28S rRNA (∼1900 bp), COI (∼900 bp), COII
(∼630 bp), histone 3 (∼350 bp) and wingless (∼300 bp).
In addition to the previously optimized primers utilized in Phylogenetic reconstruction
this study, two new oligonucleotide primers were synthesized
for improved amplification yields of COI across a broad We performed a preliminary analysis of the total combined
selection of taxa within Cercopoidea, Cerc-MidF and Cerc- dataset including all six loci to determine the behaviour of
MidR (Table S1). the data under multiple phylogenetic reconstruction paradigms,
© 2010 The Authors

Table 1. Taxa sampled.
Taxa included in 18S, 28S, H3, Wg, COI, and COII nucleotide sequence data sets. Classification follows Metcalf (1961, 1962a, 1962b; for Aphrophoridae, Old World Cercopidae, and Clastopteridae),
Maa (1963; for Machaerotidae), Fennah (1968; for tribal classification of New World Cercopidae), Hamilton (2001; for Epipygidae), Liang & Webb (2002; for Rhinaulacini), and Carvalho &
Webb (2005; for New World Cercopidae).
GenBank Accession No.

Voucher Geographical
Taxon codea source 18S 28S H3 Wg COI COII
Cercopoidea
Aphrophoridae
Aphrophorinae
Aphrophorini
Aphrophora alni (Fallén) 01-07-15-14 U.S.A. (NY) AY744783 AY744817 AY744855 GU446722 GU446983 GU447062
Aphrophora parallella Say 01-07-15-21 U.S.A. (NY) AY744785 AY744819 AY744857 GU446724 GU446985 GU447063
Aphrophora permutata Uhler 05-02-07-53 U.S.A. (CA) GU446849 GU446941 GU447168 – GU447026 GU447098
Aphrophora quadrinotata Say 01-07-15-25 U.S.A. (NY) GU446794 GU446884 GU447120 GU446725 GU446986 GU447064
Aphrophora saratogensis (Fitch) 04-11-27-37 U.S.A. (NJ) GU446836 GU446926 GU447155 – GU447014 GU447091
Microsargane vittata Fowler 03-09-03-16 Costa Rica GU446806 GU446897 – GU446737 – –
Cloviini
Clovia sp. 1 04-11-27-06 India GU446829 GU446919 – GU446751 GU447007 GU447085
Clovia sp. 2 05-03-15-45 Zambia GU446879 GU446974 GU447194 GU446787 GU447054 –
Clovia sp. 3 04-11-27-20 Malaysia GU446834 GU446924 GU447153 GU446756 GU447012 GU447089
Epicranion sp. 1 03-09-03-51 Costa Rica GU446852 GU446945 GU447171 GU446764 GU447029 GU447100
Epicranion sp. 2 04-05-11-31 Peru GU446855 GU446948 GU447174 GU446767 GU447032 GU447103
Plinia sp. 05-01-15-49 Malaysia GU446864 GU446959 GU447182 – GU447041 GU447109
Sphodroscarta gigas (Fabricius) 04-10-15-11 Peru GU446824 GU446914 – GU446746 GU447002 GU447081
Lepyroniini
Avernus ochraceiventris Schmidt 04-10-15-20 Peru GU446825 GU446915 GU447147 GU446747 GU447003 GU447082
Lepyronia coleoptrata Linnaeus 01-07-15-11 U.S.A. (NY) AY744782 AY744816 AY744854 GU446721 GU446982 GU447061
Lepyronia quadrangularis Say 01-07-18-51 U.S.A. (NY) AY744786 AY744820 AY744858 GU446726 – GU447065
Peuceptyelus sp. 05-01-15-73 Taiwan GU446865 GU446960 GU447183 GU446775 GU447042 GU447110
Philaenini
Philaenarcys bilineata (Say) 04-11-27-31 U.S.A. (NH) GU446835 GU446925 GU447154 – GU447013 GU447090
Philaenus maghresignus Drosop. 01-07-15-42 Spain AY744793 AY744827 AY744865 – GU446993 –
& Remane
Philaenus spumarius Linnaeus 01-07-15-01 U.S.A. (VT) AY744779 AY744813 AY744851 GU446718 GU446979 –
Philaenus tesselatus Melicharb 01-07-15-17 Portugal GU446797 GU446887 GU447123 – GU446992 –
Mesoptyelus sp. 05-01-15-81 Taiwan GU446866 GU446961 GU447184 – GU447043 –
Neophilaenus lineatus Linnaeus 01-07-15-03 U.S.A. (VT) AY744780 AY744814 AY744852 GU446719 GU446980 GU447060
Philagrini
Philagra sp. 05-02-05-22 Taiwan GU446844 GU446936 GU447163 – GU447022 –
Ptyelini
Cephisus siccifolius (Walker) 03-01-15-33 Costa Rica GU446805 GU446896 GU447131 – – –
Poophilus costalis (Walker) 05-03-15-40 Zambia GU446880 GU446975 GU447195 – GU447055 GU447116
Poophilus sp. 05-03-15-41 Zambia GU446881 GU446976 GU447196 – GU447056 GU447117
Ptyelus grossus (Fabricius) 05-02-05-58 Ghana GU446871 GU446966 – GU446780 GU447048 GU447113
© 2010 The Authors
Table 1. Continued.

© 2010 The Authors

Cercopidae
Callitettixinae
Callitettixini
Caloscarta sp. 01-07-15-10 Thailand GU446796 GU446886 GU447122 GU446732 GU446991 GU447068
Cercopinae
Cosmoscartini
Cosmoscarta bispecularis (White) 04-12-30-34 China GU446839 GU446929/-30 GU447158 – GU447017 GU447093
Cosmoscarta heros (Fabricius) 04-12-30-35 China GU446840 GU446931 GU447159 GU446759 GU447018 GU447094
Cosmoscarta dimidiata (Dallas) 04-11-27-13 India GU446831 GU446921 – GU446753 GU447009 GU447087
Okiscarta uchidae Matsumura 05-01-15-29 Taiwan GU446862 GU446956 – GU446773 GU447039 GU447107
Euryaulacini
Leptataspis fuscipennis Le Peletier 04-12-30-45 Malaysia GU446841 GU446932 GU447160 GU446760 GU447019 GU447095
& Serville
Locrisini
Locris actuosa Lallemand 05-02-05-51 Ghana GU446869 GU446964 – GU446778 GU447046 GU447111
Locris maculata (Fabricius) 05-02-05-54 Ghana GU446870 GU446965 GU447187 GU446779 GU447047 GU447112
Locris pullata Stål 05-02-25-63 Zambia GU446874 GU446969 GU447189 GU446782 GU447050 –
Locris rubra (Fabricius) 05-03-15-09 Zambia GU446878 GU446973 GU447193 GU446786 GU447053 –
Locris sp. 05-02-05-79 Togo GU446872 GU446967 – GU446784 – –
Rhinaulacini
Aufidus trifasciatus Stål 04-05-11-01 Australia GU446813 GU446904 GU447137 – – –
Aufidus sp. nr. biplagiatus 04-05-11-33 PNGc GU446856 GU446949 GU447175 – GU447033 –
Aufidus sp. 3 04-12-30-03 Malaysia GU446838 GU446928 GU447157 GU446758 GU447016 GU447092
Colsa orientalis (Lallemand) 06-02-15-16 Malaysia GU446877 GU446972 GU447192 GU446785 – –
Jeanneliensia sublimpida Lallemand 05-02-07-25 Ghana GU446846 GU446938 GU447165 – GU447024 –
Sounama koshunella (Matsumura) 05-01-15-30 Taiwan GU446863 GU446957/-58 GU447181 GU446774 GU447040 GU447108
Suracartini
Suracarta tricolor (Le Peletier 04-12-30-49 Malaysia GU446843 GU446935 GU447162 GU446762 GU447021 –
& Serville)
Unplacedd
Machadoa invenusta (Jacobi) 05-02-25-64 Zambia GU446875 GU446970 GU447190 GU446783 GU447051 –
Ischnorhininae
Ischnorhinini
Baetkia compressa (Le Peletier 05-01-25-49 Fr. Guiana GU446868 GU446963 GU447186 GU446777 GU447045 –
& Serville)
Baetkia maroniensis Lallemand 05-02-25-47 Fr. Guiana GU446873 GU446968 GU447188 GU446781 GU447049 GU447114
Homalogrypota coccinea (Fabricius) 04-11-27-14 Bolivia GU446832 GU446922 GU447151 GU446754 GU447010 –
Laccogrypota grandis (Distant) 04-05-11-80 Peru GU446823 GU446913 GU447146 GU446745 GU447001 GU447080
Laccogrypota praelata (Jacobi) 04-10-15-47 Peru GU446827 GU446917 GU447149 GU446749 GU447005 GU447084
Higher phylogeny of Cercopoidea
397
Table 1. Continued.

Tomaspidini
Aeneolamia albofasciata 03-09-03-37 Costa Rica GU446808 GU446899 GU447133 GU446739 – GU447076
(Lallemand)
Aeneolamia contigua (Walker) 01-07-18-73 Costa Rica AY744794 AY744828 AY744866 – – –
Aeneolamia reducta (Lallemand) 01-07-18-68 Colombia GU446799 GU446889 GU447125 GU446734 – –
Aeneolamia varia (Fabricius) 01-07-18-80 Colombia GU446802 GU446892 GU447128 – – GU447072
Aracamunia dimorpha (Distant) 04-12-30-47 Peru GU446842 GU446933/-34 GU447161 GU446761 GU447020 GU447096
Catrimania insignis (Walker) 04-05-11-13 Peru GU446817 GU446908 GU447140 – – –
Catrimania semivitrea (Walker) 04-05-11-74 Peru GU446822 GU446912 GU447145 GU446744 GU447000 GU447079
Hemitomaspis caligata (Jacobi) 05-02-07-48 Fr. Guiana GU446848 GU446940 GU447167 – – GU447097
Huaina inca (Guérin-Méneville) 03-09-03-68 Costa Rica GU446810 GU446901 GU447135 GU446741 – GU447077
Iphirhina perfecta (Walker) 03-09-03-17 Costa Rica GU446807 GU446898 GU447132 GU446738 – GU447075
Iphirhina quota (Distant) 03-09-03-64 Costa Rica GU446854 GU446947 GU447173 GU446766 GU447031 GU447102
Isozulia astralis (Distant) 04-11-27-07 Bolivia GU446830 GU446920 GU447150 GU446752 GU447008 GU447086
Isozulia sp. nr. soluta 03-09-03-13 Costa Rica GU446851 GU446943/-44 GU447170 GU446763 GU447028 GU447099
Mahanarva costaricensis (Distant) 03-01-15-32 Costa Rica AY744798 AY744832 AY744870 GU446736 GU446995 GU447074
Mahanarva tristis monagasi 04-12-30-38 Peru GU446858 GU446951 GU447177 GU446769 GU447035 –
(Fennah)
Maxantonia diversa Nast 04-05-11-68 Peru GU446821 GU446911 GU447144 GU446743 GU446999 –
Maxantonia quadriguttata (Walker) 04-10-15-24 Peru GU446826 GU446916 GU447148 GU446748 GU447004 GU447083
Monecphora pallida Lallemand 05-01-15-05 Peru GU446860 GU446953 GU447179 GU446771 GU447037 GU447106
Ocoaxo ornatipennis (Stål) 03-01-15-07 Belize GU446804 GU446894/-95 GU447130 GU446735 GU446994 –
Ocoaxo tucurricae (Lallemand) 03-09-03-62 Costa Rica GU446853 GU446946 GU447172 GU446765 GU447030 GU447101
Notozulia entreriana (Berg) 04-12-30-07 Bolivia GU446857 GU446950 GU447176 GU446768 GU447034 GU447104
Notozulia sp. 05-02-05-45 Bolivia GU446845 GU446937 GU447164 – GU447023 –
Pachacanthocnemis bella (Walker) 04-12-30-43 Peru GU446859 GU446952 GU447178 GU446770 GU447036 GU447105
Prosapia bicincta (Say) 01-07-15-28 U.S.A. (GA) AY744789 AY744823 AY744861 GU446727 GU446987 GU447066
Prosapia plagiata (Distant) 04-05-11-59 Costa Rica GU446818 GU446909 GU447141 GU446742 GU446998 GU447078
Prosapia simulans (Walker) 01-07-18-76 Costa Rica GU446801 GU446891 GU447127 – – GU447071
Sphenorhina latifascia Walker 01-07-18-81 Costa Rica GU446803 GU446893 GU447129 – – GU447073
Sphenorhina parambae (Jacobi) 05-01-25-42 Peru GU446867 GU446962 GU447185 GU446776 GU447044 –
Sphenorhina proserpina (Distant) 04-11-27-15 Bolivia GU446833 GU446923 GU447152 GU446755 GU447011 GU447088
Sphenorhina rubra (Linnaeus) 04-12-30-02 Bolivia GU446837 GU446927 GU447156 GU446757 GU447015 –
Sphenorhina distinguenda Walker 01-07-18-75 Costa Rica GU446800 GU446890 GU447126 – – GU447070
Tomaspis biolleyi (Distant) 03-09-03-66 Costa Rica GU446809 GU446900 GU447134 GU446740 – –
Zulia carbonaria (Lallemand) 01-07-18-66 Colombia GU446798 GU446888 GU447124 GU446733 – GU447069
Zulia vilior (Fowler) 01-07-18-36 Costa Rica AY744791 AY744825 AY744863 GU446729 GU446989 GU447067
Clastopteridae
Clastopterinae
Clastopterini
Clastoptera brunnea Ball 01-07-18-58 U.S.A. (UT) AY744790 AY744824 AY744862 GU446728 GU446988 –
Clastoptera laenata Fowler 04-05-11-63 Costa Rica GU446820 – GU447143 – – –
Clastoptera proteus Fitch 01-07-15-09 U.S.A. (NY) AY744781 AY744815 AY744853 GU446720 GU446981 –
© 2010 The Authors
Table 1. Continued.

