phylogenetic patterns of mimicry strategies in darnini...

Ann. soc. entomol. Fr. (n.s.), 2007, 43 (3) : 273-288

273

ARTICLE

Phylogenetic patterns of mimicry strategies in Darnini (Hemiptera: Membracidae)

Abstract. A phylogenetic analysis based on 58 morphological characters including 18 species representing 14 genera over the 15 currently known in Darnini (Hemiptera: Membracidae) confi rms the monophyly of this tribe. This result is particularly supported by the presence of cucullate setae on the ventral side of the femora. Two sister clades are inferred: the clade Funkhouseriana+ which groups four genera (Aspona, Cyphotes, Funkhouseriana, Taunaya) and exhibits a ‘bird dropping’ habitus and all other genera which exhibit a ‘dewdrop’ like habitus (Alobia, Darnis, Dectonura, Hebetica, Hebeticoides, Leptosticta, Ochrolomia, Stictopelta) or a ‘thorny’ habitus (Alcmeone, Sundarion). In the ‘dewdrop’ habitus, only the clade Ochrolomia+, is retained as a monophyletic unit. According to these results, pronotal shapes and habitus have evolved independently in each monophyletic unit and each one seems correlated with a particular type of mimicry strategy. According to the strategy, characters involved are different, a priori independent; moreover, they look coordinated regarding to the mimicry function they serve. The various evolutionary scenarios are discussed in relation to the phylogeny, and particularly in correlation with the non-gregarious behavior of these membracids, also coherent with their mimicry strategy.

Résumé. Phylogenèse des stratégies de mimétisme chez les Darnini (Hemiptera : Membracidae). Une analyse phylogénétique basée sur 58 caractères morphologiques pour 18 espèces parmi 14 des 15 genres actuellement connus de Darnini (Hemiptera : Membracidae) confi rme la monophylie de cette tribu. Ce résultat est confi rmé en particulier par la présence de soies cucullées sur la face ventrale des fémurs. Deux clades-frêres sont déduits : le clade Funkhouseriana+ qui regroupe quatre genres (Aspona, Cyphotes, Funkhouseriana, Taunaya) qui présente un aspect de «crotte d’oiseau», tandis que tous les autres genres ont un aspect de «goutte de rosée» (Alobia, Darnis, Dectonura, Hebetica, Hebeticoides, Leptosticta, Ochrolomia, Stictopelta) ou un aspect épineux (Alcmeone, Sundarion). Parmi les taxons à aspect de «goutte de rosée», seul le clade Ochrolomia+ est considéré comme monophylétique. Etant donnés ces résultats, les formes pronotales et les habitus ont évolué indépendamment dans chaque branche et chacune semble être corrélée avec un type particulier de stratégie de mimétisme. Au départ de ces stratégies, les caractères impliqués sont différents et a priori indépendants. De plus, ils paraissent coordonnés en fonction de leurs fonctions mimétiques. Les différents scénarios évolutifs sont discutés en relation avec la phylogénie et particulièrement en corrélation avec le comportement non grégaire de ces membracides, et en cohérence avec leurs stratégies mimétiques.Keywords: Mimicry strategy, phylogenetic analysis, treehoppers, solitary behavior, pronotum.

Lise Roy, Eric Guilbert* & Thierry BourgoinMuséum National d’Histoire Naturelle, Dept. Systématique et Evolution, CNRS-UMR 5202 / MNHN-USM 601, CP50,

45 rue Buff on, F-75005 Paris, France* Corresponding author

E-mail: [email protected]é le 28 juin 2007

Within the Membracoidea, the treehoppers Membracidae include nine subfamilies, 47

tribes and more than 3000 species (Dietrich 2000; Cryan et al. 2004). Only Darninae and Smiliinae appear as paraphyletic units (Dietrich et al. 2001). Th e respective position of the tribes of these two subfamilies remains in doubts. In a molecular based analysis (Cryan et al. 2000), Darnini grouped fi rst with Smiliinae and Darninae was polyphyletic. In 2001, on a morphological basis, Dietrich et al. showed

a polyphyletic Smiliinae with Darnini, Procyrtini and Hemikypthini (all Darninae) mixed together with Micrutalini and Acutalini (both Smiliinae) in the same clade. Recently, in a combined morphological and molecular analysis, Cryan et al. (2004) recognized two diff erent monophyletic clades for each Darninae and Smilliinae but the monophyly of the former could not be tested as no representatives of Hyphinoini, Procyrtini and Hemikypthini (all three Darninae) were included in the analysis. Moreover, no representative of the related subfamily Heteronotinae were tested.

Lin et al. (2004) although focused mainly on Membracinae, included more Darninae (six species, four tribes) in their analyse than any other DNA sequence-

274

L. Roy, E. Guilbert & T. Bourgoin

based analysis to date, as well as Heteronotinae. Lin et al. recovered a paraphyletic Darninae, but their analysis only included two species of Smiliinae. Th us, all the previous sequence-based analyses of Membracidae have suff ered from highly uneven and incomplete taxon sampling. Th us, the morphology-based analysis of Dietrich et al. (2001) which included representatives of all New World tribes, would seem to provide the best (although still far from defi nitive) source of information on the known status and sister-group relationships of Darnini, i.e., monophyletic sister to Hemikypthini. Unfortunately, no sequence-based analysis has included Hemikypthini.

Th e most recent work on the whole tribe Darnini was a historical review by Roy (2004). Except in this purely bibliographic study, the tribe Darnini was not studied as a whole since Goding (1929), who provided an exhaustive taxonomic review. Th e most recent taxonomic works concerning this taxon were selective studies on a single genus: Sakakibara (2005) dealt with Aspona Stål, 1862, Sakakibara & Creão-Duarte (2006) dealt with Sundarion Kirkaldy, 1904, and Creão-Duarte (1999) described the genus Funkhouseriana. McKamey published in 1998 a catalogue of the Membracoidea, where he reworded Leptosticta Butler, 1878 in the Darnini. In this catalogue, McKamey conserved the generic composition of the Darnini proposed by Deitz (1975).

Currently, Darnini include 75 species organized in 15 genera (Deitz 1975; McKamey 1998; Creão-Duarte 1999). Th eir distribution is restricted to the Neotropical and the south of Nearctic regions. Unlike most of the Membracidae, the Darnini are not aggregative but solitary, and lack any subsocial behavior. Th ey are characterized by their big pronotum, that has the same length than the body (from the front of the head to the apex of the abdomen). Two genera have suprahumeral horns: Alcmeone Stål, 1867 and Sundarion Kirkaldy, 1904. Otherwise, the shape of the pronotum is like a shell, smooth or rough, regular or battered. It covers the abdomen, and part of the wings at rest.

