Phylogeny of Moth Lacewings and Giant Lacewings ... - Sharkey lab

Phylogeny of Moth Lacewings and Giant Lacewings (Neuroptera: Ithonidae, Polystoechotidae) Using DNA Sequence Data, Morphology, and Fossils Author(s): Shaun L. Winterton and Vladimir N. Makarkin Source: Annals of the Entomological Society of America, 103(4):511-522. 2010. Published By: Entomological Society of America DOI: 10.1603/AN10026 URL: http://www.bioone.org/doi/full/10.1603/AN10026

BioOne (www.bioone.org) is an electronic aggregator of bioscience research content, and the online home to over 160 journals and books published by not-for-profit societies, associations, museums, institutions, and presses. Your use of this PDF, the BioOne Web site, and all posted and associated content indicates your acceptance of BioOne’s Terms of Use, available at www.bioone.org/page/terms_of_use. Usage of BioOne content is strictly limited to personal, educational, and non-commercial use. Commercial inquiries or rights and permissions requests should be directed to the individual publisher as copyright holder.

BioOne sees sustainable scholarly publishing as an inherently collaborative enterprise connecting authors, nonprofit publishers, academic institutions, research libraries, and research funders in the common goal of maximizing access to critical research.

SYSTEMATICS

Phylogeny of Moth Lacewings and Giant Lacewings (Neuroptera: Ithonidae, Polystoechotidae) Using DNA Sequence Data, Morphology, and Fossils SHAUN L. WINTERTON1

AND

VLADIMIR N. MAKARKIN2

Ann. Entomol. Soc. Am. 103(4): 511Ð522 (2010); DOI: 10.1603/AN10026

ABSTRACT A phylogeny of lacewing families Ithonidae and Polystoechotidae is presented based on three gene markers (16S and 18S ribosomal DNA and CAD) and 23 morphological characters. Living and fossil genera presently placed in Polystoechotidae (Fontecilla Nava´s, Platystoechotes Carpenter and Polystoechotes Burmeister) and Ithonidae (Adamsiana Penny, Allorapisma Makarkin & Archibald, Ithone Newman, Megalithone Riek, Oliarces Banks, Principiala Makarkin & Menon, Rapisma Walker and Varnia Walker) were included in phylogenetic analyses (parsimony and Bayesian) and compared with outgroups selected from various families of Neuroptera. The resulting phylogeny recovered a monophyletic clade comprising Ithonidae and Polystoechotidae as hypothesized previously. Rapismatidae as a separate family is not supported and Ithonidae are rendered paraphyletic with three extant genera previously placed in Ithonidae (Adamsiana, Oliarces, and Rapisma), recovered deep within Polystoechotidae. The fossil genera Allorapisma and Principiala formed a sister-group relationship with Rapisma, also within Polystoechotidae. Due to the lack of mutually exclusive synapomorphies for either Ithonidae or Polystoechotidae, a single more inclusive family Ithonidae is proposed, including all ithonid genera and all genera previously placed in Polystoechotidae. Synapomorphies characterizing the revised concept of Ithonidae s.l. are discussed. KEY WORDS phylogeny, Rapismatidae, Ithonidae, Polystoechotidae, lacewing

Ithonidae (moth lacewings) and Polystoechotidae (giant lacewings) are small families of robust and often hairy lacewings. Ithonidae presently comprises seven extant genera from Australia, Southeast Asia, and the New World, whereas Polystoechotidae comprise three extant New World genera. Three genera of ithonids are found in Australia; Ithone Newman (Fig. 1A) contains 10 species, whereas Megalithone Riek and Varnia Walker contain two species each (Riek 1974). Three genera also are described from the New World, and all are monotypic, although additional undescribed species are known in collections (Oswald 1998b). Oliarces clara Banks (Fig. 1B) is known from arid areas of southwestern North America. Adamsiana curoi Penny is a montane species from Central America and is unique in that females are apterous, whereas males are macropterous (Penny 1996). Narodona mexicana Nava´s is only known from the original description and a hind wing Þgure of a single specimen from Mexico (Nava´s 1929a). A single ithonid genus is Any opinions, Þndings, and conclusions or recommendations expressed in this publication are those of the authors and do not necessarily reßect the views of the National Science Foundation. 1 Corresponding author: School of Biological Sciences, University of Queensland, St. Lucia, Queensland, Australia 4072 (e-mail: [email protected]). 2 Institute of Biology and Soil Sciences, Far East Branch of the Russian Academy of Sciences, Vladivostok, 960022, Russia.

known from Southeast Asia (Rapisma McLachlan) containing ⬇20 species from montane areas in Nepal to tropical lowlands in Thailand and Malaysia (Barnard 1981, Barnard and New 1985, New 1985, Yang 1993). Rapisma was previously considered as the sole genus in the family Rapismatidae, although there seems little evidence for maintaining this separation (Penny 1996, Makarkin and Menon 2007). Polystoechotidae are represented by Platystoechotes Carpenter, Polystoechotes Burmeister (Fig. 1C), and Fontecilla Nava´s consisting of four species in total. Fontecilla graphicus Nava´s and Polystoechotes gazullai Nava´s are known only from Chile, whereas Platystoechotes lineatus Carpenter is known from the southwestern United States (California) (Oswald 1998b). A second species of Polystoechotes (P. punctatus F.) is recorded from Canada south to Panama (Oswald 1998b, Archibald and Makarkin 2006; Fig. 1). Specimens of both Ithonidae and Polystoechotidae are relatively rare in collections, although localized mass emergences have been recorded in Oliarces Banks and Ithone after seasonal rains (Riek 1974, Faulkner 1990). Larvae of Ithonidae and Polystoechotidae are known for Oliarces Banks, Megalithone, Ithone, Polystoechotes, and Platystoechotes (MacLeod 1964, Grebennikov 2004, Winterton et al. 2010). All are fossorial and in Ithonidae the mature larvae are scarabaeiform. Larvae of Ithonidae were originally re-

0013-8746/10/0511Ð0522$04.00/0 䉷 2010 Entomological Society of America

512

ANNALS OF THE ENTOMOLOGICAL SOCIETY OF AMERICA

Vol. 103, no. 4

Fig. 1. Adult Ithonidae and Polystoechotidae. (A) Ithone fulva Tillyard (Australia: Queensland). (B) Oliarces clara Banks. (USA: California). (C) Polystoechotes punctatus (F.) (USA: Washington). Photograph credits: S. L. Winterton (A and B); Karl Volkman (C). (Online Þgure in color.)

