Comparative study of the morphology of stridulatory organs of the Iberian lycaenid butterfly pupae (Lepidoptera)

Comparative Study of the Morphology of StridulatoryOrgans of the Iberian Lycaenid Butterfly Pupae(Lepidoptera)

Margarita �Alvarez,1* Miguel L. Munguira,2 and Mar�ıa Dolores Mart�ınez-Ib�a~nez3

1�Area de Ciencias, Escuela Universitaria de Magisterio ESCUNI, Avenida Nuestra Se~nora de F�atima 102, ES-28047-Madrid, Spain2Departamento de Biolog�ıa, Universidad Aut�onoma de Madrid, Cantoblanco, ES-28049-Madrid, Spain3Departamento de Zoolog�ıa y Antropolog�ıa F�ısica, Facultad de Biolog�ıa, Universidad Complutense de Madrid,ES-28040 Madrid, Spain

ABSTRACT We compared the pupal stridulatoryorgans of 35 species and one subspecies of IberianLycaenidae using scanning electron microscopy. Thestudied species belong to the tribes Theclini, Eumaeini,Lycaenini, and Polyommatini. Nine species do not showstridulatory organs on the pupae but all other speciespossess them. Stridulatory organs are formed by astridulatory plate (pars stridens) placed on the fifthabdominal segment and a file (plectron) in the sixthabdominal segment. The plate has tubercles in the The-clini and Lycaenini, tubercles, ridges, or undulations inthe Eumaenini, and tubercles, teeth, or unspecializedstructures in the Polyommatini. Morphological differen-ces can be found in the files of the different tribes,regarding the number of teeth, their form and size.Cuticular formations of the organs were studied on asurface of 2,500 mm2 and the average of ridges,tubercles, and teeth was measured searching for rele-vant taxonomic information. Stridulatory organs werethought to be an adaptation to myrmecophily but weshow that they are present both in myrmecophilousand nonmyrmecophilous species; therefore, we suggestthat this trait probably did not evolve in relation withmyrmecophily, but may be used to enhance relation-ships with ants. J. Morphol. 275:414–430, 2014. VC 2013Wiley Periodicals, Inc.

KEY WORDS: Lepidoptera; Lycaenidae; Theclini;Eumaenini; Lycaenini; Polyommatini; stridulatoryorgans; myrmecophily; Iberian Peninsula; soundproduction

INTRODUCTION

Sound production is frequent in insects, espe-cially within the Orthoptera, Hemiptera, and Cole-optera. In some groups, differences in the soundproducing organs can be considered taxonomic fea-tures (Gaiger and Vanin, 2006; Wessel, 2006).Sounds are transmitted by air or the substrateand have a special role in the intraspecific commu-nication in many insect species and are sometimesalso relevant at the interspecific level (Kirchner,1997). Stridulation is a sound production mecha-nism that consists on rubbing a file with teeth

against a sclerotized plate (Hinton, 1948; Haskell,1961; Dumortier, 1963; Downey, 1966). The stridu-latory organs usually show a similar structure andare formed by two parts: the pars stridens and theplectrum that move against each other (Dumortier,1963).

Within the Lepidoptera, sound production mayoccur either in larvae, pupae or adults. Most ofthe sound production in the Lepidoptera hasevolved in response to selection on sexual or defen-sive traits (Ewing, 1984); juvenile sound produc-tion in the Lycaenidae also mediates associationswith ants. In the adults, stridulatory sounds areemitted when ridges on the wings are rubbedagainst the legs, as in some Noctuidae (Haskell,1961). Stridulatory organs have been described insome Lycaenidae pupae species (Downey andAllyn, 1973) and have also been observed in somespecies of other Lepidoptera families such as theHesperiidae, Papilionidae, Lymantriidae, Saturni-dae, and in Sphingidae (Hinton, 1948; Dumortier,1963).

The majority of lycaenids have associations withants that can be facultative or obligate and rangefrom mutualism to parasitism. Larvae and pupaeuse chemical and acoustic signals to attract the

Additional Supporting Information may be found in the online ver-sion of this article.

Contract grant sponsor: Comunidad de Madrid and the EuropeanUnion: CAM-FEDER, GR/AMB/0266/2004 (joint project).

*Correspondence to: Margarita �Alvarez; Escuela Universitaria deMagisterio ESCUNI, Avenida Nuestra Se~nora de F�atima 102, ES-28047-Madrid, Spain. E-mail: [email protected]

Received 9 January 2013; Revised 29 September 2013;Accepted 4 October 2013.

Published online 3 December 2013 inWiley Online Library (wileyonlinelibrary.com).DOI 10.1002/jmor.20224

VC 2013 WILEY PERIODICALS, INC.