Membracidae are known for their hypertelic pronotal shapes but despite numerous studies (Poulton 1903; Funkhouser 1951; Ekkens 1972; Wood 1975, 1993; Stegmann 1998), there is no clear explanation of the meaning of such shapes. In Funkhouser (1951)’s opinion although the pronotum is highly specialized, it is considered as purely ornamental and with no phylogenetic signifi cance. However, Wood (1975) suggested that the membracid pronotum should be correlated with mimicry, crypsis, aposematic shape and color. For instance, Cyphotes nodosa Burmeister, 1836 looks like droppings of birds on leaves, while some green Stictopelta species resemble a lump or swelling, or a dewdrop on plant stem. Are these shapes purely hypertelic, or do they have an adaptive explanation?

Th e aim of this study is to review this question within a phylogenetic framework and relatively to mimicry. Guilbert (2001) already studied a similar case and showed that exaggerated traits in Tingidae (Insecta, Hemiptera) appearing simultaneously in three clades independently, could be related to an adaptive function. Could hyperthely of membracid pronotum be interpreted in a similar way, at least in Darnini?

Darnini represent a particular useful model to test this question. Indeed three diff erent general pronotal patterns are examplifi ed within this taxa: - the ‘bird

Figure 1 Habitus of the three diff erent general pronotal patterns defi ned: a, ‘bird dropping’ type (Cyphotes sp.); b, ‘thorny’ type (Sundarion sp.); c, ‘dewdrop’ type (Stictopelta sp.).

Mimicry strategies in Darnini

275

dropping’ type with a pronotum rough, irregular, gnarled, without horns and largely covering wings at rest; - the ‘dewdrop’ type represented by a pronotum smooth, regular drop like in dorsal view, without horns and also largely covering wings at rest; - the ‘thorny’ type with a pronotum rather smooth, with horns above the eyes, and laterally developed, slightly covering wings at rest (fi g. 1).

According to this, are these three habitus stochastically distributed in the phylogeny of the taxa, which will mean that such pattern have no phylogenetic background in agreement with a single ornamental function, or do they relate to any phylogenetical signal which will favor some adaptive function? Th erefore what should they signify relatively to mimicry?

Material and methods

Taxon sample

In this study, Darnini are represented by 18 species, selected as representative of 14 genera, from the 15 currently known

(Appendix 1). Peltosticta Sakakibara 1976 is not represented in this analysis because no specimen of this genus could have been examined. Stictopelta is represented by three species, Hebetica Stål 1869 and Cyphotes Burmeister 1835 are represented by two species each. Th ree taxa were selected as outgroups: Hemikyptha marginata (Fabricius 1781) (Darninae: Hemikypthini), Hyphinoe camelus Gray 1832 (Darninae: Hyphinoini), and Stictocephala alta (Walker 1851) (Smiliinae: Ceresini) as sister group of Darninae.

Figure 2 Pro- (a) and mesothoracic (b) femora in ventral view of Stictopelta sp. 2 showing the longitudinal groove and the four sites with or without cucullate setae (K1-9).

Figure 3 Apodems on internal face of abdominal sternite I (K16-19), in several Darnine species. Male on the left side, female on the right side: a, Hebetica convoluta; b, Hebeticoides sp.; c, Stictopelta adusta; d, Stictopelta sp.1; e, Alobia alutacea; f, Darnis lateralis; g, Sundarion sp. 2.

a

b

276


Character sample

Fifty eight morphological characters (Appendix 2) were included in the matrix (Appendix 3). Fifteen characters describe the tegumentar relief of the femora and trochanters. Th e ventral face of femora was divided into four areas where cucullate setae are distributed (fi g. 2). Four characters describe the apodems of the fi rst abdominal sternite (fi g. 3 and 4) considering each sex separately. Th ree characters concern the base of the ovipositor and the tegument of the abdominal sternites and tergites. Th e pronotum is described by eleven characters, fi ve of which describing the apophyses (fi g. 5–6) and the suprahumeral angles, one the postocular area (fi g. 7), three describing the surface, and two describing the internal architecture. Th e cephalic capsule is described by seven characters, related to the frontoclypeus (three characters), the postclypeus (one character), and the general aspect (three characters). Eighteen

characters describe the forewings, seven concern the membrane (four sites have been defi ned – fi g. 8 and 9), nine concern the veins, one concerns the cell shapes (fi g. 10 and 11), and one concerns the degree of sclerotization.

Phylogenetic analysis

For the phylogenetic analysis, all characters were treated as unordered and not weighted. A heuristic analysis was performed using Nona 2.0 (Goloboff 1998) to build tree(s) with 1000 replications and two starting trees per replication, and a search strategy of multiple TBR + TBR (Tree Bisection-Reconnection method of branch-swapping). Bremer’s decay index was calculated using TreeRot (Sorenson 1996) and PAUP* 4.0 (Swoff ord 2001). Th e trees were analysed with Winclada (Nixon 1999). Th e clade notation taxon+ indicates the clade including the taxon at its base and all subsequent taxa included in the same clade (De Souza Amorim 1982).

Pronotal patterns tested

According to the diff erent genera, Darnini examplify three main pronotal patterns (fi g. 1):-1. ‘bird dropping’ like pronotum: Aspona, Cyphotes, Funkhouseriana, Taunaya; -2. ‘dewdrop’ like pronotum: Alobia, Darnis, Dectonura, Hebetica, Hebeticoides, Leptosticta, Ochrolomia, Stictopelta; -3. ‘thorny’ pronotum : Alcmeone, Sundarion.Pronotal patterns are optimized onto the phylogeny, even if it concerns characters included in the analysis, considering that the problem of circularity concerns character coding and not character choice (Deleporte 1993).

ResultsTh e analysis provided four equiparsimonious trees

of 162 steps (CI = 0,42, RI = 0,64). Th e consensus tree is represented in fi gure 12.

Figure 4Photographs of male apodems (K16-17) in two diff erent species of Darnini: a, Hebetica sp. and b, Stictopelta sp. 2.

Figure 5 Pronotum longitudinally cutted showing the general structure of posterieor pronotal apophysis (K32-33) in Stictocephala alta (outgroup).


277

Darnini appear monophyletic. Th eir monophyly is supported by the presence of cucullate setae on the site two of the ventral face of prothoracic (character 2) and mesothoracic (K5) femora (see Deitz 1975). Cucul-late setae on site two of the mesothoracic ventral side of femora are lost in Dectonura laticauda (Fairmaire 1846). Cucullate setae on site one of ventral side of mesothoracic femora (K4) are present at the base of Darnini, but are lost in the clade Stictopelta adusta+ Burmeister 1835. In all Darnini except Alcmeone and Sundarion, the suprahumeral horns are lost (K25).

Two major clades are distinguished.- A fi rst clade (Funkhouseriana+) includes four

genera (Aspona Stål 1862, Cyphotes Burmeister 1835, Funkhouseriana Creão-Duarte 1999, and Taunaya Fonseca 1934), for which the monophyly is supported by the embossed tegument of abdominal sternites (K22), and the humeral angles of the pronotum laterally well developed in dorsal view (K26). Th e pronotal surface is rough (K29) and the postocular area of the pronotum (K28) turned back to simple in Taunaya+. Th e clade Aspona+ is supported by the presence of microscopic membranous relief on site one of the fore wings (K45).