ported as predatory by Tillyard (1922), but Gallard (1932) subsequently proposed that they may be phytophagous or at least saprophagous on the bark of plant roots. This is yet to be conÞrmed, but obvious similarities in larval morphology between ithonids and polystoechotids suggest a common feeding biology. Ithonid larvae were recorded by Gallard (1932) to congregate around the bases of eucalyptus trees and observed “chaÞng,” shredding and actively feeding on the bark and roots to extract either sap or decaying liquid material. Similar congregational patterns have been recorded by Faulkner (1990) around the bases of creosote bushes [Larrea tridentata (DC) Coville] by Oliarces and around trees in montane regions by Polystoechotidae. Based on detailed studies of larval head morphology, MacLeod (1964) proposed that Ithonidae and Polystoechotidae were sister-families closely related to Myrmeleontiformia (containing the families Ascalaphidae, Myrmeleontidae, Nemopteridae, Nymphidae, and Psychopsidae). Phylogenetic studies using morphology, DNA sequence data, or a

combination of the two have conÞrmed this hypothesis (Aspo¨ ck et al. 2001, Winterton 2003, Haring and Aspo¨ ck 2004, Winterton et al. 2010). The divergence time for the Ithonidae and Polystoechotidae clade is estimated as during the Late Triassic based on molecular data, with Ithonidae diverging from Polystoechotidae during the Jurassic (Winterton et al. 2010). Fossil Polystoechotidae are described from the Middle Jurassic to Late Eocene (Lambkin 1988, Ren et al. 2002, Makarkin and Archibald 2003, Archibald and Makarkin 2006), whereas “rapismatid-like” fossil ithonids have recently been described from the Early Cretaceous and Early Eocene (Makarkin and Menon 2007, Makarkin and Archibald 2009). The family-level placement of some ithonids has been questioned previously. Tillyard (1926) included Oliarces in Ithonidae that was supported by Carpenter (1951), but Lameere (1936) suggested that the genus belonged to Polystoechotidae. This was supported by a phylogenetic analysis using morphology and multiple molecular markers by Winterton et al. (2010), which strongly indicated that Oliarces should be trans-

July 2010 Table 1.

WINTERTON AND MAKARKIN: ITHONIDAE AND POLYSTOECHOTIDAE PHYLOGENY Exemplars used in this study for sequencing and/or morphological scoringa

Family/subfamily Outgroup Myrmeleontidae Nymphidae Osmylidae Ingroup Ithonidae

513

Exemplar Stilbopteryx costalis Newman Myiodactylus osmyloides Brauer Porismus strigatus (Burmeister)

Specimen voucher/code

16S

18S

CASENT8092179 EU734873 EU815255 EU860126 Winterton et al. (2010) CASENT8092210 EU734894 EU815278 EU860147 Winterton et al. (2010) HONDURAS: Cortes: Cusuco 关15⬚ 30⬘N, 88⬚ 13⬘W兴, 1900 m, 20 July 1994, Dan Curoe Makarkin and Archibald (2009) CASENT8092184 EU734865 EU815247 EU860118 Winterton et al. (2010) CASENT8092183 EU734866 EU815248 EU860119 Winterton et al. (2010) CASENT8092185 EU734885 AF012527

EU860138 Winterton et al. (2010) Makarkin and Menon (2007)

CASENT8092186 EU734899 U815282

Varnia implexa (Nava´s) Polystoechotidae Fontecilla graphicus Nava´s Platystoechotes lineatus Carpenter Polystoechotes punctatus (F.)

Collection data or reference

CASENT8092178 EU734908 EU815291 EU860159 Winterton et al. (2010)

Adamsiana curoi Penny Allorapisma chuorum Makarkin & Archibald Ithone fulva Tillyard Megalithone tillyardi Riek Oliarces clara Banks Principiala incerta Makarkin & Menon Rapisma cryptunum New & Barnard

CAD

CASC201

EU860151 THAILAND: Nakhon Nayok Khao Yai Khao Keow spirit house, 14⬚ 22.960⬘ N, 101⬚ 23.253⬘ E, 750 m, Malaise trap, 5Ð12-IX-2006, Pong Sandeo leg. T909 AUSTRALIA: Queensland: 5 km E Charleville, 17.vi.1983, R. Silcock, T.6185 EU734863 EU815244 EU860115 Winterton et al. (2010)

CASENT8092170 EU734890 EU815274 EU860143 Winterton et al. (2010) CASENT8092171 EU734893 EU815277 EU860146 Winterton et al. (2010)

a Accession numbers indicate specimens sequenced for various genes, with gene sequences deposited in GenBank (http://www.ncbi. nih.gov/). Voucher numbers identify individual specimens deposited in the California Academy of Sciences, San Francisco, CA.

ferred from Ithonidae to Polystoechotidae. Recent collections of various genera of Ithonidae and Polystoechotidae, suitable for DNA sequencing, have presented an opportunity to undertake a molecular phylogenetic study of these comparatively rare insects to try to elucidate their phylogenetic placement. A phylogeny for all extant genera of Polystoechotidae and Ithonidae is presented based on morphology and molecular DNA sequence data. Phylogenetic analyses were undertaken using maximum parsimony and Bayesian inference methods. Two extinct species of Ithonidae (Principiala Makarkin & Menon and Allorapisma Makarkin & Archibald) also are included in the analysis. The status of Ithonidae and Polystoechotidae as separate families, or alternatively, united as a single family, is discussed. We conclude that the observed phylogeny better corresponds to the latter, and a revised diagnosis and membership of the family Ithonidae s.l. is proposed. Materials and Methods Terminology. Wing venation terminology used here generally follows Oswald (1993) and genitalic terminology follows Oswald (1998a). Exemplar Selection. Ingroup taxa included all extant genera of Polystoechotidae (Polystoechotes,

Platystoechotes, and Fontecilla) and Ithonidae (Adamsiana Penny, Ithone, Megalithone, Oliarces, Rapisma, and Varnia), except Narodona Nava´s as this genus is known only from a brief description and hind wing Þgure (Table 1; Fig. 2). Two relatively complete fossils of ithonids also were included (Principiala and Allorapisma ). Outgroups were included from lacewing families Osmylidae (Porismus strigatus [Burmeister]), Nymphidae (Myiodactylus osmyloides Brauer), and Myrmeleontidae (Stilbopteryx costalis Newman). Twenty-three adult and larval characters were scored for all taxa, where data are available (Appendix 1). The two extinct species represented by compression fossils could not be scored for all characters as some could not be observed. Gene Sequencing. All gene sequences were downloaded from GenBank (Table 1) except Rapisma cryptunum New & Barnard. DNA sequencing for this additional taxon was carried out following the identical protocol outlined by Winterton et al. (2010). GenBank accession and specimen voucher numbers are presented in Table 1. Sequence Alignment and Phylogenetic Analysis. Alignment of all sequences was done manually, although CAD was aligned with reference to translated amino acid sequences (standard eukaryote) by using MacClade, version 4.06 (Maddison and Maddison

514


Vol. 103, no. 4

Fig. 2. Wings of Polystoechotidae and Ithonidae. (A) Rapisma cryptunum New & Barnard. (B) Oliarces clara Banks. (C) Platystoechotes lineatus Carpenter. (D) Varnia implexa (Nava´s). (E) Ithone fulva Tillyard. Scale ⫽ 2.0 mm. Abbreviations: 1A, 2A, 3A, Þrst to third anal veins; CuA, anterior cubitus; CuP, posterior cubitus; MA and MP, anterior and posterior branches of media; phs, prehumeral space; R1, Þrst branch of radius; Rs, radial sector; Sc, subcosta.