Clastoptera obtusa Say 01-07-15-15 U.S.A. (NY) AY744784 AY744818 AY744856 GU446723 GU446984 –
© 2010 The Authors

Clatoptera testacea Fitch 01-07-18-59 U.S.A. (WV) AY744792 AY744826 AY744864 GU446730 GU446990 –
Clastoptera xanthocephala Germar 01-07-15-06 U.S.A. (FL) GU446795 GU446885 GU447121 GU446731 – –
Clastoptera sp. 04-05-11-60 Costa Rica GU446819 GU446910 GU447142 – – –
Epipygidae
Epipyga n.sp. “a” NYSM CER089 Peru AY744795 AY744829 AY744867 – – –
Undetermined Epipygidae 1 NYSM CER090 Peru AY744796 AY744830 AY744868 – – –
Undetermined Epipygidae 2 NYSM CER091 Peru AY744797 AY744831 AY744869 – – –
Eicissus sp. 05-02-07-46 Fr. Guiana GU446847 GU446939 GU447166 – GU447025 –
Machaerotidae
Enderleiniinae
Enderleiniini
Apomachaerota reticulata Schmidt 06-02-15-17 Malaysia GU446882 GU446977 GU447197 GU446788 GU447057 –
Chaetophyes vicina Lallemand 03-09-10-72 Australia GU446811 GU446902 – GU446996 – –
Enderleinia bispina Schmidt 05-02-25-29 Uganda GU446850 GU446942 GU447169 – GU447027 –
Pectinariophyes hyalinipennis (Stål) 05-02-25-79 Zambia GU446876 GU446971 GU447191 – GU447052 GU447115
Pectinariophyes reticulata CHD AH9 Australia AY744778 AY744812 AY744850 GU446717 – GU447059
(Spångberg)
Pectinariophyes stalii (Spångberg) 03-09-10-73 Australia GU446816 GU446907 GU447139 – – –
Polychaetophyes serpulidia 03-09-10-80 Australia GU446812 GU446903 GU447136 – – –
Kirkaldy
Taihorina geisha Schumacher 05-01-15-26 Taiwan GU446861 GU446954/-55 GU447180 GU446772 GU447038 –
Hindoloidini
Hindoloides bipunctatus (Haupt) 04-05-11-10 Japan GU446814 GU446905 GU447138 – GU446997 –
Machaerotinae
Machaerotini
Grypomachaerota turbinata Schmidt 06-02-15-19 Malaysia GU446883 GU446978 GU447198 GU446789 GU447058 GU447118
Machaerota pugionata Stål 04-05-11-45 Australia GU446815 GU446906 – – – –
Machaerota takeuchii Kato 04-10-15-80 Japan GU446828 GU446918 – GU446750 GU447006 –
Outgroups
Cicadoidea
Cicadidae
Froggattoides typicus Distant CS 02.QLD.ANT.5 Australia AY744800 AY744834 AY744872 – – –
Tettigarctidae
Tettigarcta tomentosa White CS 00-09 Australia AY744803 AY744837 AY744875 – – –
Membracoidea
Aetalionidae
Gerridius fowleri (Haviland) 97-02-19-46e Guyana AY498432 AY744843 AY744881 GU446791 – –
Cicadellidae
Flexamia areolata (Ball) CHD LH038 U.S.A. (VA) AY498437 AY744845 AY744883 GU446792 – –
Membracidae
Deiroderes inermis Ramos CRB Din Tortolaf AY498425 AY744841 AY744879 GU446790 – –
399
alternate alignments of ribosomal genes performed using
d Machadoa was classified within Tomaspidini (Metcalf, 1961), but Fennah (1968) considered Tomaspidinae is not currently recognized as a tribe of Cercopinae, but Machadoa has not been
GU447119
either mafft or muscle and exclusion of third codon posi-
COII
tions and/or variable regions of ribosomal genes. Initially,
–
we employed the parsimony program tnt (Goloboff et al.,
2003) to determine the differences between tree topologies
derived from the total combined dataset, excluding hyper-
EU646029
variable regions of 18S rRNA and 28S rRNA, including
hyper-variable regions aligned with mafft and including
COI
hyper-variable regions aligned with muscle. In addition, cer-

–
–
tain data partitions were excluded from the total combined
GU446793 dataset before implementation in tnt to investigate topological
EU645960
differences when all third codon positions were excluded, only

third codon positions of nuclear genes were excluded and only
Wg
third codon positions of mitochondrial genes were excluded.

–
A combination of third codon inclusion/exclusion with inclu-

sion/exclusion of ribosomal variable regions was performed.
DQ532670
Identical dataset treatments were implemented in the maximum

AY744886
EU645915
likelihood (ML) program garli version 0.951 (Zwickl, 2006;

http://www.nescent.org/informatics/download.php?software id
H3
= 4). We investigated six dataset partitioning strategies within

a mixed model Bayesian (MMB) analysis including: (i) one
partition – total combined dataset; (ii) two partitions – (a) all
DQ532596
AY744848
EU645859
mitochondrial genes, (b) all nuclear genes; (iii) four par-

titions – (a) protein-coding mitochondrial genes, (b) protein-

28S
coding nuclear genes, (c) 18S rRNA, (d) 28S rRNA; (iv) six
partitions – (a–f) each gene locus; (v) seven partitions – (a–d)
first + second codon positions for histone 3, wingless, COI
DQ532516
AY744808
EU645792
and COII, (e) third position codons, (f) 18S rRNA, (g) 28S
rRNA; and (vi) 14 partitions – (a–l) codon position for each
18S
protein-coding gene, (m) 18S rRNA, (n) 28S rRNA. In addi-

tion, we investigated measurements of saturation in the third
codon position in all four protein-coding genes by constructing
Geographical
saturation curves based on sequence distances. For all dataset

Fr. Guiana
treatments, we recorded topological findings to select the final

Australia
St. Johng
source
dataset treatment based on the criteria of relationship stability

and dataset congruence with traditional classification.
Using the Akaike information criteria, the best-fit model
for each of the six genes and the total combined dataset
was determined using modeltest 3.7 (Posada & Crandall,
04-12-09-69
04-12-28-01
02-06-17-09
1998) partnered with paup* 4.0b10 (Swofford, 2002). A final

Voucher
codea
ML analysis was performed using garli version 0.951

listed as P. spumarius tesselatus Melichar.
(Zwickl, 2006). Using the GTR+I+G model with unfixed

parameter values, ten independent likelihood searches (each
with 1.5 million generations) were performed to ensure that
voucher codes from NCSU Genome Bank.
a globally optimal solution was reached. Nonparametric

York State Museum Genome Bank.
likelihood bootstrap trees were reconstructed using garli

Harmalia ostorius (Kirkaldy)
(500 replicates – 400 000 generations per replicate) and

formally moved into another tribe.
Fulgora laternaria Linnaeus
imported into paup for calculation of a 50% majority-rule

Tangia viridis (Walker)
consensus tree. Using treerot v.3 (Sorenson & Franzosa,

c PNG = Papua New Guinea.
2007), we calculated partitioned Bremer support values (six

British Virgin Islands.
partitions based on gene loci) by creating a constrained nodal

Table 1. Continued.
Tropiduchidae
search command file based on the ML tree topology for

Fulgoridae
US Virgin Island.
Delphacidae
implementation in paup.
Fulgoroidea
A final MMB analysis coupled with Markov chain Monte

b sometimes
Carlo was implemented in the program mrbayes v3.1.2

Taxon
a New
(Ronquist & Huelsenbeck, 2003). We conducted these analyses

with six gene partitions implemented across four independent
g
e
f
© 2010 The Authors

runs, each with ten Markov chain Monte Carlo chains 2.0; Covariance, Normal Prior mean = 1.0, stdev = 2.0;
(nine hot and one cold) run for 60 million generations. LogEvery = 1000.
Each run was started from a random tree and monitored Four node calibrations (enforced as monophyletic) were
subsequently (for fluctuating likelihood values) to ensure imposed following the time estimates approximated by
that each Bayesian analysis converged (comparable means Shcherbakov (2002) and Szwedo (personal communication);
and variances of likelihood parameters) using the program these node calibrations were input as date ranges, explained
tracer v1.4.1 (Rambaut & Drummond, 2007). All sampled below, to allow flexibility in divergence date reconstruction:
generations (every 4000) prior to stationarity were discarded (i) root height(= Auchenorrhyncha): normal prior distribution,
(burn-in). The final posterior probability tree was calculated as mean = 300.0, stdev = 12.5 for a date range of 275–325
a 50% majority-rule consensus tree (Huelsenbeck & Imennov, within two standard deviations; this date range was selected
2002; Huelsenbeck et al., 2002). as a relaxed modification of the ca. 270 mya estimate of
A final parsimony analysis was implemented in tnt the first appearance of Auchenorrhyncha in the fossil record
(Goloboff et al., 2003) under the ‘new technology search’ (Shcherbakov, 2002), the modification reflecting the rationale
treating gaps as missing and utilizing tree drifting, sectorial that this lineage had already diversified before ca. 270 mya; (ii)
searches, tree fusing (Goloboff, 1999) and ratchet (Nixon, Cercopoidea/outgroup: normal prior distribution, mean = 240,
1999). In total, 5000 random additions were completed with stdev = 10.0 for a date range of 220–260 within two stan-
the ‘new technology search’ repetition settings increased as dard deviations; this date range was selected as a relaxed
follows: ratchet set to 20 iterations, tree drift set to 20 cycles modification of the ca. 213 mya estimate (Shcherbakov, 1996)
and tree fusing set to six rounds. A nonparametric bootstrap and ca. 205 mya estimate (Shcherbakov, 2002) of the appear-
analysis (Felsenstein, 1985) was performed using paup 4.0b10 ance of Cercopoidea in the fossil record, the modification
(1000 iterations, 30 random additions per iteration). reflecting the rationale that this lineage had already diver-
sified before ca. 213 mya; (iii) Procercopidae: normal prior
distribution, mean = 210.0, stdev = 5.0 for a date range of
Topology testing
200–220 within two standard deviations; this date range was
selected as a relaxed modification of the ca. 205 mya esti-
Utilizing the ML topology we imposed constraints to test
mate (Shcherbakov, 2002) of the appearance of Procercopidae
some previous taxonomic and biogeographical hypotheses:
four independent reverse constraint analyses were conducted in the fossil record, the modification reflecting the rationale
in garli (1.5 million generations) to test: (i) the monophyly that this lineage had probably diversified before ca. 205 mya;
of Aphrophoridae; (ii) a monophyletic Clastopteridae + this extinct taxon is not included in the taxonomic sample,
Machaerotidae (Hamilton, 2001); (iii) the monophyly of New and therefore the date range of diversification for Procercop-
World (i.e. Neotropical + Nearctic) Cercopidae; (iv) the idae was assigned to the common ancestor of Cercopoidea
monophyly of Old World Cercopidae; and (v) the position of (Fig. 1, node 1); (iv) Aphrophoridae + Cercopidae: normal
Cercopidae as the most ancient of extant Cercopoidea, with prior distribution, mean = 140.0, stdev = 20.0 for a date range
Machaerotidae subsequently derived from Cercopidae (Maa, of 100–160 within two standard deviations reflecting the min-
1963). Subsequently, we performed Shimodaira–Hasegawa imum time estimate for Cercopidae at 100 mya; this date
tests (Shimodaira & Hasegawa, 1999; full optimization, 5000 range was selected as a relaxed modification of the ca. 100
bootstrap replicates) in paup to determine if we could reject mya estimate (Shcherbakov, 2002) of differentiation between
the null hypothesis for each of the five tests. Cercopidae sensu stricto and Aphrophoridae sensu stricto, the
relaxation of dates reflecting the rationale that differentiation
Divergence time estimation had already taken place by ca. 100 mya. We used normal dis-
tributions to calibrate the four nodes while enforcing a mean
A divergence time estimation was carried out using the node height and standard deviation that would reflect minimum
program beast v1.4.8 (Drummond et al., 2002, 2005, 2006; age estimates. Therefore, the mean node heights are older than
Drummond & Rambaut, 2007). To prepare command files for the minimum age estimate for the node in order to place the
execution within beast we employed beauti v1.4 (Rambaut minimum age estimate at the younger extreme of the normal
& Drummond, 2007) by importing the total combined data distribution. In addition, nine nodes were enforced to be mono-
matrix and selecting the following settings (all other settings phyletic with a uniform prior distribution of heights (lower 0.0,
remained on default): Yule Speciation Process; Relaxed upper 1000.0), including nodes 51, 57, 58, 71, 72, 73, 80, 83
Clock: Uncorrelated Lognormal; GTR gammaCategories = 4, and 90 (see Results). The program tracer v1.3 was used to
pInv; ChainLength = 30 000 000; SiteModel.alpha, Normal determine proper burn-in and to monitor the likelihoods of four
Prior mean = 1.0, stdev = 1.0, initial = 0.5; Site Model.pInv, independent analyses to ensure that equilibrium and conver-
Normal Prior mean = 1.0, stdev = 1.0, initial = 0.5; ucld. gence of likelihood distributions were obtained. A majority-
mena, Normal Prior mean = 1.0, stdev = 1.0, initial = 0.0075; rule consensus tree was calculated using paup and saved as
ucld.stdev, Normal Prior mean = 1.0, stdev = 1.0, initial = the user target tree. The tree files and the user target tree were
0.1; Mean Rate, Normal Prior mean = 1.0, stdev = 2.0; imported into the program treeannotator (Drummond &
Coefficient of Variation, Normal Prior mean = 1.0, stdev = Rambaut, 2002–2007) to calculate a rescaled posterior mean
© 2010 The Authors