- A second clade includes ten genera (Alobia Stål, 1869, Alcmeone Stål 1867, Darnis Fabricius 1803, Dectonura Butler 1878, Hebetica Stål 1869, Hebeticoides Fowler 1894, Leptosticta Butler 1878, Ochrolomia Stål 1869, Stictopelta Stål 1869, and Sundarion, Kirkaldy 1904), for which the monophyly is supported by the presence of cucullate setae on the site one of the ventral

side of metathoracic femora (K7). However, some reversions such as absence of such setae in S. adusta Burmeister 1835, D. laticauda (Fairmaire 1846) and H. convoluta (Fabricius 1781) are evident. Th e position of Alcmeone and Sundarion at the base of the clade is unresolved. Th e clade Ochrolomia+, including all these genera except Alcmeone and Sundarion, has a cephalic capsule subrectangular in anterior view (dorsal and ventral margins of cephalic capsule subparallel, K40), which is autapomorphic. Th is clade is divided into two clades: a small one grouping Darnis and Ochrolomia, and a larger one including the other genera. None of these two clades are supported by any apomorphy. In the latter clade, Leptosticta and Stictopelta sp.1 and sp.2 are grouped together, while Stictopelta adusta is grouped with the other genera (Alobia, Dectonura, Hebetica and Hebeticoides ). Th e monophyly of the clade (Leptosticta (Stictopelta sp. 1+ sp. 2)) is supported by the presence of a longitudinal groove in the middle of ventral face of the prothoracic and mesothoracic femora (K10 and 11), which are autapomorphic. It is also supported by the presence of cucullate setae on meso- and metathoracic trochanters (K14), features shared with Taunaya. Nevertheless, the genus Stictopelta is not monophyletic and the validity of the genus Leptosticta is doubtful. No apomorphy defi nes the clade including S. adusta, Hebeticoides sp., Hebetica sp., A. alutacea, D. laticauda and H. convoluta. In addition, the relation between these four last species remains unresolved, and the two species of Hebetica are not grouped together.

Figure 6Structure of posterior pronotal apophysis in 2 species of Darnini (K32-33): on the left, Funkhouseriana cerulea, on the right, Sundarion fl avum.

278


Discussion

Phylogenetic analysisWithin the Darninae, only Darnini and Procyrtini

have cucullate setae on the femora, however, they are located on the dorsal side in the Procyrtini. Few other Membracidae exhibit such setae on femora, but none possess cucullate setae on ventral side. Indeed, femoral

cucullate setae are located on the dorsal side of the metathoracic femora in Centronodini, Stegaspidinae Microcentrini, Smiliinae Quadrinareini, and some Smiliinae Tragopini or on the ablateral side in several tribes of Centrotinae (Platycentrini for instance). Dectonura is the exception among the Darnini in lacking cucullate setae on all femora.

Since the fi rst classifi cation of the Membracidae by

Figure 7 Postocular area in several Darnine species (K28). a, Aspona sp. (simple postocular area); b, Funkhouseriana cerulea (lobate postocular area); c, Hebeticoides sp. (carinate postocular area); d, Stictopelta sp1 (carinate postocular area).


279

Fairmaire (1846), Darnini and Smiliinae were often mixed. Th e general aspect is very similar in many genera. For instance, the pronotum, with humeral angles extending as lateral apophyses and a posterior apophysis which largely covers the wings at rest in some Darnini, like Cyphotes or Aspona, reminds one of Horiola Fairmaire 1846 and Tropidolomia Stål 1866 (Smiliinae: Tragopini). Funkhouseriana cerulea (Funkhouser 1951) resembles the genus Tynelia Stål 1858 (Smiliinae: Amastrini) with the battered pronotum with a smooth and metallic cuticule, and the cut-out parts showing the apical veins of the mesothoracic wings, bordered with wide brown smears. Other Amastrini resemble some Stictopelta sp. Moreover, Alcmeone species resemble some Smiliinae Ceresini. Such similarities might appear very superfi cial. Nevertheless, they resulted as mistakes among specialists, since the genus Cyphotes were fi rst included in Smiliinae by Funkhouser (1927), the type-species F. cerulea were fi rst included in genus Tynelia until the description of Funkhouseriana by Creão-Duarte (1999) and the species Ceresa robusta (Butler 1877) were included in Smiliinae Ceresini before being included by Andrade (1999) into the genus Alcmeone.

Indeed, many Darnini (shape type 1, 2 and 3) resemble some Smiliinae, but they are distinguishable by the mesothoracic wing venation and the femoral chaetotaxy. Th e fi rst diff erence between these two groups is situated on the forewing venation: confl uence of main longitudinal veins on basal half of the forewings (R and M in Smiliinae, M and Cu in Darninae ), and conformation of R4+5 and M1+2 veins (confl uent just before the apex in Smiliinae, distinct in Darninae). Such characters might be visible with naked eyes if examined with some attention, whereas the second basic diff erence requires some optical instrument (cucullate setae on the ventral side of at least one pair of femora: present in Darninae, absent in Smiliinae),

Figure 8Venation of Darnini’s forewings (K41-44 and 52-58). R radial vein; M median vein; CuA anterior cubital vein; CuP posterior cubital vein; A1 and A2 anal veins; lim apical limbus; cla clavus; Pl cubital crease; Cellc cubital cell.

Figure 9General view of forewings showing the four site with or without microscopic relief on membrane (K45-51).

Figure 10Asymmetrical shape of apical cell of forewing between M1+2 and R4+5, which is strongly curved (Taunaya rugosa X100) (K44).

Figure 11CuA vein bent, so that cubital cell appears subtriangular (Stictopelta sp. 2) (K43).

280


which could explain all the past and recent confusion between Darnini and some Smiliinae.

Optimizing the pronotal patterns onto the phylogeny shows two of the three main pronotal patterns as monophyletic units: a rough, irregular, gnarled, bird dropping-like pronotal habitus is observed in the monophyletic clade Funkhouseriana+ and the smooth, regular, dewdrop-like pronotum is present in the clade Ochrolomia+. Nevertheless, the third pattern: a rather smooth, “thorny” pronotum characterizing Alcmeone and Sundarion does not refl ect a monophyletic unit.

Accordingly, the corresponding clades are not only defi ned by their pronotal shape, but also by additional characters. Th e embossed tegument of abdominal sternites (K22) is autapomorphic for Funkhouseriana+ and the pronotum is battered and often covered with rugae, and has humeral angles more or less laterally extended.