2000). Parsimony analyses were conducted using PAUP*4.0b10 (Swofford 1999) using a branch and bound search protocol. Bootstrap support values for the parsimony analyses were calculated from 1,000 heuristic search (tree bisection and reconnection)

pseudoreplicates of resampled data sets, each with 30 random additions (constant characters excluded). Bayesian analyses were preformed using MrBayes 3.1.2 (Ronquist and Huelsenbeck 2003). The data were partitioned by data type (DNA sequence and

July 2010

WINTERTON AND MAKARKIN: ITHONIDAE AND POLYSTOECHOTIDAE PHYLOGENY

Table 2. Character matrix for 23 morphological characters for three outgroup taxa (Porismus, Myiodactylus, and Stilbopteryx) and 11 ingroup taxa representing all extant, and two extinct, genera of Ithonidae and Polystoechotidaea Porismus Myiodactylus Stilbopteryx Adamsiana Allorapisma Fontecilla Ithone Megalithone Oliarces Platystoechotes Polystoechotes Principiala Rapisma Varnia

0021001000 0000200000 000 0020110000 1000200100 001 0020001001 0000100000 001 1100010001 1100110120 002 ?101010101 0001?0???? ??2 1110110010 0100200000 002 1100000000 0010000011 112 1100000000 0010000011 112 1100000001 1000100021 002 1110110110 0100200000 002 1110100010 0100000000 002 1101110100 0001?0???0 ??2 1101010101 1/00101100 002 1101000100 0010200011 112

a Polymorphisms are indicated by a forward slash representing states 0 or 1.

morphology), locus, and by the remaining two codon positions for each protein-coding locus. A separate GTR ⫹ ␥ nucleotide substitution model was applied to each DNA partition. The mk1 model (Lewis 2001), with coding set to variable, was applied to the morphology partition. Each analysis consisted of four Markov chain Monte Carlo chains run simultaneously for 10 million generations. Trees were sampled every 1,000th generation. The Þrst 1 million trees were discarded as burn-in. A majority rule consensus tree was computed with posterior probabilities (PP) for each node. Morphological characters were treated as unordered and equally weighted, with unknown states treated as Ô?Õ Results The combined sequence length for our concatenated molecular data set was 4479 bp (CAD ⫽ 2145 bp; 18S ⫽ 1848 bp; 16S ⫽ 486 bp) once ambiguously aligned sites (i.e., hypervariable stem regions in ribosomal genes) were removed. Addition of 23 morphological characters (Table 2) increased the combined data set to 4,502 characters. Average base frequencies across the data set for 16S were greatly A/T biased (A, 34.6%; C, 11.04%; G, 17.83%; T, 36.53%) (␹2 ⫽ 17.419, df ⫽ 39, P ⫽ 0.998), whereas for 18S sequences they were in relatively equal proportions except for a slightly lower cytosine content (A, 25.73%; C, 22.27%; G, 26.69%; T, 25.30%) (␹2 ⫽ 4.757.593, df ⫽ 39, P ⫽ 1.000). Average base frequencies for CAD (all sites) were A, 32.87%; C, 15.38%; G, 20.48%; and T, 31.26% (␹2 ⫽ 77.666, df ⫽ 39, P ⫽ 0.002) and as is common with third positions, A/T bias was considerably greater (77.78%) than in either Þrst (51.04%) or second (62.56%) codon positions. Average uncorrected sequence divergences across taxa for all DNA loci varied between the outgroup Porismus and most ingroup taxa at 13.9 Ð15.5%, whereas between various ingroup genera, pairwise divergences ranged from 1.2% between closely related Ithone and Megalithone and 9.9% between the more distantly related Rapisma and Mega-

515

lithone. Ithonidae were consistently divergent from Polystoechotidae by 7.5Ð10.9% (inclusive of Rapisma and Oliarces). Inclusion of fossil exemplars in the analysis is likely to weaken the Þnal estimate due to the low amount of data available (18 informative characters). Therefore initial parsimony and Bayesian analyses of the combined molecular and morphological data were undertaken with Allorapisma and Principiala excluded. The results of the Bayesian analysis are presented in Fig. 3; six most parsimonious trees were recovered from the parsimony analysis (tree length ⫽ 2106, consistency index [CI] ⫽ 0.708, retention index [RI] ⫽ 0.492), although the resolved parts of the topology were the same as the Bayesian tree. Two polytomies in the parsimony tree represented unresolved relationships among Adamsiana, Oliarces, and Rapisma as well as Ithone, Megalithone, and Varnia, respectively. All other branches on the tree were resolved with relatively high branch support. Adamsiana, Oliarces, and Rapisma were recovered as a clade deep within Polystoechotidae, subtended by a laddered grade consisting of Platystoechotes, Polystoechotes, and Fontecilla. Ithone, Megalithone, and Varnia formed a well supported clade sister to Polystoechotidae. In a more inclusive analysis, all 14 taxa (including Allorapisma and Principiala ) were compared in a combined molecular and morphological data analysis, consisting of 616 (18%) parsimony informative characters and 3,886 parsimony uninformative characters (including 3,395 constant characters). The parsimony analysis yielded three trees (length ⫽ 2,110 steps; CI ⫽ 0.707; RI ⫽ 0.494) of similar topology to the reduced taxon set phylogeny, with Allorapisma and Principiala placed a sister to the clade consisting of Adamsiana, Oliarces, and Rapisma. A Bayesian phylogram, with parsimony bootstrap values and Bayesian posterior probability values is presented in Fig. 4. In this analysis, the two fossil genera are placed as sister to Rapisma; otherwise, the tree topology is the same as the parsimony analysis. Lower posterior probability values on certain branches correspond to lower support or unresolved topologies in the parsimony tree. Adamsiana, Oliarces, and Rapisma are again placed deep within Polystoechotidae, with relatively high branch support. Ithonidae are again a well supported clade (88% bootstrap; 0.9 PP) represented by the extant Australian genera Megalithone, Ithone, and Varnia. Ithonidae ⫹ Polystoechotidae were recovered as a monophyletic clade with relatively strong branch support (73% bootstrap; 0.9 PP). Morphological characters supporting the monophyly of the clade Ithonidae ⫹ Polystoechotidae include head retracted under prothorax (char. 1), humeral veinlet recurrent with multiple branches toward wing base (char. 2), and larval jaws short and broad for chafÞng (char. 23); characters one and 23 are apparently unique within the order and are a synapomorphy for Ithonidae ⫹ Polystoechotidae. As was found in Winterton et al. (2010), in all analyses Ithonidae was rendered paraphyletic with respect to Polystoechotidae with relatively high levels of sup-