© 2010 The Authors

node height for the clades contained in the majority-rule con- subsequent analyses, including phylogenetic, nodal support and
sensus tree. The subsequent tree file was imported into the divergence time estimation (Table S3).
program figtree v1.2.2 (Rambaut, 2007) to view and save the Conducting the same preliminary analysis using the ML
final time calibrated consensus tree from the beast analysis. program garli corroborated our preliminary parsimony results.
In addition, our investigation of partitioning strategy in
MMB analysis revealed only minor differences between
Results treatments. Therefore, we selected the most standard method
of partitioning to conduct our final analysis (strategy 4, six
Phylogenetic reconstruction partitions by gene locus).
Calculating the best-fit model using the Akaike information
The preliminary parsimony analysis revealed that topology criteria in modeltest for each of the six gene loci and
was influenced by the inclusion/exclusion of the hyper- the total combined dataset all recovered a six parameter
variable rRNA gene regions and third codon positions of model with a gamma distribution (G) and proportion of
cytochrome oxidase (CO) (Table S2). Analysis of the full invariant sites (pInv) enabled for among-site rate variation.
datasets aligned by either mafft or muscle recovered Therefore, a six parameter model with G and pInv enabled
Machaerotidae as the earliest branch of Cercopoidea, but each was employed for ML and MMB analysis. The ML topology
alignment method resulted in a paraphyletic Aphrophoridae was recovered in three of ten independent runs of the program
with respect to Clastopteridae. Cutting the hyper-variable garli with a likelihood of −53185.0041 with the following
rRNA gene regions from both alignments had little influence model parameters: r(AC) 1.2110093, r(AG) 2.9158277, r(AT)
on the overall topology and appeared to be noncritical in 1.8995722, r(CG) 0.83445686, r(CT) 4.4071287, bfreqA
recovering familial-level relationships. However, removing 0.2233444, bfreqC 0.25383427, bfreqG 0.26892854, bfreqT
third codon positions for all protein-coding genes resulted in a 0.25389279, alpha 0.55650228 and pinv 0.5833645 (Fig. 1).
topology with four monophyletic families for each alignment A likelihood bootstrap analysis recovered 64 of 108 nodes
method, but the earliest branching family was incongruent (59.3%) with a value of 70 or greater, whereas lower bootstrap
between muscle and mafft alignments (see Table S2, ‘No values (less than 70) were concentrated at deeper-level or
CO 3rd’). Further inclusion/exclusion treatments of third codon backbone nodes (Fig. 1, Table 2). Terminal nodes, those
positions by gene revealed that CO third positions had great recovering relationships within genera or between two genera,
influence over topology, whereas histone 3 and wingless exhibited bootstrap values of 70 or greater for 76% of the
were noncritical in recovering deep-level relationships. When nodes (51 of 67) (Table 2).
excluding CO third positions, Clastopteridae was recovered The four independent runs of the MMB analysis all
outside of Aphrophoridae, but differed in placement between converged on a single likelihood distribution with a mean
alignments, which suggests that the alignment of hyper- likelihood of −52400.0 (run1: mean = −52400.0 stdev =
variable rRNA gene regions had influence on topology in 0.467, run2: mean = −52400.0 stdev = 0.398, run3: mean =
combination with excluded CO third positions (Table S2). −52400.0 stdev = 0.345 and run4: mean = −52400.0 stdev =
Only the muscle alignment was consistent in early branching 0.306). The majority-rule consensus topology (first 5 000 000
relationships when CO third positions were excluded in discarded as burn-in) recovered posterior probabilities with a
combination with included or excluded hyper-variable rRNA value of 80 or greater for 92 of 108 nodes (85%) of the ML
regions. Calculating saturation curves for CO third codon topology (Fig. 2, Table 2). Only eight nodes were incongruent
positions recovered a much steeper curve with a distinct plateau (DNS) with the ML topology (Table 2), which was primarily
when compared with the more direct curves calculated from caused by the placement of the Clastopteridae (node 14, Fig. 1)
histone 3 and wingless third codon positions. Therefore, we within the Aphrophoridae (node 21, Fig. 1) as well as minor
selected the more conservative alignment (muscle) to recover relationship incongruities within clade 59 Cercopidae (Fig. 1,
consistent relationships regardless of inclusion/exclusion of Table 2).
hyper-variable rRNA gene regions (Table S2). We excluded Parsimony analysis recovered four equally parsimonious
CO third codon positions for the final analyses based partly on trees with a cost of 9765 (consistency index = 0.305, retention
topological influence and levels of saturation. Consequently, index = 0.609 for strict consensus topology). We recovered
the final molecular dataset (5317 bp) derived from alignments topological congruence for 83 of the 108 nodes illustrated
performed in muscle, excluding CO third codon positions but in the ML topology (Fig. 1). The predominant topological
including hyper-variable rRNA gene regions, was used in all differences recovered between parsimony and ML analysis
Fig. 1. Maximum likelihood phylogram recovered in ten independent garli analyses (likelihood: −53185.0041). Nodes are numbered and
correspond with the branch support values and posterior probabilities presented in Table 2. Circles with white centres are present at nodes that
are congruent between strict consensus parsimony and maximum likelihood topologies. Nodes with support measure (MLB, maximum likelihood
bootstrap; PB, parsimony bootstrap; Br, total Bremer) equal to or above 80 (MLB), 80 (PB) and 3.0 (Br) (high) are indicated with a solid black
circle in the centre of the outer circle. Nodes with measures less than (low) the aforementioned values are without a central black circle. Branches
are coloured according to family group.
© 2010 The Authors

Table 2. Node support values calculated with maximum likelihood nonparametric bootstrap, parsimony nonparametric bootstrap, posterior probabilities from mixed model Bayesian analysis and
total/partitioned Bremer values for nodes recovered in both the maximum likelihood and parsimony topologies. Node numbers correspond to those presented in Fig. 1.
Maximum Total Bremer Bremer

likelihood Parsimony Posterior Bremer cytochrome cytochrome Bremer Bremer Bremer Bremer
Node no. bootstrap bootstrap probability support oxidase I oxidase II histone 3 wingless 18S rRNA 28S rRNA
1 97 95 100 23.0 12.5 0.0 −8.5 −6.5 11.0 14.5

2 100 100 100 36.0 −3.3 2.1 3.6 −8.1 7.2 34.5
3 <50 87 100 9.0 2.5 −0.5 −2.5 −0.5 4.0 6.0
4 <50 <50 65 1.0 0.0 0.0 0.0 0.0 0.0 1.0
5 <50 54 70 4.0 −1.0 0.0 0.9 0.0 2.1 2.0
6 <50 <50 92 2.0 3.5 −0.5 0.5 1.0 −1.0 −1.5
7 63 76 100 8.0 −3.0 3.1 0.3 −4.9 −1.1 13.6

8 51 <50 89 2.0 −0.4 0.0 1.5 −1.5 1.0 1.4
9 100 99 100 10.0 0.0 0.0 8.0 0.0 1.0 1.0
10 <50 <50 <50 2.0 0.0 0.0 1.0 0.0 0.0 1.0
11 83 80 100 3.0 0.0 0.0 0.0 0.0 0.0 3.0
12 97 92 100 4.0 0.0 0.0 −0.1 0.0 1.1 3.0
13 <50 59 100 5.0 −7.5 1.7 0.5 −5.3 0.0 15.6
14 100 100 100 21.0 −5.0 3.1 15.5 −5.2 0.9 11.7
15 100 100 100 50.0 −3.0 4.0 12.5 −2.0 8.5 30.0
16 71 90 98 4.0 0.0 0.0 3.5 0.0 0.5 0.0
17 88 58 100 2.0 0.0 0.0 1.7 0.0 0.1 0.2
18 68 <50 72 0.0 0.0 0.0 0.5 0.0 −0.5 0.0
19 80 63 95 2.0 0.0 0.0 1.8 0.0 0.2 0.0
20 <50 <50 DNS 3.0 −0.5 −0.5 7.0 −5.5 1.0 1.5
21 <50 <50 DNS 3.0 −1.7 1.6 6.5 −3.4 0.1 −0.2
22 100 100 100 78.0 3.3 10.7 23.2 −9.1 21.6 28.3
23 <50 <50 DNS 5.0 6.7 1.9 −11.6 −1.7 −1.4 11.1
24 100 100 100 46.0 6.6 −0.2 9.1 −1.0 11.4 20.0
25 82 <50 100 0.0 0.0 0.0 0.0 0.0 0.0 0.0
26 99 99 100 12.0 −1.0 0.0 2.0 0.0 2.0 9.0
27 99 100 100 7.0 2.0 0.0 0.0 0.0 0.0 5.0
28 88 68 66 1.0 1.0 0.0 0.0 0.0 0.0 0.0
29 <50 <50 <50 0.0 0.0 0.0 0.0 0.0 0.0 0.0
30 78 63 100 5.0 23.8 0.8 −42.9 −1.5 −5.0 29.8
31 100 100 100 26.0 7.0 −0.5 8.5 1.5 −1.0 10.5
32 100 100 100 13.0 2.0 0.0 8.0 4.0 0.0 −1.0
33 72 58 100 5.0 24.0 0.4 −42.9 −1.3 −4.4 29.1
34 80 67 100 5.0 23.8 0.8 −43.0 −1.6 −5.0 30.0
35 57 <50 86 0.0 0.0 0.0 0.0 0.0 0.0 0.0
36 93 70 100 4.0 0.0 0.0 2.0 0.0 0.0 2.0
37 100 100 100 3.0 0.0 0.0 −1.0 0.0 0.0 4.0
38 90 62 100 6.0 2.2 1.5 0.5 −0.2 0.5 1.5
39 51 92 96 16.0 5.0 9.0 0.0 0.0 1.0 1.0
40 <50 <50 69 1.0 1.0 −1.0 1.0 −1.0 0.0 1.0
© 2010 The Authors
Table 2. Continued.