Th e subrectangular cephalic capsule in anterior

view (K40) is synapomorphic for Ochrolomia+ except for the two basal genera Alcmeone and Sundarion. In these genera, the pronotum is smooth, shiny and with a regular surface, resembling a drop, and covering part of the wings. Th e humeral angles are not laterally extended.

Ochrolomia was considered as a subgenus of Darnis by Stål (1869) but raised to generic rank by Goding (1894). Funkhouser (1951) noted the great similarity between the two genera. In this clade, Ochrolomia and Darnis are grouped together and basal to all others. Th e only character distinguishing the two genera is the ratio of the distance between ocelli and the distance between the eye and the nearest ocellus (Goding 1894). Nevertheless, this character is not clearly diagnostic for all specimens of the genera examined here. A revision of both genera is needed to determine their composition and monophyly.

Stictopelta can be divided into two groups depending on the femoral characters, the postocular pronotal

Figure 12 Consensus tree of four equiparcimonious trees of 162 steps (CI=42, RI=63), produced by NONA using Fitch parsimony with 1000 replicates and 2 starting trees per replicate, and a search strategy of multiple TBR + TBR resulting from a heuristic analysis of the 21 taxa described by 58 characters. Th e numbers below nodes refer to the Bremer’s decay index. Th e three groups distinguished on the basis of their pronotal patterns are precised.


281

area, the vertex and general pronotal shapes. A fi rst group with a slender pronotum includes sp. 1 and sp. 2, and a second group with a more stocky, compact pronotum, includes S. adusta. Such a division explains why S. adusta is not included in the clade with the other Stictopelta species. Nevertheless, the systematics of the genus Stictopelta will not be discussed here.

Th e monospecifi c genus Leptosticta was created by Stål (1869) as subgenus of Stictopelta, and was raised to generic level by Butler (1878). Frequently treated as a synonym of Stictopelta (Metcalf & Wade’s 1965 catalog), it was reinstated as valid genus by Sakakibara (1976). Here, Leptosticta is included in a clade comprising Stictopelta sp. 1 and sp. 2. According to our results this species should be included in Stictopelta, but a detailed revision is needed to determine the status of Leptosticta.

Th e relationships between the more derived

species of the group remains uncertain. Dectonura is monospecifi c, and D. laticauda is known only by the type specimen. It is here sister group of Hebetica convoluta. In addition, Hebetica and Hebeticoides, supposed closely related genera according to Fowler (1894), are not grouped together. Hebeticoides is here sister group of Alobia, Dectonura and Hebetica. Alobia is also monospecifi c, and its position remains unresolved.

To be complete, it is to be noted that Dectonura and Leptosticta were both invalidated (Funkhouser 1951 and Goding 1929, respectively) and put in synonymy, before being reinstated as genus by Metcalf and Wade (1965) and Sakakibara (1976), respectively.

Phylogenetic relationships within the ‘thorny’ group (the third group represented by Alcmeone and Sundarion) remain uncertain. It does not form a monophyletic group.

Figure 13 Optimization of the characters involved in the fi rst scenario presented. Circles represent fast optimization, and dots represent slow optimization.

282


Evolutionary hypothesisTwo diff erent scenarios can be drawn, considering

the two major clades of the Darnini. A fi rst scenario involves the fi rst clade Funkhouseriana+. A second scenario involves the second clade excluding Alcmeone and Sundarion: the Ochrolomia+ group. Th e characters involved in these scenarios refl ect the evolution of the habitus toward two of the three general patterns observed in the Darnini (table 1).

First scenario. Five characters are involved in the evolution of this group (Figure 13). Th ree characters concern the shape and the position of the vertex and the frontoclypeus. Th e frontoclypeus is distinctly delimited in the vertex in F. cerulea+ group (K35). Th is character is plesiomorphic and shared also with Sundarion, in deltran optimization. It is a reversal in this clade and in Sundarion with acctran optimization. Th e subparallel position of the vertex and the frontoclypeus to the body axis (K34) is apomorphic in the group (Cyphotes nodosa + Cyphotes sp.). Cyphotes

species clearly exhibit this state while it is not as visible in Aspona. Also, the inferior part of the frontoclypeus is prominent in Funkhouseriana+ (K36), while it is truncate in Ochrolomia+. Th is character state is shared with Alcmeone and Sundarion. Th e fourth character is the rough texture and battered relief of the pronotum (K29), which is apomorphic to the group (Aspona, Cyphotes, Taunaya). F. cerulea has a battered pronotum but a smooth pronotal surface. Th e fi fth character is the modifi cation of the internal cuticular leaf of the posterior pronotal apophysis (K33). Aspona, Cyphotes, and Taunaya have a membranous internal leaf and the area not clung are reduced and ramifi cated (it is not clung only along some grooves of the external leaf ). Th is internal leaf is not as sclerotized as the external leaf in F. cerulea, but it is not membranous and is not clung to the external leaf in almost the entire apophysis. Th e outline is more and more jumbled by the battering and the rugae of the pronotal surface, which becomes larger; and the body has a tendency

Figure 14 Same as fi gure 8 for the second scenario. Th e optimization does not change whether it is deltran or acctran.


283

towards a fl atness by changing the orientation of the vertex and frontoclypeus, which might give them a weevil-like appearance. Th us, species such as Cyphotes nodosa resemble bird droppings.

Second scenario. Six characters are involved in the evolution of this group (fi gure 14). Th e covering of the wings by the pronotum at rest becomes more important through the clade Stictopelta+ (K23). Th e wings are almost uncovered in Alcmeone and Sundarion, at the base of this clade. Th e covering is reduced in Darnis and

Ochrolomia. It is moderate in Leptosticta, Stictopelta sp. 1, Stictopelta adusta, and Hebeticoides. It is developed at the apex of the group in Stictopelta sp. 2, Hebetica and Alobia. At the same time, the width of the cephalic capsule increases and its height decreases in anterior view (K38) for the more distal species (S. adusta+ group). Darnis, Ochrolomia, Stictopelta, and Leptosticta have an ovoid vertex, while Hebetica, Hebeticoides, and Alobia have a rectangular vertex (K38). Th e shape in S. adusta is midway between these. In addition, Darnis

Table 1. Summary of the characters involved in the two scenarios of mimicry strategy, and their optimization.