516


Vol. 103, no. 4

Fig. 3. Bayesian phylogram of relationships of extant taxa of Ithonidae and Polystoechotidae based on analyses of DNA sequence data and 23 morphological characters. Branch length represents number of changes. Bayesian posterior probability and parsimony bootstrap support values are presented in order on each branch. Outgroups are Osmylidae (OSMY), Nymphidae (NYMP), and Myrmeleontidae (MYRM).

port. Three genera presently placed in Ithonidae, i.e., Adamsiana, Oliarces, and Rapisma, were placed as a monophyletic clade deep within Polystoechotidae, although no discreet morphological characters could be found to support inclusion in this clade in the family. Adamsiana, Rapisma, and Oliarces are supported as a clade by wing trichosors being absent (char. 10) and pale wing venation (char. 11). When the fossil genera Allorapisma and Principiala were included in the analysis, they formed a sister-group relationship with Rapisma, although branch support within the clade comprising Adamsiana, Rapisma, Oliarces, Allorapisma , and Principiala became nonexistent. The sister-group relationship between Allorapisma and Principiala is supported by the divergent branching pattern of the forewing medial vein (char. 14) (Makarkin and Menon, 2007, Þgs. 3 and 5; Makarkin and Archibald 2009, Þgs. 2 and 3). The Australian genera of Ithonidae (Ithone, Megalithone, and Varnia) were recovered as a monophyletic clade in all analyses (bootstrap support ⫽ 88 Ð99%; PP ⫽ 0.89 Ð1.0) with Megalithone weakly supported as sister to Varnia in the pruned analysis. Five characters are synapomorphies for this clade: forewing with two radial branches originating on R (char. 13), mediuncus triangular (char. 19), male ectoprocts enlarged (homoplasious in Oliar-

ces) (char. 20), male sternite 9 shovel-like (char. 21), and female sternite 9 forming a psammorotrum (char. 22). Discussion The results of this analysis using a combination of DNA sequence data and morphology support revised deÞnitions and taxon composition for both Polystoechotidae and Ithonidae, although treating the clade as a single family should also be considered. Ithonidae, based on these results, comprises only the endemic Australian genera Ithone, Megalithone, and Varnia. Synapomorphies supporting Ithonidae in this context include forewing with two radial branches originating on R, mediuncus triangular, male sternite 9 shovel-like, and female sternite 9 forming a psammorotrum. Here, genera previously placed in Ithonidae, i.e., Rapisma, Oliarces, and Adamsiana, were recovered within Polystoechotidae along with Polystoechotes, Platystoechotes, and Fontecilla. Consequently, Polystoechotidae are predominantly distributed in the New World (Þve genera), with the single, relatively species-rich genus Rapisma, distributed throughout Southeast Asia. Multiple fossils of putative Polystoechotidae have been described, but deÞnitive

July 2010


517

Fig. 4. Phylogram of relationships of genera of Ithonidae and Polystoechotidae based on matching topologies from Bayesian and parsimony analyses of DNA sequence data and 23 morphological characters. Two additional extinct genera (Allorapisma Makarkin & Archibald, Principiala Makarkin & Menon) of Ithonidae also are included. Branch length represents number of changes. Bayesian Posterior Probability and bootstrap support values are presented in order on each branch. Wing drawings of Principiala and Allorapisma are modiÞed after Makarkin and Menon (2007) and Makarkin and Archibald (2009), respectively.

518


family-level assignments have been problematic for some taxa (Lambkin 1988, Ren et al. 2002, Makarkin and Archibald 2003, Archibald and Makarkin 2006). The reason for this confusion is that synapomorphies deÞning Polystoechotidae are mostly lacking and characters used to differentiate the family are often plesiomorphies (Makarkin and Archibald 2003, Archibald and Makarkin 2006). The placement of Oliarces and Rapisma is not surprising with similarities between the two genera, and with Polystoechotidae, noted by previous authors. Lameere (1936) suggested that Oliarces belonged to Polystoechotidae, and Grebennikov (2004) described a close morphological similarity of Oliarces Þrst instars to a putative polystoechotid. Makarkin and Archibald (2009) questioned the identity of this larva, suggesting that it could be an ithonid. Tillyard (1916, 1919b) suggested that Rapisma and Oliarces do not belong in Ithonidae but require separate family status rather than placement in Polystoechotidae. Barnard (1981) supported the separation of Rapismatidae as a distinct family and listed eight characters that separated this family from Ithonidae. In his description and discussion of Adamsiana, Penny (1996) considered each of these characters in turn, with the new genus exhibiting character states of both Ithonidae and Rapismatidae and therefore suggested that the separation of the two families was not necessary. Neither Barnard (1981) nor Penny (1996) considered Polystoechotidae in their studies. Based on internal female genitalic morphology, Szira´ki (1998) indicated that Rapisma showed a closer similarity with Polystoechotidae than Ithonidae. Our results are congruent with this conclusion, with Rapisma placed within Polystoechotidae and that the genus does not deserve separate family status. The Mexican ithonid genus Narodona is known only from the original description and Þgure of the hind wing, with the type probably destroyed (Monserrat 1985). Based on the wing Þgure, this genus shows a clear afÞnity with Adamsiana (Penny 1996) and is most likely closely related. Given that Adamsiana is placed within Polystoechotidae in this analysis, it is clear that Narodona is probably a polystoechotid as well. The incomplete yet highly autapomorphic wing venation of Principiala and Allorapisma suggests placement within Polystoechotidae. The sister-group relationship with Rapisma supports previous suggestions by Makarkin and Menon (2007) of a close relationship among these genera based on overall similarities in wing venation and body shape. Is there justiÞcation for maintaining two separate families, Ithonidae and Polystoechotidae? The sister group relationship between Ithonidae ⫹ Polystoechotidae is supported by two morphological synapomorphies: head retracted under prothorax in adults and larval jaws very short and broad. Although larval biological data are incomplete for extant taxa (and nearly impossible for extinct taxa), it is likely that being fossorial, along with the probable phytophagous feeding biology, represent synapomorphies for this