41 100 100 100 18.0 1.0 9.0 3.0 7.0
© 2010 The Authors

−1.0 −1.0
42 <50 <50 99 4.0 −4.4 1.0 −0.2 −3.3 −1.0 12.0
43 100 100 100 13.0 2.0 −1.0 5.0 1.0 3.0 3.0
44 <50 <50 81 0.0 0.0 0.0 0.0 0.0 0.0 0.0
45 100 95 100 21.0 3.5 0.0 0.0 −0.8 7.5 10.8
46 100 100 100 21.0 −0.4 2.8 5.5 −8.2 0.0 21.3
47 <50 <50 76 0.0 0.0 0.0 0.0 0.0 0.0 0.0
48 91 93 100 9.0 0.7 3.3 1.0 −0.3 1.0 3.3
49 100 100 100 32.0 1.5 8.5 14.5 −1.0 3.0 5.5
50 100 100 100 17.0 1.0 0.0 1.0 0.0 9.0 6.0
51 <50 <50 99 5.0 23.8 0.8 −43.0 −1.6 −5.0 30.0
52 100 100 100 34.0 −4.5 5.7 9.4 −1.3 −2.3 27.0
53 75 75 100 8.0 −1.6 2.1 2.4 −9.8 −1.1 15.9
54 95 87 100 7.0 1.0 0.0 6.0 −1.0 0.0 1.0
55 70 85 88 5.0 4.0 1.0 0.0 −1.0 0.0 1.0
56 69 55 98 3.0 2.0 0.5 0.0 0.0 0.0 0.5
57 <50 <50 <50 0.0 0.0 0.0 0.0 0.0 0.0 0.0
58 <50 <50 <50 1.0 −5.4 1.7 0.5 −3.3 −2.0 9.6
59 100 100 100 20.0 −8.3 4.7 10.0 −10.9 −4.6 29.0
60 <50 <50 DNS 1.0 −5.4 1.7 0.4 −3.3 −2.0 9.7
61 <50 <50 DNS 0.0 0.0 0.0 0.0 0.0 0.0 0.0
62 <50 <50 DNS 0.0 0.0 0.0 0.0 0.0 0.0 0.0
63 100 100 100 16.0 −4.4 1.7 4.5 −3.3 0.0 17.6
64 53 <50 93 0.0 0.0 0.0 0.0 0.0 0.0 0.0
65 55 <50 51 0.0 0.0 0.0 0.0 0.0 0.0 0.0
66 100 99 100 11.0 −4.4 1.7 −0.6 −3.3 −2.0 19.7
67 83 96 100 11.0 8.0 0.0 5.0 −2.0 −1.0 1.0
68 92 75 100 2.0 2.0 1.0 −2.0 −1.0 0.0 2.0
69 100 100 100 18.0 9.0 0.0 0.0 4.0 2.0 3.0
70 65 77 86 1.0 3.0 0.0 0.0 0.0 0.0 −2.0
71 54 59 99 5.0 23.8 0.8 −43.0 −1.6 −5.0 30.0
72 77 <50 100 0.0 0.0 0.0 0.0 0.0 0.0 0.0
73 <50 <50 97 0.0 0.0 0.0 0.0 0.0 0.0 0.0
74 73 72 100 3.0 −5.5 1.0 3.5 −2.0 −2.0 8.0
75 <50 <50 DNS 2.0 −6.5 2.0 3.5 −2.0 −3.0 8.0
76 71 66 78 5.0 −2.0 17.0 0.0 −7.0 −2.0 −1.0
77 99 99 100 15.0 −0.4 2.6 11.9 −5.9 −2.3 9.1
78 100 100 100 18.0 4.0 0.0 5.0 7.0 1.0 1.0
79 100 100 100 23.9 −2.4 1.7 6.4 0.7 −2.0 19.6
80 <50 <50 76 0.0 0.0 0.0 0.0 0.0 0.0 0.0

405
Table 2. Continued.

81 86 73 100 0.0 0.0 0.0 0.0 0.0 0.0 0.0

82 71 <50 100 0.0 0.0 0.0 0.0 0.0 0.0 0.0
83 66 <50 100 0.0 0.0 0.0 0.0 0.0 0.0 0.0

84 <50 <50 100 0.0 0.0 0.0 0.0 0.0 0.0 0.0
85 100 100 100 12.0 3.0 5.0 1.0 3.0 −2.0 2.0
86 86 81 100 3.0 −4.7 2.0 2.7 −0.3 −2.0 5.3
87 <50 <50 85 0.0 0.0 0.0 0.0 0.0 0.0 0.0
88 <50 <50 DNS 0.0 0.0 0.0 0.0 0.0 0.0 0.0
89 100 100 100 17.0 −2.6 1.3 8.3 3.0 −2.0 9.0
90 <50 <50 98 0.0 0.0 0.0 0.0 0.0 0.0 0.0
91 89 88 100 8.0 −6.0 3.0 4.0 −2.3 −2.0 11.3
92 99 99 100 8.0 1.0 0.0 0.0 4.0 1.0 2.0
93 <50 66 81 3.0 −1.5 1.0 −2.0 −3.5 −1.0 10.0
94 97 98 100 9.0 0.0 0.0 −2.0 0.0 1.0 10.0
95 100 100 100 23.0 3.1 1.2 14.1 −8.9 0.9 12.7
96 77 61 100 2.0 −1.0 0.0 2.5 0.0 0.0 0.5
97 <50 <50 <50 0.0 0.0 0.0 0.0 0.0 0.0 0.0
98 69 <50 92 1.0 0.0 0.0 0.0 0.0 0.0 1.0
99 68 <50 100 1.0 −5.4 1.7 0.4 −3.3 −2.0 9.7
100 67 58 100 1.0 −5.8 1.6 −0.2 −3.2 −2.0 10.6
101 99 100 100 14.0 −2.5 1.6 3.9 −2.9 −1.0 14.9
102 94 98 100 5.0 −10.7 1.3 3.3 −1.7 −2.0 14.7
103 100 100 100 13.0 0.0 4.0 2.0 7.0 0.0 0.0
104 90 84 100 6.0 −3.0 1.0 −1.0 −2.0 −3.0 14.0
105 100 100 100 13.0 −0.4 2.1 7.2 −1.2 −0.8 6.1
106 78 69 95 2.0 0.0 0.0 1.7 0.0 −0.7 1.0
107 60 <50 90 1.0 −5.4 1.7 0.4 −3.3 −2.0 9.7
108 93 78 100 1.0 0.0 0.0 0.0 0.0 0.0 1.0
DNS, does not support.
© 2010 The Authors
to influence the early branching order of Machaerotidae and

Clastopteridae as either group was recovered as the first branch
within Cercopoidea (see Table S2). Final dataset analyses
demonstrated congruence between ML and parsimony for the
placement of Clastopteridae outside of Aphrophoridae as the
second branch within Cercopoidea (Fig. 1). Support for this
arrangement was low with bootstrap values less than 50 and a
Bremer value of just 3 for node 20 (Table 2).
Aphrophoridae, as defined currently, was nonmonophyletic
in all analyses; this nonmonophyly was due to two separate
Fig. 2. Summary topology from the mixed model Bayesian analysis topological placements: (i) the inclusion within Aphrophoridae
with posterior probabilities for represented nodes and terminal groups of the exemplars of Epipygidae (in both ML and parsimony
(in parentheses). Branches are coloured according to family group. analyses) and (ii) the exclusion of Microsargane vittata Fowler
(this species is classified within Aphrophoridae, but in these
were concentrated almost entirely within the Cercopidae analyses was placed consistently in Cercopidae at node 57;
(node 51; Fig. 1). Nonparametric bootstrap analysis recovered Fig. 1). As outlined above, the nonmonophyly of Aphrophori-
moderate support across the topology with 55 of the 108 nodes dae as reconstructed by MMB analysis resulted from the
(51%) exhibiting a value of 70 or greater (Table 2). As with inclusion of Clastopteridae in addition to Epipygidae within
ML bootstrap analysis, those nodes recovering relationships the aphrophorid lineages (Fig. 2). We recovered strong sup-
within genera or between two genera exhibited higher bootstrap port for the monophyly of Epipygidae (Fig. 1, node 41), but
values of 70 or greater for 64% of the nodes (43 of 67) weaker support for the nodes responsible for recovering Epipy-
(Table 2). Partitioned Bremer values were calculated for all gidae within Aphrophoridae (Fig. 1, nodes 40, 29, 23 and
nodes congruent between the ML and parsimony topologies 21; Table 2). Cercopidae was recovered as monophyletic in
(83 nodes) and are represented in Table 2. all analyses, given the inclusion of the previously mentioned
Microsargane vittata (Fig. 1; see Discussion). Within Cercop-
idae, we recovered moderate support for a monophyletic New
Testing traditional classification World subfamily Ischnorhininae (Fig. 1, node 71; Table 2).
Topological tests of the five previous hypotheses for
A monophyletic Cercopoidea was recovered in all analyses relationships within Cercopoidea resulted in the failure to reject
(ML, MMB and parsimony) with strong nodal support the null hypothesis (the constraint tree was not significantly
(node 1, Table 2). The two included ribosomal genes in different from the unconstrained tree) in all cases (Table S4).
combination with COI provided a strong phylogenetic signal Two of the tests returned constrained tree likelihood scores
for the recovery of Cercopoidea (node 1), but histone 3 and considerably less optimal than the unconstrained ML analysis,
wingless provided a contradictory signal (−8.5 and −6.5, but these results were still not enough to return a significant
respectively), whereas COII remained neutral (Table 2). A result. For example, excluding Epipygidae from its position
well-supported monophyletic Machaerotidae was recovered within the Aphrophoridae returned a tree with a score 13.66
in all analyses as the first branch within Cercopoidea, but greater than the unconstrained analysis, whereas constraining
weak to moderate support uniting the remaining cercopoid Cercopidae to be sister to aphrophorids, clastopterids and
groups together reflected that molecular data lacked strong machaerotids returned a tree likelihood 12.35 greater than the
character support for this arrangement (node 13, Table 2, unconstrained analysis.
Fig. 1). Both wingless and COI provided a contradictory
signal for the recovery of a monophyletic Machaerotidae, but
the signal from 28S, 100% bootstrap values and a 100% Divergence time estimation
posterior probability are conclusive evidence of monophyly
(node 2, Fig. 1, Table 2). Closely mimicking this same pattern, Divergence time estimation using the program beast
we recovered a monophyletic Clastopteridae in all analyses resulted in four independent runs converging on the same
with very high Bremer values, 100% bootstrap values and likelihood distribution (Run1: posterior mean −53574.1,
a 100% posterior probability (node 14, Table 2, Fig. 1). stdev 0.526; likelihood mean −53345.53, stdev 0.223;
However, the placement of Clastopteridae within Cercopoidea Run2: posterior mean −53573.7124, stdev 0.4504; likeli-
was less clear between analyses and preliminary dataset hood mean −53345.5548, stdev 0.2116; Run3: posterior mean
treatment. For example, MMB recovered Clastopteridae within −53575.222, stdev 0.7581; likelihood mean −53345.8407,
the Aphrophoridae (Fig. 2) with strong support. In addition, stdev 0.6217; Run4: posterior mean −53575.1045, stdev
the preliminary parsimony analyses all indicated that the 0.5521; likelihood mean −53345.7138, stdev 0.2363). A burn-
third codon positions of COI and COII heavily influenced in of 1 000 000 generations was applied and the maximum
the recovery of the group within Aphrophoridae (Table S2). clade stability tree (Fig. 3) was calculated using 29 000 trees
Furthermore, the alignments of the length variable regions of from Run2 (run with smallest standard deviation). The specific
the two ribosomal genes based on mafft or muscle appeared node height for each node represented in Fig. 3 is the mean of
© 2010 The Authors