Autapomorphy for Homoplasy ReversionChar. 1 and 4 Cucullate setae on site 1 of the ventral face of the profemora and the mesofemora

Whole DarniniAbsent in Hebeticoides sp. + clade

Char. 9 Cucullate setae on site 3 of the ventral face of the metafemora Present in the whole Ochrolomia + clade

Absent in D.laticauda and in Stictopelta sp . 1

Char. 12 and 13 Ventral crest extending the small lobe of the apex of the profemora and the mesofemora towards the base absent (0); present and reaching the middle of the femora (1); present all along the femora (2)

Present all along femora in L. flaviceps+ clade and partly present on femora in S. adusta+ clade

Abd

omen

Char. 22 Tegument of abdominal sternitesRegularly embossed with round pits in the whole clade F. cerulea + (state =1)

Char. 23 Pronotum coverage of fore wings

Partly covering to largely covering in the F.cerulea + clade and the Ochrolomia+ clade

Char. 26 Humeral angles in dorsal viewLaterally developed in the whole clade F.cerulea +

Char. 28 Postocular pronotal areaSimple in the T. rugosa + clade (ie for the whole F. cerulea + clade except the basal OTU)

Char. 29 Rugae on the pronotal surface Present in T. rugosa + clade (ie for the whole F. cerulea + clade except the basal OTU)

Char. 33 Internal tegument of bitegumentar apophysis

Membranous in T. rugosa+ clade and in Ochrolomia + clade

Char. 34 Position of vertex and frontoclypeus according to body axis Subparallel in genus Cyphotes

Char. 35 Frontoclypeus delimitation in vertex Not distinct in genus Cyphotes

Char. 36 Inferior part of frontoclypeus (in anterior view) )

Truncate in the whole clade Ochrolomia+ and in C. nodosa

Char. 38 Lateral height of vertex in comparison with median height of vertex, including frontoclypeus

Of equal height for S. adusta +

Char. 40 Dorsal and ventral margins of cephalic capsule Subparallel in Ochrolomia + clade

Win

gs Char. 44 Apical cell of fore wings between R4+5 and M1+2 Asymmetrical the whole F. cerulea + clade

Bird dropping' like and 'dewdrop' like habitus

Dewdrop' like habitus

Bird dropping' like habitus

Pron

otum

Cep

halic

cap

sule

Table 1 : Autapomorphic characters which are involved in

Involved in mimesis strategies

Legs

284


and Ochrolomia, which are basal to the other genera mentioned above, and Leptosticta, Stictopelta sp. 1 and S. sp. 2, have a double regular line of cucullate setae on ventral face of pro- and mesothoracic femora (K1 and 4). Th e ventral crest extending the small lobe of the apex of the pro- and mesothoracic femora towards the base (K12 and 13) is absent in these former taxa, except in Stictopelta sp. 1 and S. sp. 2 where it is present all along the femora. It is also absent in S. adusta, but becomes shorter and smaller in Hebetica, Hebeticoides, Alobia, and Dectonura. It disappears on prothoracic femora in H. convoluta. Th e frontoclypeus becomes less prominent by widening and fl attening of the vertex, thus allowing these species to hug the substrate. Th e swollen metopidium hides the vertex better when the latter is fl attened, and then the pronotum, resembling a drop or a seed could have a mimetic function. In addition to these attributes, femora of some species such as Leptosticta and Stictopelta have grooves which may hide tibiae, or crests which hold tibiae close to femora. Th ese adaptations allow the pronotum to hide the legs at rest.

Comparison of the two mimetic strategies On the whole, the characters involved in the

two mimetic strategies are not the same, even if some character states are shared and involved in the two scenarios (Table 1). Th e posterior apophysis of pronotum is almost completely covering wings at rest in both Funkhouseriana+ monophyletic unit and most of Ochrolomia+ taxa (K23), the membranous internal tegument of apophysis is in reversion in Funkhouseriana+, so that it becomes the same as in Ochrolomia+ (K33), and the inferior part of the frontoclypeus is truncate in Ochrolomia+ (K36) and in C. nodosa. Nevertheless, the two sets of characters have changed coherently in each group. Characters 1, 4, 9, 12, 13, 23, 33, 36, 38 and 40 are involved in the mimetic strategy of the Ochrolomia+ clade, whereas characters 22, 23, 26, 28, 29, 33, 34, 35, 36 and 44 are involved in the Funkhouseriana+ unit. Th e evolution that seems to characterize the two groups produces the same eff ect (camoufl age), through diff erent ways. Both groups become fl atter, by changing the vertex relative orientation (Funkhouseriana+, K34) or by truncating it (Ochrolomia+, K40). Th e cephalic capsule and the legs become invisible in both groups. In the Funkhouseriana+ group, the outline is blurred on its whole by imprecise, hazy contours (K26, 28 and 29), which gives the insect the appearance of a fuzzy little mass. Th e cephalic capsule and legs are visible but not easily identifi able because of the lack of sharp contours. In contrast, among the Ochrolomia+ group, the outline is precise, and has sharp contours. Th e pronotum is relatively growing all along the clade:

the metopidium becomes higher and globular and the posterior apophysis covers better part of the wings (K23) in intermediate and distal taxa. Th e frontoclypeus becomes less conspicuous: it is fl atter than in the other clade in anterior view (K36 and 40), and its relative height decreases also, compared with its width, in distal taxa, whereas it stays perpendicular to body axis (K34). Th e pronotum becomes so conspicuous and the cephalic capsule so fl at in anterior view and invisible in dorsal view, that one can see only the dewdrop-shaped or seed-shaped pronotum. One can almost distinguish neither the cephalic capsule under it nor the legs, which are closely held in place and are situated mostly under the body at rest.

Mimicry and subsocialityIn treehoppers that are known to be generally

subsocial, the mimetic function is often enhanced by their aggregative habits. For instance, many Membracis sp. (Membracinae) colonies look like fl owers around and along a plant stem. Th e vertically fl attened and brilliantly coloured habitus of such species allows a predator to easy detect an isolated specimen. Th e grouping of numerous specimens on a stem, each in a characteristic position, allows them to be confused with clusters of fl owers. Th e same eff ect can be seen in colonies of treehoppers of genus Umbonia (Membracinae), which resemble certain plant thorns (Cocroft 1999a,b).

On the contrary, Darnini species are not known to be subsocial. Th ey cannot benefi t from the camoufl age eff ect of aggregation. Th eir probable mimicry must be eff ective for each separate individual. In the two scenarios discussed above, supposed mimed objects are non-living forms (bird-droppings, dewdrops) or, if living, mimed forms seem not to be a (normal) part of the plant (blister on a leaf or a stem, seed), which are not regularly repeated in nature.

Nevertheless, unlike aggregative treehoppers above cited, the animal aspect seems to be individually blurred in Darnini. Legs are almost invisible in both Darnini scenarios, whereas they are visible in Membracis sp. and Umbonia sp. Forewings are largely covered at rest in Darnini, whereas they are always visible in Membracis sp. and Umbonia sp. Th e repetition of the same shape can be enough to make body parts such as legs and forewing inconspicuous in Membracis sp. and Umbonia sp. In Darnini, the blurring is more individual.