Vol. 103, no. 4

clade. Convincing mutually exclusive synapomorphies supporting both families are lacking, either as previously deÞned or based on our combined analyses. It seems that the most reasonable conclusion is to combine Ithonidae (including Rapismatidae) with Polystoechotidae as a single family containing all the genera presently included in these two families. In conclusion, it is clear from the combined morphological and DNA sequence data presented here that the separation of Ithonidae and Polystoechotidae is not warranted. Our results strongly support either the transfer of Adamsiana, Oliarces, and Rapisma to Polystoechotidae, or combining all the taxa into a single family. Morphological characters deÞning Polystoechotidae relative to Ithonidae are few, but there are some putative synapomorphies deÞning a single family, including adult and larval morphology and larval biology. Consequently, we have chosen the latter option as a more reasonable alternative to maintaining Ithonidae and Polystoechotidae as separate, but poorly deÞned, family groups. Further studies are required on the group, in particular, acquisition of DNA sequence data for Narodona, Adamsiana, and Varnia as well as biological studies of the larval morphology and feeding biology of all species. Family Ithonidae Newman, 1853, sensu novo Ithonesidae Newman, 1853: ccii [incorrect formation of family name]. Type genus: Ithone Newman, 1838. Ithonidae: Tillyard, 1916: 274 [corrected name for Ithonesidae]. Polystoechotidae Handlirsch, 1906 Ð1908 [1906]: 42. Type genus: Polystoechotes Burmeister, 1839. Syn. nov. Rapismiden Kru¨ ger, 1922: 139. Type genus: Rapisma MacLachlan, 1866. Rapismidae: Kru¨ ger, 1923: 72, 73. Rhapismidae Nava´s, 1929b: 376 (as “fam. nov.”). Type genus: Rhapisma [unjustiÞed emendation of Rapisma]. Rapismatidae: Barnard, 1981: 122 (corrected name for Rhapismidae). Diagnosis. Robust, medium-sized to large neuropterans (forewing length, ⬇15Ð50 mm) distinguished from similar families by the presence of the following features: Head more or less retracted under pronotum; forewing elongate oval to subtriangular (broad-triangular in Brongnairtiellidae, Osmylopsychopidae); antehumeral space long and broad in both wings, disproportionately large in genera with relatively narrow costal space; humeral veinlet well developed, recurrent, pectinate branched in both wings (in hindwing with at least one branch) (absent in hindwing of Hemerobiidae; absent in both wings of Dilaridae); apical area between C and R1 (or Sc⫹R1) markedly expanded (except perhaps Rapisma), with long, mainly forked veinlets of R1 or Sc⫹R1 in both wings (not markedly expanded in Hemerobiidae, Dilaridae); distal nygma present between two proximal branches of Rs in both wings (absent in Brongnairtiellidae, Psy-

July 2010


chopsidae, Hemerobiidae). Larval jaws relatively short and very broad basally; larva fossorial, associated with plant roots. Genera Included. Kasachstania PanÞlov 1980, Osmyloides PanÞlov 1980 and Paleopterocalla Oswald, 2007 (⫽Pterocalla PanÞlov 1980; all from the Late Jurassic of Karatau, Kazakhstan); Principiala Makarkin et Menon, 2007 (Early Cretaceous of Brazil and England); Palaeopsychops Andersen, 2001 (Early Eocene of Denmark and western North America); Allorapisma Makarkin et Archibald, 2009 (Early Eocene of western North America); Ithone Newman, Megalithone Riek, Varnia Walker (Australia); Rapisma (Oriental Region); Adamsiana Penny, Oliarces Banks, Narodona Nava´s, Polystoechotes Burmeister, Platystoechotes Carpenter and Fontecilla Nava´s (North and South America). Comments. Ithonidae in a broader sense (i.e., inclusive of Polystoechotidae and Rapismatidae) as deÞned here includes six extinct and 10 extant genera restricted to Australia, the Oriental region and the New World. The family originated during the TriassicJurassic, more widely distributed during the Jurassic than the present (with many undescribed taxa from Chinese Jurassic deposits). Key to Extant Genera of Ithonidae s.l. 1. Wing venation pale green or white; wings rarely maculate . . . . . . . . . . . . . . . . . . . . . . . 2 Ñ Wing venation dark; wings usually strongly maculate or smoky infuscate . . . . . . . . . . 4 2. Wing venation brilliant white; body dark (southwestern United States) . . . . . Oliarces Ñ Wing venation and body mostly pale (vivid green in live individuals) . . . . . . . . . . . . 3 3. Wings usually spotted or banded; subcostal veinlets connected by numerous cross-veins so that serial ranks of cells present; female macropterous (Oriental region) . . . Rapisma Ñ Wings without infuscate patterning, membrane hyaline; subcostal veinlets not connected by cross-veins, single rank of cells; female apterous (New World) . . Adamsiana 4. Forewing R vein with additional branch, basal to pectinate Rs vein (Fig. 2E) (Australia) . . . 5 Ñ Forewing R vein with a single pectinate Rs vein (New World) . . . . . . . . . . . . . . . . 7 5. Wings mottled; cross-veins interlinking subcostal veinlets, creating multiple ranks of cells . . . . . . . . . . . . . . . . . . . . . . Varnia Ñ Wings uniform infuscate; single row of cells in costal space . . . . . . . . . . . . . . . . . . . 6 7. Fore basitarsus at most no longer than apical tarsomere; male ectoprocts enlarged and widened at apex . . . . . . . . . . . . . Megalithone Ñ Fore basitarsus much longer than apical tarsomere; male ectoprocts narrow and rounded at apex; wing uniform infuscate; single row of cells in costal space . . . . . . . . . . . . . Ithone 8. Hind wing without prominent maculae (pterostigma slightly marked) . . . P . olystoechotes

519

Ñ Hind wing with prominent maculae, particularly in posterior region . . . . . . . . . . . . 9 9. Six to nine R1-Rs cross-veins in both wings (southwestern United States). . . . . . . Platystoechotes Ñ Less than three R1-Rs cross-veins in both wings (Chile) . . . . . . . . . . . . . . Fontecilla Appendix 1: Descriptions of Morphological Character States 1. Head. (0) clearly visible from above; (1) partially concealed beneath anterior margin of prothorax (Fig. 1). The derived state of this character is a synapomorphy for Polystoechotidae ⫹ Ithonidae and has been noted previously by Makarkin and Menon (2007). 2. Forewing Humeral Veinlet. (0) simple, crossvein-like (Oswald 1998a: Þg. 1); (1) recurrent, branched (Fig. 2BÐE). The presence of the well developed recurrent and branched humeral veinlet is characteristic of the order Neuroptera. It occurs in psychopsoid-like families, Ithonidae, Polystoechotidae, Hemerobiidae, and some Berothidae and Mantispidae. In this analysis, the state is considered a synapomorphy of the genera of Ithonidae and Polystoechotidae; it has probably evolved independently at least three times, in Hemerobiidae, Psychopsidae (and related fossil groups), and the clade Berothidae ⫹ Mantispidae. 3. Proximity of Forewing Subcostal (Sc) and Radial (R) Veins. (0) Sc and R1 completely separate distally (Fig. 2A and B, C, E); (1) closely approximated but not fused (Fig. 2C); (2) fused (Oswald, 1998a: Þg. 1). Most of the ingroup exhibit the plesiomorphic state (0) with the Sc and R1 veins separate along entire length, whereas Fontecilla, Polystoechotes, and Platystoechotes have the veins closely approximated distally (1). Only the outgroups have Sc and R1 fused toward the apex of the wing (2). In the majority of the fossil species of Polystoechotidae not included in this analysis, however, these veins are fused. 4. Number of Forewing Subcostal (Sc-R1) CrossVeins. (0) one, near wing base; (1) numerous crossveins along length of subcostal area. The plesiomorphic state was most numerous, with the derived state present in Porismus, Varnia, Rapisma, Allorapisma , and Principiala . 5. Number of Forewing Rs Branches. (0) 4 Ð 8; (1) 9 Ð12. Relatively few Rs branches is considered in this analysis as plesiomorphic and is found in Porismus, Stilbopteryx, and Ithonidae. The polarity of this character is not clear and in general, the states of character may depend largely on the wing size. 6. Forewing Costal Space. (0) relatively narrow basally and equal width along entire length (Fig. 2B); (1) distinctly dilated basally (Fig. 2C). In Porismus, the costal space is narrow basally and distinctly dilated at some distance from the base; state (0) is used in the analysis. This character is highly variable among all ingroup and outgroup taxa. 7. Forewing Subcostal Veinlets. (0) forked (Fig. 2AÐE); (1) simple (Carpenter 1940: Þg. 73). The forewing subcostal veinlets have extensive terminal