all node heights in the distribution. A 95% confidence interval Microsargane belongs in the family Cercopidae (with subfa-
was calculated from the posterior distribution of node heights. milial affiliation undetermined) and, therefore, here we transfer
Microsargane to Cercopidae, incertae sedis.
Interestingly, in a phylogenetic analysis of a bacterial
Discussion
endosymbiont (Sulcia muelleri of phylum Bacteroidetes) of
The results of the maximum parsimony (MP) and ML Auchenorrhyncha, the DNA sequence generated from the
analyses (Fig. 1) agree concerning the higher phylogeny endosymbiont of Microsargane vittata (listed as Microsargane
of Cercopoidea; those analyses strongly support the mono- sp.) was related more closely to the endosymbiont of
phyly of Cercopoidea and of four major included lineages, Cercopidae than to the endosymbionts of Aphrophoridae
corresponding to the families Machaerotidae, Clastopteri- (Moran et al., 2005), providing corroborating evidence for our
dae, Aphrophoridae (including Epipygidae) and Cercopidae placement of Microsargane within Cercopidae.
(if redefined to include the genus Microsargane). The results
of the Bayesian analysis (Fig. 2) also support the monophyly
of Machaerotidae, Cercopidae (including Microsargane) and Machaerotidae
Clastopteridae; however, the latter was placed within a para-
The approximately 110 described species of Machaerotidae
phyletic Aphrophoridae. In all analyses, the four included
are distributed throughout much of the Palaeotropics (Metcalf,
exemplars of Epipygidae constituted a monophyletic lineage
1960; Maa, 1963); there are no New World species of
within Aphrophoridae (as in Fig. 1, node 41).
Machaerotidae. This family is diagnosed morphologically by
The classification of Cercopoidea remains problematic at
the following suite of character states: scutellum longer than
nearly all higher levels. Not only is there disagreement as
wide, deep antennal pits concealing the antennal bases and
to the number and constitution of spittlebug families, but the
nonoverlapping forewing apices (when at rest) (Dietrich,
subfamilial and tribal classifications are in taxonomic disarray
2005). Behaviourally, this family is unique within Cercopoidea
or abandoned entirely (as in the case of Ischnorhininae;
in that machaerotid nymphs construct calcified tubes affixed to
Carvalho & Webb, 2005). Although the present analyses
their host plants, and live within these tubes immersed in a
support several conclusions regarding the monophyly and
clear liquid excretion derived from ingested xylem sap; just
constituency of various higher groups of Cercopoidea, many
after hatching and before each nymphal moult, machaerotids
critical taxa are not represented here and therefore we regard
apparently produce small amounts of the spittle froth typical
the conclusions of this study as evidence for preliminary
of Cercopoidea (Evans, 1940; Maa, 1963; Marshall, 1964;
taxonomic recommendations only. Thus, with one exception,
Marshall & Marshall, 1966; Rakitov, 2002). This life-history
we refrain from making official changes to the classification
strategy differs considerably from those of the other cercopoid
of Cercopoidea here; the one exception is the placement of the
families, which vary with regard to size and consistency
genus Microsargane.
of produced spittle masses, but which do not produce such
tubes.
Microsargane: a family-level misclassification revealed Machaerotidae was recovered uniformly as the monophyletic
sister group to all other lineages of Cercopoidea (Fig. 1,
The genus Microsargane is currently classified within node 13; Fig. 2), in contrast to two other notable hypotheses
Aphrophoridae. Morphologically, species of Microsargane do on the evolutionary position of this spittlebug family. First,
present an aphrophorid gestalt, despite being brightly coloured Maa (1961) asserted, based on their restricted distribution
(a condition atypical among Aphrophoridae); the compound and lack of known fossils, that Machaerotidae is the most
eyes of Microsargane are similar in shape and position to derived lineage of spittlebugs. Later, Maa (1963) speculated
those of many Aphrophoridae (excluding Epipygidae, in which that Machaerotidae was derived from the family Cercopidae
the compound eyes touch the forewing bases), and the fron- based on the supposition that Cercopidae is the most ancient
toclypeus of Microsargane is similar in its convex shape to of extant spittlebug lineages, a supposition based, in turn, on
those of Aphrophoridae species. Nevertheless, Microsargane the observation that Cercopidae nymphs, like cicada nymphs,
vittata (the exemplar of Microsargane included in these anal- tend to feed on subterranean parts of their host plants (Evans,
yses) was recovered consistently within the family Cercopi- 1940; Maa, 1963). It should be noted that although some
dae (as in Fig. 1, node 57), although with only weak support Cercopidae do feed on plant parts at or just below the soil
(Fig. 1, nodes 51 and 57; Table 2). The taxonomic affinity of surface (Leite et al., 2005; Carvalho & Webb, 2005), nymphs
Microsargane within Cercopidae remains enigmatic, however, of many cercopid species feed on plant parts that are far
as Microsargane (a Neotropical genus) was recovered consis- above the soil surface (Thompson, 1997; Carvalho & Webb,
tently among the Old World cercopid clades. DNA sequence 2005; Cryan & Svenson, personal observations). Second,
data for the single specimen of Microsargane vittata included Hamilton (2001) hypothesized that Machaerotidae is included
here were incomplete (missing data from COI, COII and his- (as Machaerotinae) within a redefined family Clastopteridae,
tone 3) and, therefore, the exact placement of Microsargane based on the deep antennal bases that both groups apparently
within Cercopidae has yet to be tested rigorously. Neverthe- have, as well as the assertion that only a single hindwing vein
less, based on the results of these analyses, we conclude that differentiates the two groups taxonomically.
© 2010 The Authors

Fig. 3. Chronogram based on divergence time estimations calculated within beast. Confidence intervals are presented as shaded grey bars for each
node representing the upper and lower bound (95%).
Neither of these two evolutionary scenarios regarding Clastopteridae

Machaerotidae is supported by the genetic evidence-based
analyses presented here. Rather, the reconstruction of Machae- Species of Clastopteridae are typically the smallest spittle-
rotidae (a monophyletic clade, separate from Clastopteridae) as bugs (individuals of most clastopterid species are ovoid and
the most anciently divergent lineage within Cercopoidea repre- generally 2-4 mm in length); this family is morphologically
sents a novel and intriguing hypothesis. Divergence date anal- diagnosed by having the following suite of character states:
ysis (Fig. 3) estimated that the extant Machaerotidae diverged scutellum longer than wide, deep antennal pits concealing
around the Triassic–Jurassic boundary, between 197.3 and the antennal bases and broadly overlapping forewing apices
215.96 mya (95% confidence interval, mean = 206.57 mya); (when at rest) (Dietrich, 2005). Behaviourally, clastopterid
this estimation is somewhat older than the early tertiary origin nymphs produce small spittle masses consisting of tiny air
bubbles suspended in a relatively viscous fluid (Hamilton,
of Machaerotidae that Maa (1963) approximated based on the
2001). The approximately 80 described species of Clastopteri-
incomplete fossil record of Cercopoidea (indeed, no fossils of
dae are classified in two genera, Clastoptera and Iba. Species
Machaerotidae have been described to date).
of Clastoptera occur throughout the New World, whereas Iba
The 27 genera of Machaerotidae are taxonomically divided
(with only two described species) is apparently restricted to the
into the subfamilies Machaerotinae (characterized by an
Philippines (Metcalf, 1962b); specimens of Iba were unavail-
enlarged, spine-like scutellar process) and Enderleiniinae (lack-
able for inclusion in the present analyses. Therefore, the inclu-
ing the spine-like scutellar process) (Maa, 1963). Three species
sion of Iba within Clastopteridae could not be tested in this
of Machaerotinae, representing the genera Machaerota and study.
Grypomachaerota, were included in these analyses. Although In his redefinition of the higher taxa of Cercopoidea,
morphological evidence (notably, the scutellar spine) supports Hamilton (2001) considered Clastoptera and Iba as con-
the monophyly of Machaerotinae, this group was not recovered stituents of the subfamily Clastopterinae within an expanded
in these analyses. We suspect that taxonomic and data sam- family Clastopteridae, which also included the subfamilies
pling bias prevented the monophyly of Machaerotinae from Machaerotinae (see above) and several genera included pre-
being supported here. The tribe Enderleiniini was recovered viously in the family Aphrophoridae. Although important
as monophyletic except for the exclusion of Apomachaerota as putatively ‘intermediate’ groups, those aphrophorid gen-
reticulata Schmidt, which Maa (1963) discussed as having a era (Abbaloma, Beesoniella, Grellaphia, Nyanja, Patriziana,
number of characters that differentiate this monotypic genus Pseudomachaerota, Sepulia and Tremapterus; Hamilton, 2001)
from the rest of the tribe; thus, the phylogenetic placement were unfortunately unavailable for inclusion in these analyses,
of Apomachaerota as excluded from Enderleiniini is perhaps and so their relationships with Clastopteridae were untested.
unsurprising. The phylogeny of the family Machaerotinae will However, the included exemplars of Clastoptera were recov-
be explored in more detail in a subsequent analysis. ered consistently as a strongly supported monophyletic clade
© 2010 The Authors

(as in Fig. 1, node 14), that was never united in a single lineage tribes. Species of Aphrophoridae occur worldwide (Metcalf,
with Machaerotidae, providing evidence that Clastopteridae, as 1962a) and in most terrestrial habitats. This family is
defined by Metcalf (1962b), is a natural clade. diagnosed morphologically by having the following suite of
The placement of Clastopteridae was somewhat vari- character states: scutellum about as long as wide, relatively
able, however, depending on analytical permutations. MP shallow antennal pits not concealing the antennal bases,
and ML analyses placed Clastopteridae as the sister group oblong compound eyes not touching forewing bases and a
to Aphrophoridae + Cercopidae (as in Fig. 1, nodes 13 convex frontoclypeus (Dietrich, 2005). Behaviourally, nymphs
and 20), a divergence estimated as having occurred dur- typically produce conspicuous spittle masses on exposed
ing the mid-Jurassic, between 145.39 and 202.58 mya (95% portions of the host plants (Hamilton, 1982; Liang & Fletcher,
confidence interval, mean = 174.74 mya) (Fig. 3). This phy- 2003; Ribeiro et al., 2005); although nymphs of some species
logenetic placement seems consistent with trends in certain live solitarily (that is, each larva produces and develops
morphological characters, such as the deep antennal pits that in its own spittle mass), a number of aphrophorid species
hide the antennal bases, also present in Machaerotidae (Hamil- develop gregariously as immatures (Bales & Furniss, 1984;
ton, 2001). The Bayesian analysis (Fig. 2), however, placed Biedermann, 2003; Wise et al., 2006). In the extreme, species
Clastopteridae within a paraphyletic Aphrophoridae (which in the Afrotropical genus Ptyelus and the Neotropical genus
also includes Epipygidae); this placement seems less likely Cephisus are well known for this nymphal group-living
from a morphological standpoint, as synapomorphies exclu- habit, in which each shared spittle mass might harbour
sively uniting these groups are, as yet, unknown. from tens to hundreds of individuals (Wagner et al., 1991;
Tsacas & Couturier, 1993; Cryan & Svenson, personal
observation). Adult Aphrophoridae tend towards subdued,
Epipygidae rather cryptic coloration, and several species [most notably
Philaenus spumarius Linnaeus, Ptyelus goudoti (Bennett) and
Epipygidae is the smallest of described cercopoid families in
P. grossus (Fabricius)] exhibit remarkable intraspecific colour
terms of biodiversity; Hamilton (2001) erected Epipygidae to
polymorphism (Synave, 1954; Thompson & Halkka, 1973;
include four described and 27 undescribed species. Diagnostic
Thompson, 1984), apparently due to the occurrence of several
features given for this group include the position of the
alleles at a single locus (Halkka et al., 1973, 1975).
compound eyes (almost touching the forewing base and
In describing his hypothesis of aphrophorid polyphyly,
concealing the prothorax laterally) and unusual, elytriform
Hamilton (2001) surmised that Aphrophoridae represents a
or ‘crumpled-looking’ forewings that are either punctate
‘miscellaneous assembly of genera’ based on his interpreta-
and reticulate or glossy and highly sculptured (Hamilton,
tions of various morphological characters, including foreleg
2001). Although the biology of epipygid species is unknown,
articulation and wing folding. In a new classification scheme,
Hamilton (2001) speculated that epipygid nymphs may not be
some aphrophorid genera were recommended to a redefined
competent spittle mass producers and that epipygid adults may
Clastopteridae (see above), a few aphrophorid species were
either be nonfeeding or infrequent feeders.
included in the new family Epipygidae and most Aphrophori-
Four species of Epipygidae were included in these analyses,
dae were included in Cercopidae, presumably as the subfamily
representing three genera (one of which is undescribed). In
Aphrophorinae (Hamilton, 2001).
all analyses, the epipygids were recovered as a strongly
Exemplars of 28 species (representing 17 genera in six
supported, monophyletic clade (as in Fig. 1, node 41);
tribes) currently classified within Aphrophoridae were included
however, this clade was always placed within the family
in these analyses; the results of MP and ML analyses
Aphrophoridae. Morphologically, there seems to be little
consistently recovered 27 of these in a monophyletic clade
to separate these groups; in Dietrich’s (2005) key to the
(Fig. 1, node 21; only Microsargane vittata was excluded
families of Cicadomorpha, Epipygidae and Aphrophoridae
from this lineage; see above) that also included Epipygidae
were distinguished based on the shape of the frontoclypeus
(see above). We take these results as a strong indication that
(flattened or laterally concave in the former, convex in the
Aphrophoridae + Epipygidae constitutes a distinct lineage;
latter) and the position of the compound eyes (touching the
divergence analyses estimated that this lineage originated
forewing base in Epipygidae and not reaching the forewing
around the beginning of the Cretaceous, between 122.78 and
base in Aphrophoridae). Hamilton’s (2001) observation of the
atypical epipygid forewings was not used by Dietrich (2005) 168.34 mya (95% confidence interval, mean = 145.38 mya)
to separate Epipygidae from Aphrophoridae. On the basis of (Fig. 3), somewhat earlier than Shcherbakov’s (1996, 2002)
our analytical results, we conclude that Epipygidae represents Aphrophoridae divergence estimate of 100 mya based on the
a monophyletic lineage within Aphrophoridae, and probably fossil record. The Bayesian analysis (Fig. 2) also recovered the
deserves either a tribal or subfamilial designation. Aphrophoridae + Epipygidae clade; however, Clastopteridae
was also included in this group (but see above).
At the tribal level, the included exemplars of Philaenini
Aphrophoridae were recovered in a monophyletic clade (Fig. 1, node 24),
and Aphrophorini was recovered as monophyletic except for
Metcalf’s (1962a) catalogue of the family Aphrophoridae the exclusion of Microsargane (see above). However, our
listed 820 species distributed among 151 genera in nine results suggest that Metcalf’s (1962a) classification (as listed
© 2010 The Authors