ConclusionTh e phylogeny which is suggested here involves 18

species, which are representative of each of the 14 tested genera, logically selected among the 75 species included


285

in the tribe. Th e genus Peltosticta is not represented. Some genera such as Leptosticta and Stictopelta are in need of a revision. Th us, the evolutionary hypothesis must be improved with more taxa. Nevertheless, we believe that the evolutionary trends concerning the mimicry adaptation would remain the same. Moreover, as many Darnini resemble species of other tribes and subfamilies in pronotal shape, the same shapes could have arisen independently multiple times in the evolution of the Membracidae. Such a trend supports the hypothesis that the pronotum of these groups has an adaptive function, leading to evolutionary parallelism. A more powerful test of the hypothesis of adaptive function could incorporate these multiple independent lineages of membracids with similar pronotal shapes. Th e scenarios of adaptation to mimicry in Darnini could therefore partially explain the diversity of pronotal shapes in Membracidae.Acknowledgments. We are grateful for the fi nancial support to the Société Entomologique de France (Bourse Germaine Cousin), for his help and advices to Dr Lewis L. Deitz (North Carolina State University, Raleigh), and for the loan of some specimens to Dr Stuart McKamey (Systematic Entomology Laboratory, United States Department of Agriculture, Washington).

ReferencesAndrade G.S. de 1999. Nomenclatural and taxonomic notes on

Alcmeone robusta (Butler) comb.n. (Hemiptera, Auchenorrhyncha, Membracidae). Revista Brasileira de Zoologia 16: 267-268.

Bates H. W. 1862. Contributions to an insect fauna of the Amazon valley, Lepidoptera : Heliconidae. Transaction of the Linnean Society of London 23: 495-566.

Benjamin S. P. 2004. Taxonomic revision and phylogenetic hypothesis for the jumping spider subfamily Ballinae (Araneae, Salticidae). Zoological Journal of the Linnean Society 142: 1-82.

Brower A. V. Z. 1995. Locomotor mimicry in butterfl ies? A critical review of the evidence. Philosophical Transaction of the Royal Society of London B 347: 413-425.

Brower A. V. Z. 1996. Parallel race formation and the evolution of mimicry in Heliconius butterfl ies : a phylogenetic hypothesis of mitochondrial DNA sequences. Evolution 50: 195-221.

Butler A.G. 1878. On various genera of the homopterous familly Membracidae, with description of new species and a new genus in the collection of the British Museum. Systematic Entomology 2: 337-361.

Cocroft R. B. 1999a. Off spring-parent communication in a subsocial treehopper (Hemiptera: Membracidae: Umbonia crassicornis). Behaviour 136: 1-21.

Cocroft R. B. 1999b. Parent-off spring communication in response to predators in a subsocial treehopper (Hemiptera: Membracidae: Umbonia crassicornis). Ethology 105: 553-568.

Creão-Duarte A. J. 1999. Funkhouseriana, um novo gênero de Darninae (Hemiptera, Membracidae). Revista Brasileira de Zoologia 16: 905-908.

Cryan J. R., Wiegmann B. M., Deitz L. L., Dietrich C. H. 2000. Phylogeny of the treehopper (Insecta: Hemiptera: Membracidae): evidence from two nuclear genes. Molecular Phylogeny and Evolution 21: 72-85.

Cryan J. R., Wiegmann B. M., Deitz L. L., Dietrich C. H., Whiting M. F. 2004. Treehopper trees: Phylogeny of Membracidae (Hemiptera: Cicadomorpha: Membracoidea) based on molecules and morphology. Systematic Entomology 29: 441-454.

Deleporte P. 1993. Characters, attributes, and tests of evolutionary scenarios. Cladistics 9: 427-432.

De Souza Amorim D. 1982. Classifi caçao por sequenciaçao : Uma proposa para a denominaçao dos ramos retardados. Revista Brasileira de Zoologia 11: 1-9.

Deitz L. L. 1975. Classifi cation of the higher categories of the New World treehoppers (Homoptera: Membracidae). North Carolina Agriculture Experimental Station Technical Bulletin 225: 1-177.

Dietrich C.H. 2000. Treehoppers (Hemiptera: Cicadomorpha: Membracoidea). http://www. inhs.uiuc.edu/~dietrich/treehome.html.

Dietrich C. H., McKamey S. H., Deitz L. L. 2001. Morphology -based phylogeny of the treehopper family Membracidae (Hemiptera: Cicadomorpha: Membracoidea). Systematic Entomology 26: 213-239.

Ekkens D. 1972. Peruvian treehopper behavior (Homoptera: Membracidae). Entomological News 83: 257-271.

Fairmaire L. 1846. Revue de la tribu des Membracides. Annales de la Socièté Entomologique de France 4(2ème série): 235-320.

Fowler W. W. 1894. Class Insecta, orderRhynchota, suborder Hemiptera-Homoptera (continued), fam. Membracidae. Biologia Centrali Americana 2: 1-173.

Funkhouser W. D. 1927. General catalogue of the Hemiptera. Fasc. I: Membracidae. Smith College, Northampton, Mass., 581 p.

Funkhouser W. D. 1951. Homoptera fam. Membracidae. Genera Insectorum 208: 1-383.

Goding F. W. 1894. Bibliographical and synonymical catalogue of the described Membracidae of the North America. Illinois Natural History Survey Bulletin 3: 391- 482,

Goding F. W. 1929. Th e Membracidae of South America and the Antilles (IV): subfamilies Hoplophorioninae, Darninae, Smiliinae, Tragopinae (Homoptera). Transaction of the American Entomological Society 55: 197-330.

Goloboff P. 1998. Nona. Ver. 2.0. [Computer softare] Fundacion e Instituto Miguel Lillo, Miguel Lillo 205, 4000 S. M. de Tucuman, Argentina.

Guilbert E. 2001. Phylogeny and evolution of exaggerated traits among the Tingidae (Cimicomorpha, Heteroptera). Zoologica Scripta 30: 313-324.

Lin C. P., Danforth B. N., Wood T. K. 2004. Molecular phylogenetics and evolution of maternal care in Membracine treehoppers. Systematic Biology 53: 400-421.

McKamey S. H. 1998. Taxonomic catalogue of the Membracoidea (exclusive of leafhoppers). Memoirs of the American Entomological Institute 60: 1-177.

Metcalf Z. P., V. Wade 1966. General Catalogue of the Homoptera. Bibliography of the Membracoidea and Fossil Homoptera (Homoptera: Auchenorhyncha) (1963). Suppl. to Fasc. 1 Membracoidea (1965) and A Catalog of the Fossil Homoptera (1966) North Carolina State College, Raleigh.

Miller J. S. 1996. Phylogeny of the Neotropical moth tribe Josini (Notodontidae: Dioptinae): a hidden test of Müllerian mimicry. Zoological Journal of the Linnean Society 118: 1-45.

Müller F. 1879. Ituna and Tyrinis; a remarkable case of mimicry in butterfl ies. Transaction of the Entomological Society of London 1879: 20-24.

Nixon K. C. 1999. Winclada (beta) ver. 0.9.9.9. Published by the author, Ithaca, N.Y.

Pasteur G. 1982. A classifi catory review of mimicry systems. Annual Revue of Ecology and Systematics 13: 169-199.