520


branching in all ingroup taxa; the amount of branching is much less in Myiodactylus. 8. Forewing Subcostal Veinlets. (0) not interconnected (Fig. 2B and E); (1) interconnected by crossveins, forming multiple ranks of cells between costal margin and Sc (Fig. 2A and C, D). The derived state, with the subcostal veinlets interconnected and forming multiple ranks of cells is found in Principiala , Allorapisma , Varnia, Platystoechotes, and Rapisma. The genus Fontecilla possesses a few cross-veins, irregularity placed but not is serial ranks. 9. Forewing Gradate Series. (0) irregularly arranged, more than two series present (Fig. 2A and D, E); (1) two distinct series only (Fig. 2C). The derived state, with gradates arranged in two series is found only in Platystoechotes, Polystoechotes and Fontecilla. 10. Forewing Trichosors. (0) present along entire wing margin; (1) absent entirely, or present in only small portion of wing margin. The presence of wing trichosors is a synapomorphy for Neuroptera. The trichosors are frequently reduced in number or completely absent in many families, as found here in Stilbopteryx and the clade Adamsiana, Oliarces, Allorapisma , and Rapisma. 11. Wing Venation. (0) dark (black or brown) (Fig. 2CÐE); (1) pale (white to light green) (Fig. 2A and B). Pale wing venation is derived and is present in Myiodactylus and ingroup genera Oliarces, Rapisma, and Adamsiana. In Adamsiana, Myiodactylus, and Rapisma, the body and wings are green in living specimens, whereas in Oliarces the body is dark and wing veins are vivid white (Fig. 1B). 12. Forewing Shape. (0) symmetrical, largely ovoid (Fig. 2E); (1) asymmetrical tending toward falcate (Fig. 2C). An asymmetrical forewing is a synapomorphy for Polystoechotidae, although it is secondarily symmetrical in Oliarces, Allorapisma , Principiala , and some Rapisma. 13. Forewing Radial Branches. (0) single, pectinate Rs originating from R (Fig. 2AÐC); (1) two branches originating from R, the distal branch being the pectinate Rs (Fig. 2D and E). The plesiomorphic condition occurs throughout Neuroptera. The derived condition represents a synapomorphy for Ithonidae s.s. (i.e., Ithone, Megalithone, and Varnia) and has evolved independently in Hemerobiidae and some Dilaridae and Kalligrammatidae. Carpenter (1951) incorrectly Þgured the wing of Ithone fusca Newman by omitting this basal free radial branch, instead showing only a single, pectinate Rs originating on R. 14. Fork of Forewing MP. (0) branches closely proximate, diverging acutely and only composed of two to three branches reaching wing margin (Fig. 2AÐE); (1) branches widely divergent composed of more than Þve branches reaching wing margin (Makarkin and Menon 2007: Þgs. 3 and 5; Makarkin and Archibald 2009: Þgs. 2 and 3). The derived state is found only in the fossil genera Allorapisma and Principiala . 15. Hindwing Basal Cross-Vein r-m Shape. (0) sigmoid (Fig. 2E); (1) straight (Fig. 2B); (2) reduced or apparently absent (New 1983: Þg. 2). The shape of the

Vol. 103, no. 4

hindwing sigmoid vein was considered evidence for coalescence of MA with the radial vein in the forewing (e.g., Carpenter 1951), but now is interpreted as a cross-vein or “a cross-vein brace” (e.g., see Kukalova´Peck and Lawrence 2004: Þg. 5). The sigmoid form (0) is plesiomorphic for Neuropterida and is exhibited by Megalithone, Ithone and Polystoechotes. Stilbopteryx, Principiala , Adamsiana and Oliarces have a straight vein. In Porismus the vein is apparently absent or greatly reduced (New 1983), but it is sigmoid in other Osmylidae genera (e.g., Osmylus Latreille). Rapisma and Varnia show interspeciÞc polymorphism in this character as the vein is present or absent between species. 16. Female Wings. (0) macropterous; (1) apterous. The apterous female is autapomorphic for Adamsiana (Penny 1996). 17. Tarsal Claw Shape. (0) narrow, without projections (Riek 1974: Þgs. 1 and 2); (1) relatively broad basally with small projection (Barnard 1981: Þg. 7). The derived state of this character is autapomorphic for Rapisma (Penny 1996, Makarkin and Menon 2007). 18. Tibial Spurs. (0) present (Riek 1974: Þgs. 1 and 2); (1) absent. Tibial spurs are likely plesiomorphic throughout Neuroptera with numerous instances of secondary reduction, here exempliÞed by Myiodactylus and ingroup genera Adamsiana and Rapisma. 19. Male Genitalia, Mediuncus Shape. (0) as a lobed plate with ventral processes (Barnard 1981: Þgs. 12 and 15); (1) triangular (Acker 1960: Þgs. 58 and 60; as coxopodite 9); (2) reduced or absent (Penny 1996: Þgs. 4 and 5). Based on examinations of specimens of Oliarces in this study, it is clear that previous interpretations of the gonarcus being divided medially into two parts as illustrated by (Acker 1960: Þg. 52) are incorrect, with only a single plate present. An interpretation of the gonarcus in Ithone as triangular is incorrect as well (Barnard 1981), as the darkly sclerotized mediuncus is triangular (1) in Ithone, Megalithone, and Varnia, with a weakly sclerotized and arched gonarcus present immediately dorsal of the mediuncus. The mediuncus is a medial, lobed, platelike structure as the plesiomorphic state (0) in the outgroups, Rapisma, Fontecilla, Polystoechotes, and Platystoechotes, whereas it is weakly formed in Oliarces and Adamsiana (2). 20. Male Ectoprocts. (0) not enlarged (Penny 1996: Þg. 3); (1) enlarged (Riek 1974: Þgs. 4 and 8). The male ectoprocts are enlarged in Ithone, Megalithone, Varnia, and Oliarces. 21. Male Sternite 9. (0) simple, not elongated (Barnard 1981: Þg. 29); (1) elongated and shovel-like (Riek 1974: Þgs. 4 and 8). The derived state is exhibited by extant members of Ithonidae s. str. (i.e., Megalithone, Ithone, and Varnia). This character is not observable in the extinct species Principiala and Allorapisma . 22. Female Sternite 9 (ⴝnine gonocoxites). (0) not enlarged (Barnard 1981: Þgs. 8 and 9); (1) enlarged (Riek 1974: Þgs. 6 and 41). The enlarged female sternite 9 of Megalithone, Varnia, and Ithone forms a psammorotrum (Tillyard 1919a).