in Table 1) for other aphrophorid tribes is not supported by and the number of post-tibial spines. In a more recent review of
these data. For example, Cloviini was grossly polyphyletic, New World Cercopidae, Carvalho & Webb (2005) concurred
recovered in no less than four lineages [Fig. 1, nodes 22, with regard to the monophyly of New World Cercopidae, but
31, 34 (Plinia) and 45 (Sphodroscarta)]; likewise, exemplars reverted to the earlier name for the group, Ischnorhininae.
of Lepyroniini were recovered in three separate lineages Both studies discussed that the New World cercopids are char-
[Fig. 1, nodes 39, 45 (Avernus) and 46 (Peuceptyelus)]. acterized by (among other features) the complete fusion of
Ptyelini was polyphyletic due to the exclusion of Cephisus, the subgenital plates to the pygofer, whereas the subgenital
the only New World member of the tribe sampled here. plates of Old World Cercopidae are either partially fused (as
Clearly, the definition and constituency of aphrophorid tribes in Cosmoscartinae) or entirely free (Fennah, 1968; Carvalho
must be re-evaluated based on a more comprehensive analysis & Webb, 2005).
of Aphrophoridae, a family that, itself, needs more stable, Although not articulated in either study, the descriptions
synapomorphy-based, definition. of the degrees of fusion of the subgenital plate suggest not
only the derived position of Ischnorhininae with respect to Old
Cercopidae World cercopids, but also a potential lack of synapomorphic
support for the monophyly of Old World Cercopidae (as
Cercopidae is the largest of the extant spittlebug families Cercopinae). Indeed, the results of the present analyses support
in terms of both biodiversity and typical body size. Metcalf both of these claims: (i) the monophyly of the New World
(1961) catalogued nearly 1400 species in the family Cercopi- subfamily Ischnorhininae (sensu Carvalho & Webb, 2005) was
dae, classified into two subfamilies, 17 tribes and 142 genera. recovered with moderate to strong support in all analyses;
The actual number of species in this family is much higher, and (ii) the Old World Cercopidae (Cercopinae, sensu Fennah,
however; for example, Metcalf (1961) recorded 294 species 1968) was recovered as a paraphyletic group (as in Fig. 1, node
of Cercopidae as occurring in the New World alone; but in a 58) from which Ischnorhininae was derived. Divergence date
more recent checklist, Carvalho & Webb (2005) documented analysis estimated the origin of Ischnorhininae during the mid-
that number as 475. Cretaceous, between 91.06 and 134.02 mya (95% confidence
Adults of many cercopid species have bright coloration in interval, mean = 112.18 mya) (Fig. 3).
conspicuous patterns on the forewings, pronotum, head and/or Carvalho & Webb (2005) did not adopt a tribal classification
legs. As in some other spittlebugs (see above), intraspecific within the New World subfamily Ischnorhininae, as many of
polymorphism with regard to coloration and patterning has the tribes recognized previously by Metcalf (1961) included
been documented in several species of Cercopidae (Carvalho & taxa occurring in both Old and New World regions and many
Webb, 2005; Thompson & León González, 2005). This family of the characters used by Fennah (1968) to delimit tribes
is diagnosed morphologically by having the following suite of of New World spittlebugs are neither consistent nor reliable.
character states: scutellum about as long as wide, relatively Exemplars of two of Fennah’s (1968) New World tribes
shallow antennal pits not concealing the antennal bases were included here; except for one species of Laccogrypota,
and globular compound eyes (Dietrich, 2005). Behaviourally, members of Fennah’s Ischnorhinini (here represented by the
nymphs of Cercopidae tend to feed on herbaceous monocots genera Homalogrypota, Laccogrypota and Baetkia) formed a
(often grasses) that exhibit associative nitrogen fixation via monophyletic lineage (Fig. 1, node 77). However, these results
root zone bacteria (Thompson, 2004), and typically produce suggest that the tribe Tomaspidini is paraphyletic with respect
conspicuous, individual spittle masses. to Ischnorhinini.
The 58 species of Cercopidae (representing both Old and Although the taxonomic sample included here does not allow
New World taxa) included in these analyses were recov- for comprehensive examination of Old World Cercopidae tribal
ered in a monophyletic lineage (Fig. 1, node 51; along with structure, some trends are emerging. For example, Liang &
the aphrophorid species Microsargane vittata; see above). Webb (2002) noted that members of the tribe Rhinaulacini have
Although bootstrap support (Table 2) for Cercopidae was coeloconic sensilla on the antennae similar to those observed
weak, the posterior probability and Bremer values demon- in Locrisini by Boulard & Boulard (1979). Corroborating that
strated a strong signal (and therefore support) for the mono- observation, results of the present analyses placed members of
phyly of Cercopidae including Microsargane vittata. This clade Rhinaulacini and Locrisini together in a single lineage (Fig. 1,
is estimated to have diverged around the beginning of the node 59). However, Rhinaulacini is paraphyletic with respect
Cretaceous, between 122.78 and 168.34 mya (95% confidence to a monophyletic Locrisini. Additionally, the Old World tribe
interval, mean = 145.38 mya) (Fig. 3), somewhat earlier than Cosmoscartini was recovered as paraphyletic with respect to
Shcherbakov’s (1996, 2002) Cercopidae divergence estimate Euryaulacini and Suracartini. These results, when considered
of 100 mya based on the fossil record. together, suggest that the generic constituency of the Old World
Fennah (1968) recognized two subfamilies (the Old World tribes needs to be examined in greater detail.
Cercopinae and the New World Tomaspidinae) within Cer-
copidae that differed in composition and definition from the The history of spittlebug diversification
two subfamilies catalogued by Metcalf (1961). To support his
new subfamilial classification, Fennah cited several morpho- Shcherbakov (1996, 2002) estimated, based on fossil record
logical features, including male and female genitalic features evidence, that Cercopoidea originated during the early Jurassic
© 2010 The Authors

(approximately 205 mya); that date was partially used to cal- current classification, especially at the subfamilial and tribal
ibrate the divergence date analysis in this study, but was left levels, does not accurately reflect the evolutionary history of
to vary to reflect the rationale that Cercopoidea had already the spittlebugs.
diversified before its first appearance in the fossil record.
The divergence date estimates generated here (Fig. 3) indicate
that Cercopoidea originated during the Early Triassic, approx- Supporting Information
imately 243.55 ± 15.96 mya (older than Shcherbakov’s date
estimate, even given the statistical margin of error). During Additional Supporting Information may be found in the
that time period, Pangaea was an intact supercontinent and online version of this article under the DOI reference: DOI:
the mean global temperatures had risen to almost 22◦ C from 10.1111/j.1365-3113.2009.00520.x
12–15◦ C of the Permian (Grimaldi & Engel, 2005). Floristic
changes are evident from fossil deposits, with ‘new’ radiations
Table S1. Oligonucleotide primer sequences.
of a variety of plant groups, including the angiosperm-like Table S2. Preliminary parsimony dataset treatments.
Bennettitales (Grimaldi & Engel, 2005). Given the close asso-
Table S3. Descriptive statistics for data partitions.
ciation of Auchenorrhyncha with plants, the hypothesis that
large-scale floristic changes might affect diversification pat- Table S4. Topological tests.
terns of these phytophagous insects is unsurprising. Please note: Neither the Editors nor Wiley-Blackwell
The results of this study indicate that the family Machaeroti- are responsible for the content or functionality of any
dae diverged during the Triassic–Jurassic boundary, approxi-
supporting materials supplied by the authors. Any queries
mately 206.57 ± 9.27 mya, while Pangaea remained intact. The
(other than missing material) should be directed to the
family Clastopteridae diverged during the Jurassic, approxi-
corresponding author for the article.
mately 174.74 ± 29.35 mya; it was during this period of time
that Pangaea separated into Laurasia and Gondwana (com-
pleted by about 155 mya). The African/Australasian distribu-
tion of extant machaerotids and almost exclusively New World Acknowledgements
distribution of clastopterids are somewhat difficult to explain
given such ancient origins with continuous or contiguous Juras- We thank D. Peck, K. G. A. Hamilton and especially V.
sic landmasses. Perhaps local/regional extinctions and lineage Thompson for their valuable advice and encouragement for this
sorting might explain these patterns, but the causes of these project. For assistance with, and exceptional companionship
extinctions are only speculative. during, fieldwork, we thank M. Adams, C. Bartlett, D. Kondo,
The early Cretaceous (approximately 145.38 ± 22.6 mya) T. McCabe, K. Miller, G. Morse, J. Robertson, H. Romack,
F. Shockley, V. Thompson and J. Urban. For generously
origins of Aphrophoridae and Cercopidae apparently coincided
providing specimens used in this analysis, we are grateful to C.
with the origin and radiation of the angiosperms and further
Bartlett, C. Dietrich, S. Drosopoulos, K. Hill, K. Morishima,
increases in mean global temperatures (Grimaldi & Engel,
M. Moulds, N. Nazdrowicz, R. Rakitov, C. Simon, D. Takiya,
2005). Given the predominantly tropical distribution of extant
D. Tallamy, V. Thompson, M. Whiting, D. Yanega, J. Zahniser
spittlebug diversity, it seems probable that these diversification
and the Costa Rica ALAS Project Coordinators (R. Colwell, J.
events occurred on Gondwana after its separation from
Longino, H. Hespenheide). We deeply appreciate G. Carvalho,
Laurasia and the modern distribution of spittlebugs is probably
M. Fletcher, K. G. A. Hamilton, H.-T. Shih and V. Thompson
explained, at least in part, by vicariance events as Gondwana
for their kind assistance with specimen identification. We
fragmented further. Certainly, the estimated time of divergence
thank V. Thompson and J. Urban for providing insightful
of the New World Cercopidae (subfamily Ischnorhininae)
review comments on a manuscript draft of this paper. This
during the mid-Cretaceous, approximately 112.179 ± 21.12 material is based upon work supported by the National Science
mya, seems to coincide with the separation of South America Foundation, under grants DEB-0813897 and DEB-0529679,
from Africa, which would provide a defensible explanation for and by the New York State Museum. Any opinions, findings
the monophyly of Ischnorhininae. and conclusions or recommendations expressed in this material
are those of the authors and do not necessarily reflect the views
of the National Science Foundation or the New York State
Phylogenetic research needs in Cercopoidea Museum.
Many phylogenetic and classificatory questions regarding