Poulton E. B. 1903. Suggestion as to the meaning of the shapes and colours of the Membracidae in the struggle for existence, p. 273-285 in: Buckton G. B. (ed.), Monograph of the Membracidae. Lowell Reeve, London.

Roy L. 2004. Revue historique des Darnini (Hemiptera : Membracidae) au niveau générique. Notes fauniques Gembloux 51: 33-47.

Sakakibara A. M. 1976. Gênero e espécies novas de Darnini (Homoptera, Membracidae) Revista Brasileira de Biologia 36: 505-511.

Sakakibara A. M. 2005. Th e species of Aspona Stål and nomenclatural notes (Hemiptera, Cicadomorpha, Membracidae) Revista Brasileira de Entomologia 49: 462-464.

286


Sakakibara A. M., Creão-Duarte A. J. 2006. Th e species of membracids described by C. Berg and M. Spinola classifi ed in Sundarion Kirkaldy (Hemiptera, Cicadomorpha, Membracidae) Revista Brasileira de Entomologia 50: 31-32.

Simmons R. B., Weller S. J. 2002. What kind of signal do mimetic tiger moths send? A phylogenetic test of wasp mimicry systems (Lepidoptera: Arctiidae: Euchromiini). Proceedings of the Royal Society of London B 269: 983-990.

Sorenson M. D. 1996. TreeRot. University of Michigan, Ann Arbor.Stål C. 1869. Hemiptera Fabriciana. Fabricianska Hemipterarter, efter de i

Köpehnhamn och Kiel förvarade typexemplaren granskade och beskrifne. 2. Kongeiga Svenska Vetenskaps-Akademiens Handlingar 8: 1-130.

Stegmann U. E. 1998. An exaggerated trait in insects: Th e prothoracic skeleton of Stictocephala bisonia (Auchenorrhyncha: Membracidae). Journal of Morphology 238: 157-178.

Swoff ord D. L. 2001. PAUP* - Phylogenetic Analysis Using Parsimony (*and other methods). Version 4.0B 10. Sinauer Associates, Sunderland, Massachusetts.

Wood T. K. 1975. Studies of the function of the Membracid pronotum (Homoptera) II. Histology. Proceedings of the Entomologial Society of Washington 77: 78-81.

Wood T. K. 1993. Diversity in the new world Membracidae. Annual Revue of Entomology 38: 409-435.

Yen S. H., Robinson G. S., Quicke D. L. J. 2005. Phylogeny, systematics and evolution of mimetic wing patterns of Eterusia moths (Lepidoptera, Zygaenidae, Chalcosiinae). Systematic Entomology 30: 358-397.

Appendix 1. List of taxa examinedTaxa not included in the matrix are indicated by (*)

Alcmeone Stål 1867 A. lata (Walker 1851)* A. picea (Fairmaire 1846)Alobia Stål 1869 A. alutacea Stål 1869Aspona Stål 1862 Aspona sp.Cyphotes Burmeister 1835 C. nodosa Burmeister 1835 Cyphotes sp.1 Cyphotes sp.2* Cyphotes sp.3*Darnis Fabricius 1803 D. lateralis Fabricius 1846Dectonura Butler 1878 D. laticauda (Fairmaire 1846)Funkhouseriana Creão-Duarte 1999 F. cerulea (Funkhouser 1951)Hebetica Stål 1869 H. convoluta (Fabricius 1781) Hebetica sp.Hebeticoides Fowler 1894 Hebeticoides sp.Leptosticta Butler 1878 L. fl aviceps Butler 1878Ochrolomia Stål 1869 Ochrolomia sp. O. suturalis (Germar)*

Peltosticta Sakakibara 1976 P. yonkei Sakakibara 1976*Stictopelta Stål 1869 S. adusta Burmeister 1835 Stictopelta sp.1 Stictopelta sp.2 Stictopelta sp.3Sundarion Kirkaldy 1904 Sundarion sp.1 Sundarion sp.2*Taunaya Fonseca 1934 T. rugosa Fonseca 1934

Appendix 2. List of characters and states used in the analysis

Respective defi nitions of the various areas of chaetotactic characters as recognised and used in the phylogenetical analysis for the ventral side of femora and for some microscopic relief on forewings:

Femora (K1-9, cf. fig. 2):Site 1: basal quarter of ventral side of femurSite 2: area surrounded by the basal limit of the articular area

(site 3) and the distal limit of site 1Site 3: articular area, which is located just below the femoro-

tibial articulation. It forms a fovea, containing a part of the unstretched tibia, extending more or less forward base of the femur and laterally surrounded by some carenae and the apical lobe.

Site 4: chitinous apophysis, usually translucent, rounded, anteroventrally located, at the apex of femur and covering partially the base of the tibia, when unstretch.

Forewing (K46-51, cf. fig. 9)Site 1: Apical limbusSite 2: Area located just above the cubital creaseSite 3: Area located just below the cubital creaseSite 4: Apex of the cubital cell, between CuA and CuP,

anterad to the limbus crease.

Characters1. Cucullate setae on site 1of the profemora absent (0);

present (1);2. Cucullate setae on site 2 of the profemora absent (0);

present (1);3. Cucullate setae on site 3 of the profemora absent (0);

present (1);4. Cucullate setae on site 1 of the mesofemora absent (0);



present (1);7. Cucullate setae on site 1 of the metafemora absent (0);



present (1);10. Longitudinal groove on the ventral face of the profemora

absent (= profemora section rounded) (0), present in mid part profemora (1), present only in distal part of the profemora (2)


287

11. Longitudinal groove on the ventral face of the mesofemora absent (0), present in mid part mesofemora (1), present only in distal part of the mesofemora (2);

12. Ventral crest extending the small lobe of the apex of the profemora towards the base absent (0); present and reaching the middle of the femora (1); present all along the femora (2);

13. Ventral crest extending the small lobe of the apex of the mesofemora towards the base absent (0); present and reaching the middle of the femora (1); present all along the femora (2);

14. Cucullate setae on mesotrochanters absent (0); present (1);15. Cucullate setae on metatrochanters absent (0); present (1);16. Apodem length of fi rst abdominal sternite (cf. fi g. 3–

4) in males reaching at most half the length of sternite II (0); reaching distal boundary of sternite II (1); extending beyond distal boundary of sternite II (2);

17. Apodem of fi rst abdominal sternite (cf. fi g. 3) in males parallely oriented to the body axis (0); perpendicularly oriented to the body axis (1); in an intermediate position between parallel and perpendiculate (2);

18. Apodem length of fi rst abdominal sternite (cf. fi g. 3) in females reaching at most half the length of sternite II (0); reaching distal boundary of sternite II (1); extending beyond distal boundary of sternite II (2);

19. Apodem of fi rst abdominal sternite (cf. fi g. 3) in females parallely oriented to the body axis (0); perpendicularly oriented to the body axis (1); in an intermediate position between parallel and perpendiculate (2);