July 2010


23. Larval Jaw Shape. (0) elongate and straight for piercing prey; (1) elongate, but curved for seizing prey; (2) relatively short, with very broad base (Grebennikov 2004: Þgs. 7 and 24, 31). Elongate and straight jaws modiÞed for piercing (0) is plesiomorphic for Neuroptera (Winterton et al. 2010) and is found in Porismus; elongate and incurved apically jaws of Myiodactylus and Stilbopteryx are used for seizing prey (1). Larval stages known for Ithone, Megalithone, Polystoechotes, Oliarces, and Platystoechotes and all exhibit the derived state of short jaws with a board base (2) (MacLeod 1964, Faulkner 1990, Grebennikov 2004, Winterton et al. 2010). This type of jaw is obviously not suitable for piercing or seizing prey but instead is used for scraping or chafÞng the surface of plant roots, as observed in Ithone by Gallard (1932). All ingroup taxa were scored for this state. This type of specialized feeding biology and head morphology is a putative synapomorphy for the entire clade Ithonidae ⫹ Polystoechotidae.

Acknowledgments We thank Justin Bartlett for help with preparation of wing images. Specimens of Rapisma were collected as part of the Thailand Inventory Group for Entomological Research project supported by National Science Foundation (NSF) project DEB-0542864. We also thank Drs. John Oswald (Texas A&M University) and Norman Penny (California Academy of Sciences) for loan of specimens of Adamsiana. This research was supported by NSF project DEB-0236861 (to S.L.W.).

References Cited Acker, T. S. 1960. The comparative morphology of the male terminalia of Neuroptera (Insecta). Microentomology 24: 25Ð 83. Andersen, S. 2001. Silky lacewings (Neuroptera: Psychopsidae) from the Eocene-Paleocene transition of Denmark with a review of the fossil record and comments on phylogeny and zoogeography. Insect Syst. Evol. 32: 419 Ð 438. Archibald, S. B., and V. N. Makarkin. 2006. Tertiary giant lacewings (Neuroptera: Polystoechotidae): revision and description of new taxa from western North America and Denmark. J. Syst. Paleon. 4: 119 Ð155. Aspo¨ ck, U., J. D. Plant, and H. L. Nemeschkal. 2001. Cladistic analysis of Neuroptera and their systematic position within Neuropterida (Insecta: Holometabola: Neuropterida: Neuroptera). Syst. Entomol. 26: 73Ð 86. Barnard, P. C. 1981. The Rapismatidae (Neuroptera)Ðmontane lacewings of the Oriental Region. Syst. Entomol. 6: 121Ð136. Barnard, P. C., and T. R. New. 1985. New species in the Rapisma malayanum-complex (Neuroptera: Rapismatidae). Neuroptera Int. 3: 165Ð173, 307 (errata). Burmeister, H.C.C. 1839. Handbuch der Entomologie. Zweiter Band. Besondere Entomologie. Zweite Abtheilung. Kaukerfe. Gymnognatha. (Zweite Ha¨lfte; vulgo Neuroptera). Theod. Chr. Friedr. Enslin., Berlin, Germany. Carpenter, F. M. 1940. A revision of the Nearctic Hemerobiidae, Berothidae, Sisyridae, Polystoechotidae and Di-

521

laridae (Neuroptera). Proc. Am. Acad. Arts Sci. 74: 193Ð 218. Carpenter, F. M. 1951. The structure and relationships of Oliarces (Neuroptera). Psyche 58: 32Ð 41. Faulkner, D. K. 1990. Current knowledge of the biology of the moth-lacewing Oliarces clara Banks (Insecta: Neuroptera: Ithonidae), pp. 197Ð203. In M. W. Mansell and H. Aspo¨ ck (eds.), Advances in Neuropterology, Third International Symposium, Kruger National Park, South Africa, 3Ð 4 February 1988. Department of Agricultural Development, Pretoria, South Africa. Gallard, L. 1932. Notes on the feeding habits of the brown moth-lacewing, Ithone fusca. Aust. Nat. 8: 168 Ð170. Grebennikov, V. V. 2004. Grub-like larvae of Neuroptera (Insecta): a morphological review of the families Ithonidae and Polystoechotidae and a description of Oliarces clara. Eur. J. Entomol. 101: 409 Ð 417. Handlirsch, A. 1906 –1908. Die fossilen Insekten und die Phylogenie der rezenten Formen. Ein Handbuch fu¨ r Palaöntologen und Zoologen. W. Engelmann, Leipzig. Germany. Haring, E., and U. Aspo¨ ck. 2004. Phylogeny of the Neuropterida: a Þrst molecular approach. Syst. Entomol. 29: 415Ð 430. Kru¨ ger, L. 1922. Hemerobiidae. Beitra¨ge zu einer Monographie der Neuropteren-Familie der Hemerobiiden. Stett. Entomol. Zeit. 83: 138 Ð172. Kru¨ ger, L. 1923. Neuroptera succinica baltica. Die im baltischen Bernstein eingeschlossenen Neuropteren des Westpreussischen Provinzial-Museums (heute Museum fu¨ r Naturkunde und Vorgeschichte) in Danzig. Stett. Entomol. Zeit. 84: 68 Ð92. Kukalova´ -Peck, J., and J. F. Lawrence. 2004. Relationships among coleopteran suborders and major endoneopteran lineages: evidence from hind wing characters. Eur. J. Entomol. 101: 95Ð144. Lambkin, K. J. 1988. A re-examination of Lithosmylidia Riek from the Triassic of Queensland, Australia with notes on Mesozoic osmylid-like fossil Neuroptera (Insecta: Neuroptera). Mem. Qld. Mus. 25: 445Ð 458. Lameere, A. 1936. Precis de Zoologie. Res. Inst. Zool. Torley-Rousseau, Brussels 5: 1Ð160. Lewis, P. O. 2001. A likelihood approach to estimating phylogeny from discrete morphological character data. Syst. Biol. 50: 913Ð925. MacLeod, E. G. 1964. Comparative morphological studies on the head capsule and cervix of larval Neuroptera (Insecta). Ph.D. dissertation, Harvard University. Cambridge, MA. Maddison, W., and D. Maddison. 2000. MacClade version 4.0. Sinauer, Sunderland, MA. Makarkin, V. N., and S. B. Archibald. 2009. A new genus and Þrst Cenozoic fossil record of moth lacewings (Neuroptera: Ithonidae) from the Early Eocene of North America. Zootaxa 2063: 55Ð 63. Makarkin, V. N., and F. Menon. 2007. First record of fossil Ôrapismatid-likeÕ Ithonidae (Insecta, Neuroptera) from the Lower Cretaceous Crato Formation of Brazil. Cret. Res. 28: 743Ð753. Makarkin, V. N., and S. B. Archibald. 2003. Family afÞnity of the genus Palaeopsychops Andersen with description of a new species from the Early Eocene of British Columbia, Canada (Neuroptera: Polystoechotidae). Ann. Entomol. Soc. Am. 96: 171Ð180. McLachlan, R. 1866. A new genus of Hemerobidae, and a new genus of Perlidae. Trans. R. Entomol. Soc. Lond. 15: 353Ð354.