Cercopoidea remain unanswered. More detailed investigations
with expanded taxonomic sampling are needed to address out- References
standing questions pertaining to generic and tribal relationships
Adis, J. & Schubert, H.O.R. (1984) Ecological research on arthropods
within each major spittlebug lineage. Once the phylogeny of in central Amazonian forest ecosystems with recommendation for
Cercopoidea is reliably reconstructed at these lower levels, then study procedures. Trends in Ecological Research for the 1980s (ed.
the classification of Cercopoidea will undoubtedly need a care- by J. H. Cooley and F. B. Golley), pp. 111–114. Plenum Press,
ful, phylogeny-based re-evaluation. It seems evident that the New York, New York.
© 2010 The Authors

Arenz, C.L. & Joern, A. (1996) Prairie legacies – invertebrates. Edgar, R.C. (2004) MUSCLE: multiple sequence alignment with
Prairie Conservation: Preserving North America’s Most Endangered high accuracy and high throughput. Nucleic Acids Research, 32,
Ecosystem (ed. by F. B. Sanson and F. L. Knopf), pp. 91–110. 1792–1797.
Island Press, Washington, DC. Evans, J.W. (1940) Tube-building cercopids (Homoptera, Machaeroti-
Bales, F.M. & Furniss, M.M. (1984) Bionomics of the cone spittlebug, dae). Transactions of the Royal Society of South Australia, 64,
Aphrophora canadensis (Homoptera: Cercopidae) on mugo pine in 70–75.
Idaho. Great Basin Naturalist, 44, 338–348. Felsenstein, J. (1985) Confidence limits on phylogenies: an approach
Bekker-Migdisova, E.E. (1962) Superorder Rhynchota. Insects with using the bootstrap. Evolution, 39, 783–791.
proboscis. Fundamentals of Paleontology, Vol. 9, Arthropoda, Fennah, R.G. (1968) Revisionary notes on the New World genera
Tracheata, Chelicerata (ed. by B. B. Rohdendorf). Akademiya of cercopid froghoppers (Homoptera: Cercopoidea). Bulletin of
Nauk SSSR, Moscow [in Russian; English translation 1991, Entomological Research, 58, 165–190.
Smithsonian Institution, Washington, DC]. Genecodes (2006) Sequencher Version 4.6. Genecodes Co., Ann Arbor,
Biedermann, R. (2003) Aggregation and survival of Neophilaenus Michigan.
albipennis (Hemiptera: Cercopidae) spittlebug nymphs. European Goloboff, P.A. (1999) Analyzing large data sets in reasonable times:
Journal of Entomology, 100, 493–499. solutions for composite optima. Cladistics, 15, 415–428.
Boulard, M. & Boulard, J. (1979) Un organe énigmatique chez les Goloboff, P.A., Farris, J.S. & Nixon, K. (2003) TNT: Tree Analysis
femelles de Locris (Homoptera Cercopidae). Annales de la Societé Using New Technology. Version 1.0, Beta test v. 0.2. Instituto Miguel
Entomologique de France (N.S.), 15, 513–523. Lillo, San Miguel de Tucumán, Argentina.
Carr, D.E. & Eubanks, M.D. (2002) Inbreeding alters resistance Grimaldi, D. & Engel, M.S. (2005) Evolution of the Insects. Cam-
to insect herbivory and host plant quality in Mimulus guttatus bridge University Press, Cambridge.
(Scrophulariaceae). Evolution, 56, 22–30. Halkka, O., Halkka, L., Raatikainen, M. & Hovinen, R. (1973) The
Carson, W.P. & Root, R.B. (1999) Top-down effects of insect genetic basis of balanced polymorphism in Philaenus (Homoptera).
herbivores during early succession: influence on biomass and plant Hereditas, 74, 69–80.
dominance. Oecologia, 121, 260–272. Halkka, O., Halkka, L., Raatikainen, M., Hovinen, R. & Vasarainen,
Carvalho, G.S. & Webb, M.D. (2005) Cercopid Spittlebugs of the New A. (1975) Genetics of Philaenus colour polymorphism: the 28
World (Hemiptera, Auchenorrhyncha, Cercopidae). Pensoft, Sofia. genotypes. Hereditas, 79, 308–310.
Clark, W.E., Diaz, G.I. & van Cleave, H.W. (1980) Taxonomy and Hamback, P.A. (2001) Direct and indirect effects of herbivory: feeding
biology of spittlebugs of the genera Aeneolamia Fennah and by spittlebugs affects pollinator visitation rates and seedset of
Prosapia Fennah (Cercopidae) in Northeastern Mexico. Folia Rudbeckia hirta. Ecoscience, 8, 45–50.
Entomologica Mexicana, 34, 13–24. Hamilton, K.G.A. (1982) The Insects and Arachnids of Canada.
Cronin, J.T. & Abrahamson, W.G. (1999) Host-plant genotype and Part 10: The Spittlebugs of Canada. Homoptera: Cercopidae.
other herbivores influence goldenrod stem galler preference and Agricultural Canada Publishers Number 1740.
performance. Oecologia, 121, 392–404. Hamilton, K.G.A. (2001) A new family of froghoppers from the Amer-
Cryan, J.R. (2005) Molecular phylogeny of Cicadomorpha (Insecta: ican tropics (Hemiptera: Cercopoidea: Epipygidae). Biodiversity, 2,
Hemiptera: Cicadoidea, Cercopoidea, and Membracoidea): adding 15–21.
evidence to the controversy. Systematic Entomology, 30, 563–574. Hamilton, K.G.A. (2005) Foreword. Cercopid Spittlebugs of the New
Cryan, J.R., Liebherr, J.K., Fetzner, J.W. Jr & Whiting, M.F. (2001) World (Hemiptera, Auchenorrhyncha, Cercopidae) (by G. S. Carvalho
Evaluation of relationships within the endemic Hawaiian Platynini and M. D. Webb), pp. 6–7. Pensoft, Sofia.
(Coleoptera: Carabidae) based on molecular and morphological Hanna, M. (1966) The spittlebugs of Michigan (Homoptera: Cercopi-
evidence. Molecular Phylogenetics and Evolution, 21, 72–85. dae). Papers of the Michigan Academy of Science, Arts, and Letters,
Cryan, J.R., Wiegmann, B.M., Deitz, L.L. & Dietrich, C.H. (2000) 51, 39–73.
Phylogeny of the treehoppers (Insecta: Hemiptera: Membracidae): Heinrichs, E.A. & Barrion, A.T. (2004) Rice-feeding Insects and
evidence from two nuclear genes. Molecular Phylogenetics and Selected Natural Enemies in West Africa. Biology, Ecology, Iden-
Evolution, 17, 317–334. tification. International Rice Research Institute, Los Banos (Philip-
Dietrich, C.H. (2002) Evolution of Cicadomorpha (Insecta, Hemi- pines), WARDA - The Africa Rice Center, Abidjan (Cote d’Ivoire).
ptera). Denisia, 4, 155–170. Holmann, F. & Peck, D.C. (2002) Economic damage caused by spit-
Dietrich, C.H. (2005) Keys to the families of Cicadomorpha and sub- tlebugs (Homoptera: Cercopidae) in Colombia: a first approximation
families and tribes of Cicadellidae (Hemiptera: Auchenorrhyncha). of impact on animal production in Brachiaria decumbens pastures.
Florida Entomologist, 88, 502–517. Neotropical Entomology, 31, 275–284.
Drummond, A.J. & Rambaut, A. (2007) BEAST: Bayesian evolution- Holzinger, W.E., Kammerlander, I. & Nickel, H. (2003) The Auchen-
ary analysis by sampling trees. BMC Evolutionary Biology, 7, 214. orrhyncha of Central Europe. Vol. 1: Fulgoromorpha, Cicadomorpha
Drummond, A.J., Nicholls, G.K., Rodrigo, A.G. & Solomon, W. excl. Cicadellidae. Brill, Leiden.
(2002) Estimating mutation parameters, population history and Huelsenbeck, J.P. & Imennov, N.S. (2002) Geographic origin of
genealogy simultaneously from temporally spaced sequence data. human mitochondrial DNA: accommodating phylogenetic uncer-
Genetics, 161, 1307–1320. tainty and model comparison. Systematic Biology, 51, 673–688.
Drummond, A.J., Rambaut, A., Shapiro, B. & Pybus, O.G. (2005) Huelsenbeck, J.P., Larget, B., Miller, R.E. & Ronquist, F. (2002)
Bayesian coalescent inference of past population dynamics from Potential applications and pitfalls of Bayesian inference phylogeny.
molecular sequences. Molecular Biology and Evolution, 22, 1185–1192. Systematic Biology, 51, 673–688.
Drummond, A.J., Ho, S.Y.W., Phillips, M.J. & Rambaut, A. (2006) Ivey, C.T., Carr, D.E. & Eubanks, M.D. (2004) Effects of inbreeding
Relaxed phylogenetics and dating with confidence. PLoS Biology, in Mimulus guttatus on tolerance to herbivory in natural environ-
4, e88. ments. Ecology, 85, 567–574.
Eden-Green, S.J., Balfas, R., Sutarjo, T. & Jamalius. (1992) Charac- Katoh, K., Misawa, K., Kuma, K. & Miyata, T. (2002) MAFFT: a
teristics of the transmission of Sumatra disease of cloves by tube- novel method for rapid multiple sequence alignment based on fast
building cercopoids, Hindola spp. Plant Pathology, 41, 702–712. Fourier transform. Nucleic Acids Research, 30, 3059–3066.
© 2010 The Authors