20. Two tegumentar expansions from posterior part of sternite V and covering base of ovipositor absent (0); present (1);

21. Sclerotization of abdominal tegument homogeneous (0); heterogeneous (1). We consider homogenous either a strong or a weak sclerotization on all the abdomen and heterogenous a sclerotization markedly stronger on sternites and laterotergites than on tergites;

22. Tegument of abdominal sternites smooth (0); regularly embossed with round pits (1);

23. Pronotum not or slightly covering fore wings when at rest (0); partly covering fore wings (not reaching costal cells) (1); almost entirely covering fore wings (reaching costal cells or costa) (2);

24. Posterior pronotal apophysis with round apex and without air pocket (0); with subtriangular apex and without air pocket (1); with subtriangular apex and an air pocket (2);

25. Suprahumeral horns absent (0); present (1);26. Humeral angles in dorsal view simple (0); laterally

developed (1);27. Humeral angles laterally developed acute (0);

auriculiform (1);28. Postocular pronotal area (cf. fi g. 7) simple (0); carinate

(1); lobate (2). A simple postocular area is not prominent at all (fi g. 5a), a carinate one is is slightly prominent laterally, with a ventral carina, but does not extend below the eye (fi g. 5c, d) and a lobate one has a little lobe extending slightly above the eye (fi g. 5b);

29. Rugae on the pronotal surface absent (0); present (1);30. Pronotal surface without setae (0); with numerous setae

(1);31. Pronotal surface strongly punctate (0); slightly punctate

(1). Pronotal surface is considered strongly punctate when pits are so large and/or deep that most part of the surface is situated in the pits. On the contrary, a slightly punctate pronotal surface is mostly situated between pits;

32. Posterior pronotal apophysis (cf. fi g. 6) monotegumentar (0); bitegumentar (1) : Th e tegument of the posterior apophysis is composed of two leaves, which stick in some areas one to the other, and, in other areas, are separated one from the other by an air layer. We described the former case as “monotegumentar”, since the two leaves are distinct only one in such areas, and the

latter case as “bitegumentar”. Note that the inferior leaf is usually membranous or at least not as sclerotized as the superior one.

33. Internal tegument of bitegumentar apophysis sclerotized (0); membranous (1);

34. Vertex and frontoclypeus perpendicular to body axis (0); subparallel to body axis (1);

35. Frontoclypeus not distinctly delimitate in vertex (0); distinctly delimitate in vertex (1). Th e frontoclypeus is more or less markedly visible in the middle of the vertex. It is considered distinctly delimitate when the suture is conspicuous, not fused in the vertex, making the inferior margin discontinuous. It appears not distinctly delimitate when the suture is so reduced that it seems fused into the vertex;

36. Inferior part of frontoclypeus (in anterior view) truncate (0); prominent (1);

37. Postclypeal carina straight (0); sinuate (1). Th e postclypeal carina is ventrally and longitudinally situated at the base of the rostrum and may be straight or sinuate in lateral view;

38. Lateral height of vertex in comparison with median height of vertex, including frontoclypeus, of equal height (0); half to two third the median height of vertex, including frontoclypeus (1);

39. Frontoclypeus narrow (a third the width of vertex) (0); frontoclypeus wide (more than half the width of vertex) (1);

40. Dorsal and ventral margins of cephalic capsule subparallel (0); dorsal margin straight, while ventral margin convex (1);

41. Apex of CuA and CuP vein joined at the base of cubital crease on a simple point (0); separate by a small vein (1);

42. Area strongly sclerifi ed at the base of forewings absent (0); present (1);

43. CuA vein (cf. fi g. 11) straight (0); bent (1);44. Apical cell of fore wings between R4+5 and M1+2

symmetrical (0); asymmetrical, because of strong curvature of R4+5 (cf. fi g. 10) (1);

45. Microscopic relief on membrane on site 1 of fore wings absent (0); present (1);

46. Microscopic relief on membrane on site 1 of fore wings squamiform (0); conical (1);






52. Fork of radial vein with two nodes (0); one node (1); three nodes (2);

53. R vein bifurcating between R1+2+3 and R4+5 (0); between R1 and R2+3+4+5 (1):

54. Transversal vein in the fork of R vein of fore wings absent (0); present (1);

55. Position of transversal vein in the fork of r vein basal to r-m vein (0); distal to r-m vein (1);

56. Confl uence of M and Cu on basal third of forewings simply joining (0); intermingled (1);

57. Confl uence of M and Cu linear (0); sinuate (1);58. Confl uence of M and Cu more than a third the length

of the wing (0); less than a quarter the length of the wing (1).

288


Appendix 3. Matrix of 21 taxa and 58 characters used in the analysis

None are ordered neither weighted.

Species Characters numbers

1234567891111111111222222222233333333334444444444555555555

0123456789012345678901234567890123456789012345678

Alcmeone picea 11011011000000011??–100110–200110001010111010–111100011001

Alobia alutacea 01001111?00110012120102200–10100–000000001100–0–0–0010–101

Aspona sp. 110011010000000????101120110111110110101?11111110–1001––––

Cyphotes nodosa 110110010000000??110?1220100111?11100111111110100–00010111

Cyphotes sp.1 11011?110000000??12101220110111111110111111110101011–1––––

Darnis lateralis 11011011100000011111101000–20000–0000100110?0–0–0–0001–001

Dectonura laticauda 00?00?00001110010??–00?200–10001100000001100???????010––––

Funkhouseriana cerulea 1101110000111????11?1112010200010?11010111110–111100010001

Hebetica convoluta 010011011010100200–?002000–201111000?0001?000–0–0–0010–110

Hebetica sp. 01?01111100110020??–00?000–2011??000100000000–0–0–0010––––

Hebeticoides sp. 01?01?111111100100–1?01000–20010–000?00011000–0–0–0–010–––

Hemikyptha marginata 00000000000000012121100110–200110001010111000–110–00011001

Hyphinoe camelus 0000000000000001212–0?0110–2011100110111110?0–0–110––?1?01

Leptosticta fl aviceps 11?11?111220011??0–0101000–200111000110000100–0–0–00111100

Ochrolomia sp. 11011011100000001??–101000–20000–000010011000–0–0–0001–001

Stictocephala alta 000000000000000120–?000110–001110011110110010–0–0–02–––001

Stictopelta adusta 010010011000000100–0201200–10010–000000010100–0–111?010100

Stictopelta sp.1 1111111102222112010?001000–1001??000110010010–0–0–000111–0

Stictopelta sp.2 1101101112222111202–?02000–10000–000110010100–0–0–0010–100

Sundarion sp. 110110110000000200–0000110–200110011010111000–111100011001

Taunaya rugosa 11?11000000001111??–1112010010111011010111110–11111001–011

phylogenetic patterns of mimicry strategies in darnini...

Documents