522


Monserrat, V. J. 1985. Lista de los tipos de Mecoptera y Neuroptera (Insecta) de la coleccio´ n L. Nava´s, depositados en el Museo de Zoologõá de Barcelona. Misc. Zool. 9: 233Ð243. Nava´ s, L. 1929a. Insectos neotropicos. 5.a serie. Rev. Chil. Hist. Nat. 33: 17Ð24. Nava´ s, L. 1929b. Insectos exo´ ticos Neuro´ pteros y aÞnes del Museo Civico de Ge´ nova. Ann. Mus. Civ. Stor. Nat. Giacomo Doria 53: 354 Ð389. New, T. R. 1983. Revision of the osmylid subfamilies Porisminae and Eidoporisminae (Insecta: Neuroptera). Aust. J. Zool. 31: 763Ð770. New, T. R. 1985. A new species of Rapismatidae (Neuroptera) from Sabah. Neuroptera Int. 3: 133Ð135. Newman, E. 1838. Entomological Notes. Entomol. Mag. Lond. 5: 168 Ð181. Newman, E. 1853. Proposed division of Neuroptera into two classes. Zoologist 11 (Appendix): clxxxiÐ cciv. Oswald, J. D. 1993. Revision and cladistic analysis of the world genera of the family Hemerobiidae (Insecta: Neuroptera). J. NY Entomol. Soc. 101: 143Ð299. Oswald, J. D. 1998a. Osmylops Banks (Neuroptera: Nymphidae): generic review and revision of the armatus species group. J. Neuropt. 1: 79 Ð108. Oswald, J. D. 1998b. Rediscovery of Polystoechotes gazullai Nava´s (Neuroptera: Polystoechotidae). Proc. Entomol. Soc. Wash. 100: 389 Ð394. Oswald, J. D. 2007. A new replacement name for Pterocalla PanÞlov, 1980 (Neuroptera: Polystoechotidae), a junior homonym of Pterocalla Rondani, 1848 (Diptera: Ulidiidae). Proc. Entomol. Soc. Wash. 109: 257Ð258. Panfilov, D. V. 1980. Novye predstaviteli setcharokrylykh (Neuroptera) iz yury Karatau [New representatives of lacewings (Neuroptera) from the Jurassic of Karatau], pp. 82Ð111. In V. G. Dolin, D. V. PanÞlov, A. G. Ponomarenko, and L. N. Pritykina (eds.), Iskopaemye nasekomye mezozoya. Naukova Dumka, Kiev, Russia. Penny, N. D. 1996. A remarkable new genus and species of Ithonidae from Honduras (Neuroptera). J. Kans. Entomol. Soc. 69: 81Ð 86. Ren, D., M. S. Engel, and W. Lu. 2002. New giant lacewings from the Middle Jurassic of Inner Mongolia, China (Neuroptera: Polystoechotidae). J. Kans. Entomol. Soc. 75: 188 Ð193.

Vol. 103, no. 4

Riek, E. F. 1974. The Australian moth-lacewings (Neuroptera: Ithonidae). J. Aust. Entomol. Soc. 13: 37Ð54. Ronquist, F., and J. P. Huelsenbeck. 2003. MrBayes 3: Bayesian phylogenetic inference under mixed models. Bioinformatics 19: 1572Ð1574. Szira´ ki, G. 1998. Female internal genitalia of some Neuroptera of phylogenetic interest. Acta Zool. Fenn. 209: 243Ð 247. Swofford, D. L. 1999. PAUP*. Phylogenetic analysis using parsimony (*and other methods), version 4.0b4a. Sinauer, Sunderland, MA. Tillyard, R. J. 1916. Studies in Australian Neuroptera. No. iv. The families Ithonidae, Hemerobiidae, Sisyridae, Berothidae, and the new family Trichomatidae; with a discussion of their characters and relationships, and descriptions of new and little-known genera and species. Proc. Linn. Soc. New South Wales 41: 269 Ð332. Tillyard, R. J. 1919a. Studies in Australian Neuroptera. No. 8. Revision of the family Ithonidae, with descriptions of a new genus and two new species. Proc. Linn. Soc. New South Wales. 44: 414 Ð 437. Tillyard, R. J. 1919b. The panorpoid complex. Part 3: the wing venation. Proc. Linn. Soc. New South Wales 44: 533Ð718. Tillyard, R. J. 1922. The life-history of the Australian mothlacewing, Ithone fusca, Newman (order Neuroptera Planipennia). Bull. Entomol. Res. 13: 205Ð223. Tillyard, R. J. 1926. The insects of Australia and New Zealand. Angus and Robertson, Sydney, Australia. Winterton, S. L. 2003. Molecular phylogeny of Neuropterida with emphasis on the lacewings (Neuroptera). Entomol. Abh. 61: 158 Ð160. Winterton, S. L., N. B. Hardy, and B. M. Wiegmann. 2010. On wings of lace: phylogeny and Bayesian divergence time estimates of Neuropterida (Insecta) based on morphological and molecular data. System. Entomol. (doi: 10.1111/j.1365-3113.2010.00521.x). Yang, C. K. 1993. The montane lacewings (Neuroptera: Rapismatidae) new to China, with descriptions of four new species. Sci. Treat. Syst. Evol. Zool. 2: 145Ð153. Received 14 February 2010; accepted 4 April 2010.