Katoh, K., Kuma, K., Toh, H. & Miyata, T. (2005) MAFFT version 5: Importance (ed. by M. R. Wilson and L. R. Nault), pp. 131–149.
improvement in accuracy of multiple sequence alignment. Nucleic CIE, London.
Acids Research, 33, 511–518. Nixon, K.C. (1999) The parsimony ratchet, a new method for rapid
Kosztarab, M., O’Brien, L.B., Stoetzel, M.B., Deitz, L.L. & Frey- parsimony analysis. Cladistics, 15, 407–414.
tag, P.H. (1990) Problems and needs in the study of Homoptera Pacheco, J.M., Silva, C.R.S., Ruvolo, M.C.C. & Schiavone, C. (1984)
in North America. Systematics of the North American Insects Biologia das cigarrinhas das pastagens (Homoptera, Cercopidae):
and Arachnids: Status and Needs. Virginia Agricultural Experi- ninfas de Deois flavopicta. Resumos Congresso Brasileiro de
ment Station Information Series 90-1 (ed. by M. Kosztarab and Entomologia, 9, 43.
C. W. Schaefer), pp. 119–145. Virginia Polytechnic Institute and Panzer, R., Sillwaugh, D., Gnaedinger, R. & Derkovitz, G. (1995)
State University, Blacksburg, Virginia. Prevalence of remnant dependence among the prairie and savanna-
Leite, L.G., Machado, L.A., Goulart, R.M., Tavares, F.M. & Filho, inhabiting insects of the Chicago region. Natural Areas Journal, 15,
A.B. (2005) Screening of entomopathogenic nematodes (Nemata: 101–116.
Rhabditida) and the efficiency of Heterorhabditis sp. against the sug- Parman, V.R. & Wilson, M.C. (1982) Alfalfa crop responses to
arcane root spittlebug Mahanarva fimbriolata (Fabr.) (Hemiptera: feeding by the meadow spittlebug (Homoptera: Cercopidae). Journal
Cercopidae). Neotropical Entomology, 34, 785–790. of Economic Entomology, 75, 481–486.
Liang, A-P. & Fletcher, M.J. (2003) Review of the Australian Peck, D., Castro, U., López, F., Morales, A. & Rodríguez, J. (2001)
aphrophorid spittlebugs (Hemiptera: Aphrophoridae). Australian First records of the sugar cane and forage grass pest, Prosapia
Journal of Entomology, 42, 84–93. simulans (Homoptera: Cercopidae), from South America. Florida
Liang, A-P. & Webb, M.D. (2002) New taxa and revisionary notes in Entomologist, 84, 402–409.
Rhinaulacini spittlebugs from southern Asia (Homoptera: Cercopi- Pires, C.S.S., Price, P.W. & de Oliveira, R.C. (2000) Distribution of
dae). Journal of Natural History, 36, 729–756. the spittlebug Deois flavopicta Stål (Homoptera: Cercopidae) on
Maa, T.C. (1961) Remarks on the distribution of the Machaeroti- wild and cultivated host species. Anais da Sociedade Entomológica
dae (Hemiptera: Cercopoidea). Pacific Insects Monograph, 2, do Brasil, 29, 401–412.
95–100. Posada, D. & Crandall, K.A. (1998) Modeltest: testing the model of
Maa, T.C. (1963) A review of the Machaerotidae (Hemiptera: DNA substitution. Bioinformatics, 14, 817–818.
Cercopoidea). Pacific Insects Monograph, 5, 1–166. Rakitov, R.A. (2002) Structure and function of the Malpighian tubules,
Maddison, D.R. & Maddison, W.P. (2003) MacClade. Version 4.06. and related behaviors in juvenile cicadas: evidence of homology with
Sinauer Associates, Sunderland, Massachusetts. spittlebugs (Hemiptera: Cicadoidea and Cercopoidea). Zoologischer
Marshall, A.T. (1964) Spittle-production and tube-building by cer- Anzeiger, 241, 117–130.
copoid nymphs (Homoptera). 2. The cytology and function of the Rambaut, A. (2007) FigTree version 1.2.2 [WWW document]. URL
granule zone of the Malpighian tubules of tube-building nymphs. http://tree.bio.ed.ac.uk/software/figtree/ [accessed in 2009].
Quarterly Journal of Microscopical Science, 105, 415–422. Rambaut, A. & Drummond, A.J. (2007) Tracer v1.4.1 [WWW doc-
Marshall, A.T. & Marshall, P.M. (1966) The life-history of a ument]. URL http://tree.bio.ed.ac.uk/software/tracer/ [accessed in
tube-dwelling Cercopoid: Machaerota coronata Maa (Homotera: 2009].
Machaerotidae). Proceedings of the Royal Entomological Society of Redak, R.A., Purcell, A.H., Lopes, J.R.S., Blua, M.J., Mizell, R.F.
London, Series A: General Entomology, 41, 17–20. III & Andersen, P.C. (2004) The biology of xylem fluid-feeding
Metcalf, Z.P. (1960) General Catalogue of the Homoptera. Fascicle insect vectors of Xylella fastidiosa and their relation to disease
VII Cercopoidea. Part 1 Machaerotidae. North Carolina State epidemiology. Annual Review of Entomology, 49, 243–270.
University, Raleigh, North Carolina. Ribeiro, G.T., Mendonça, M. da C., de Mesquita, J.B., Zanuncio, J.C.
Metcalf, Z.P. (1961) General Catalogue of the Homoptera. Fascicle & Carvalho, G.S. (2005) Spittlebug Cephisus siccifolius damaging
VII Cercopoidea. Part 2 Cercopidae. North Carolina State Univer- eucalypt plants in the State of Bahia, Brazil. Pesquisa Agropecuária
sity, Raleigh, North Carolina. Brasileira, 40, 723–726.
Metcalf, Z.P. (1962a) General Catalogue of the Homoptera. Fascicle Rodman, D.H. & Miller, D.J. (1992) Enzyme activities associated with
VII Cercopoidea. Part 3 Aphrophoridae. North Carolina State salivary glands of the froghopper Eoscarta carnifex (F) (Homoptera,
University, Raleigh, North Carolina. Cercopidae): possible role of salivary catalase in phytotoxicity.
Metcalf, Z.P. (1962b) General Catalogue of the Homoptera. Fascicle Australian Journal of Zoology, 40, 365–370.
VII Cercopoidea. Part 4 Clastopteridae, pp. 59. North Carolina State Ronquist, F. & Huelsenbeck, J.P. (2003) MrBayes 3: Bayesian phy-
University, Raleigh, North Carolina. logenetic inference under mixed models. Bioinformatics, 19,
Meyer, G.A. (1993) Comparison of the impacts of leaf- and sap- 1572–1574.
feeding insects on growth and allocation of goldenrod. Ecology, Sanz, N.T. (1997) Checklist of Pests and Diseases of Selected Crops
74, 1101–1106. of Belize. Belize National Plant Protection Service, Ministry of
Meyer, G.A. & Root, R.B. (1993) Effects of herbivorous insects and Agriculture and Fisheries, Central Farm, Cayo District.
soil fertility on reproduction of goldenrod. Ecology, 74, 1117–1128. Shcherbakov, D.E. (1996) Origin and evolution of the Auchenorrhyn-
Meyer, G.A. & Whitlow, T.H. (1992) Effects of leaf and sap feeding cha as shown by the fossil record. Studies on Hemipteran Phy-
insects on photosynthetic rates of goldenrod. Oecologia, 92, logeny (ed. by C. W. Schaefer), pp. 31–45. Entomological Society
480–489. of America, Lanham, Maryland.
Moran, N.A., Tran, P. & Gerardo, N.M. (2005) Symbiosis and insect Shcherbakov, D.E. (2002) The 270 million year history of Auchenor-
diversification: an ancient symbiont of sap-feeding insects from rhyncha (Homoptera). Denisia, 4, 29–35.
the bacterial phylum Bacteroidetes. Applied and Environmental Shimodaira, H. & Hasegawa, M. (1999) Multiple comparisons of log-
Microbiology, 71, 8802–8810. likelihoods with applications to phylogenetic inference. Molecular
Nault, L.R. (1987) Origin and evolution of Auchenorrhyncha- Biology and Evolution, 16, 1114–1116.
transmitted, plant infecting viruses. Proceedings of the 2nd Inter- Sorenson, M.D. & Franzosa, E.A. (2007) TreeRot, Version 3. Boston
national Workshop on Leafhoppers and Planthoppers of Economic University, Boston, Massachusetts.
© 2010 The Authors

Swofford, D.L. (2002) PAUP*. Phylogenetic Analysis Using Parsi- (Homoptera: Cercopidae) en Costa Rica. Manejo Integrado de Pagas
mony (*and Other Methods). Version 4.0b10. Sinauer Associates, y Agroecología (Costa Rica), 75, 43–51.
Sunderland, Massachusetts. Tsacas, L. & Couturier, G. (1993) Une nouvelle espèce de Cladochaeta
Synave, H. (1954) Contribution a l’étude des Cercopidae Afrícains. de l’Équateur inquiline des nymphes de Cephisus erythrocephalus
Les genres Cordia Stål et Ptyelus LePeletier et Serville (Hemiptera- [Diptera, Drosophilidae; Homoptera, Aphrophoridae]. Revue Franç-
Homoptera). Volume Jubilaire Victor Van Straelen, Directeur de aise d’Entomologie (Nouvelle série), 15, 85–90.
l’Institute Royal des Sciences Naturelles de Belgique, 1925–1954, Urban, J.M. & Cryan, J.R. (2007) Evolution of the planthoppers
2, 879–886. (Insecta: Hemiptera: Fulgoroidea). Molecular Phylogenetics and
Tamura, K., Dudley, J., Nei, M. & Kumar, S. (2007) MEGA4: Molec- Evolution, 42, 556–572.
ular Evolutionary Genetics Analysis (MEGA) software version 4.0. Urban, J.M. & Cryan, J.R. (2009) Entomologically famous, evolu-
Molecular Biology and Evolution, 24, 1596–1599. tionarily unexplored: the first phylogeny of the lanternfly family
Thompson, V. (1984) Distributional evidence for thermal melanic Fulgoridae (Insecta: Hemiptera: Fulgoroidea). Molecular Phyloge-
color forms in Philaenus spumarius, the polymorphic spittlebug. netics and Evolution, 50, 471–484.
American Midland Naturalist, 111, 288–295. Wagner, M.R., Atuahene, S.K.N. & Cobbinah, J.R. (1991) Forest
Thompson, V. (1994) Spittlebug indicators of nitrogen-fixing plants. Entomology in West Tropical Africa: Forest Insects of Ghana.
Ecological Entomology, 19, 391–398. Kluwer, Dordrecht.
Thompson, V. (1997) Spittlebug nymphs (Homoptera: Cercopidae) Wilson, M.R. & Claridge, M.F. (1991) Handbook for the Identification
in Heliconia flowers (Zingiberales: Heliconiaceae): preadaptation of Leafhoppers and Planthoppers of Rice. CAB International, Oxon.
and evolution of the first aquatic Homoptera. Revista de Biologia Wise, M.J., Kieffer, D.L. & Abrahamson, W.G. (2006) Costs and
Tropical, 45, 905–912. benefits of gregarious feeding in the meadow spittlebug, Philaenus
Thompson, V. (1999) Spittlebugs associated with actinorhizal host spumarius. Ecological Entomology, 31, 548–555.
plants. Canadian Journal of Botany, 77, 1387–1390. Zajac, M.A. & Wilson, M.C. (1984) The effects of nymphal feeding
Thompson, V. (2004) Associative nitrogen fixation, C4 photosynthesis, by the meadow spittlebug, Philaneus spumarius (L.) on strawberry
and the evolution of spittlebugs (Hemiptera: Cercopidae) as major yield and quality. Crop Protection, 3, 167–175.
pests of Neotropical sugar cane and forage grasses. Bulletin of Zwickl, D.J. (2006) Genetic algorithm approaches for the phylogenetic
Entomological Research, 94, 189–200. analysis of large biological sequence datasets under the maximum
Thompson, V. & Halkka, O. (1973) Color polymorphism in some likelihood criterion. PhD dissertation, University of Texas, Austin,
North American Philaenus spumarius (Homptera: Aphrophoridae) Texas.
populations. American Midland Naturalist, 89, 348–359.
Thompson, V. & León González, R. (2005) La identificación y Accepted 8 November 2009
distribución de los salivazos de la caña de azúcar y los pastos First published online 1 March 2010
© 2010 The Authors

View publication stats

Family-Level Relationships of The Spittlebugs and

Uploaded by

Copyright:

Available Formats

Family-Level Relationships of The Spittlebugs and

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Family-Level Relationships of The Spittlebugs and

Uploaded by

Copyright:

Available Formats

See discussions, stats, and author profiles for this publication at: https://www.researchgate.

Family-level relationships of the spittlebugs and froghoppers (Hemiptera:

Article in Systematic Entomology · March 2010

Jason R. Cryan Gavin J. Svenson

SEE PROFILE SEE PROFILE

The user has requested enhancement of the downloaded file.

Family-level relationships of the spittlebugs and

Abstract. The spittlebug superfamily Cercopoidea (Hemiptera: Cicadomorpha) com-

Introduction discharged from the insect’s alimentary system and supple-

© 2010 The Authors

© 2010 The Authors

© 2010 The Authors

GenBank Accession No.

GenBank Accession No.

© 2010 The Authors

GenBank Accession No.

GenBank Accession No.

© 2010 The Authors

alternate alignments of ribosomal genes performed using

hyper-variable regions aligned with muscle. In addition, cer-

differences when all third codon positions were excluded, only

third codon positions of mitochondrial genes were excluded.

A combination of third codon inclusion/exclusion with inclu-

Identical dataset treatments were implemented in the maximum

likelihood (ML) program garli version 0.951 (Zwickl, 2006;

= 4). We investigated six dataset partitioning strategies within

mitochondrial genes, (b) all nuclear genes; (iii) four par-

titions – (a) protein-coding mitochondrial genes, (b) protein-

protein-coding gene, (m) 18S rRNA, (n) 28S rRNA. In addi-

saturation curves based on sequence distances. For all dataset

treatments, we recorded topological findings to select the final

dataset treatment based on the criteria of relationship stability

1998) partnered with paup* 4.0b10 (Swofford, 2002). A final

ML analysis was performed using garli version 0.951

(Zwickl, 2006). Using the GTR+I+G model with unfixed

a globally optimal solution was reached. Nonparametric

likelihood bootstrap trees were reconstructed using garli

(500 replicates – 400 000 generations per replicate) and

imported into paup for calculation of a 50% majority-rule

consensus tree. Using treerot v.3 (Sorenson & Franzosa,

2007), we calculated partitioned Bremer support values (six

partitions based on gene loci) by creating a constrained nodal

search command file based on the ML tree topology for

A final MMB analysis coupled with Markov chain Monte

Carlo was implemented in the program mrbayes v3.1.2

(Ronquist & Huelsenbeck, 2003). We conducted these analyses

© 2010 The Authors

© 2010 The Authors

© 2010 The Authors

© 2010 The Authors

Maximum Total Bremer Bremer

1 97 95 100 23.0 12.5 0.0 −8.5 −6.5 11.0 14.5

7 63 76 100 8.0 −3.0 3.1 0.3 −4.9 −1.1 13.6

Maximum Total Bremer Bremer

41 100 100 100 18.0 1.0 9.0 3.0 7.0

© 2010 The Authors

80 <50 <50 76 0.0 0.0 0.0 0.0 0.0 0.0 0.0

Maximum Total Bremer Bremer

81 86 73 100 0.0 0.0 0.0 0.0 0.0 0.0 0.0