JOURNAL OF MORPHOLOGY 275:414–430 (2014)

ants (Fiedler et al., 1996; Akino et al., 1999; Trav-assos and Pierce, 2000; Pierce et al., 2002; Barberoet al., 2009) and some authors suggest that soundproducing organs might have evolved in relationto myrmecophily (Pierce et al., 2002). There is,therefore, some interest to show if those lycaenidsthat could be defined as nonmyrmecophilous havestridulatory organs and produce sounds, becausethis would support the idea that sound productiondid not initially evolve in relation to myrmecoph-ily. Currently, larval structures for sound produc-tion are still not known in full detail despitecareful descriptions by Downey and Allyn (1979),Kitching et al. (1985, 1999), and Barbero et al.(2009). The pupae of Lycaenidae have stridulatoryorgans at the end of the fifth and the beginning ofthe sixth pupal abdominal segments (Hinton,1948; Dumortier, 1963; Downey, 1966; Downeyand Allyn, 1973, 1978; Barbero et al., 2012;Kaminski et al., 2012). The stridulatory plate ofthe pupal fifth abdominal segment has a charac-teristic structure in different species, withtubercles, teeth, or ridges. Behind the stridulatoryplate, there is an intersegmental region formed bya transparent membrane and behind this, we finda file on the sixth abdominal segment. This fileshows teeth in all species (Kleemann in Schild,1877, 1876; Prell, 1913; Hinton, 1948; Downey,1966; Downey and Allyn, 1973, 1979).

The pupae produce sounds by rubbing the twosclerotized abdominal structures against eachother (Hinton, 1948; Haskell, 1961; Dumortier,1963; Downey, 1966). We have occasionallyobserved this behavior in pupae kept under labo-ratory conditions by observing pupal movementwith a stereomicroscope while the pupae was overa phonendoscope membrane that allowed hearingthe sounds (see also Fiedler, 1992). The muscularactions involved in these movements have notbeen studied in detail.

The stridulatory organs have been described indetail by Downey and Allyn (1973) for 65 species ofLycaenidae coming from different world regions.Regarding species that are present in the study area,the stridulatory organs of Polyommatus icarus andthe stridulatory plate of Lycaena phlaeas have beenpreviously described by Downey (1973). Other speciessuch as Satyrium ilicis, Glaucopsyche alexis, Phenga-ris nausithous, and Polyommatus dorylas were men-tioned by Downey as having stridulatory organs, butwithout providing images or descriptions.

In previous studies made in the Iberian Penin-sula, Mart�ın Cano (1982) showed stridulatoryorgans in the different species of the Lycaenidaefamily. In this article, we add new data to this pre-vious work and describe for the first time thedetailed morphology of the stridulatory organs ata quantitative and qualitative level for 36 Iberianlycaenids, representative of the fauna of thisregion. Data from the different species are com-

pared to show differences or similarities amongrelated species and some quantitative measure-ments are developed to explore the possible taxo-nomic value of these structures.

MATERIAL AND METHODSSpecies Collection

Field collected specimens came from different places inSpain: Almeria, �Avila, Badajoz, Burgos, Cantabria, C�aceres,Granada, Guadalajara, Huesca, Ja�en, Madrid, Segovia, andTeruel. Part of the material was specifically collected for thisstudy in the field between March 2004 and October 2006, dur-ing a total of 13 samplings trips. The rest of the material camefrom the immature stages collection of the Biology Departmentof the Universidad Aut�onoma of Madrid (Table 1 and Support-ing Information S1).

The specimens were collected by searching for the larvae onthe foodplants or shaking the branches of shrubs and trees andcollecting the larvae with a butterfly net. The collected larvaewere fed in the laboratory with their specific foodplants untilpupation. Then, these pupae were used for recording experi-ments and kept until the adults emerged. After adult emer-gence, the cast skins of the specimens were preserved and usedfor the scanning electron microscopy (SEM). The emergedadults were useful to confirm species identifications, adultswere identified following the keys of Fauna Ib�erica (Garc�ıa Bar-ros et al., 2013).

For the species Aricia cramera, Lycaena bleusei, and Zizeeriaknysna, adult females were collected in the field and kept inthe laboratory until eggs were laid. Then the larvae hatchingfrom the eggs were fed in captivity until pupation.

Sample Preparation

The different origin of the specimens caused that not all ofthem were in the same conditions. Pupae collected in the fieldwere fixed and conserved in 70% ethanol, dried at room temper-ature during 24 h in order to achieve total dehydration. Thecast skins coming from pupae bred in laboratory conditionswere also conserved in ethanol and dried during 24 h to facili-tate posterior mounting. Collection material was also trans-ferred to vials with ethanol to make manipulation easier. Tomount the specimens, they were washed with water andscreened with a stereomicroscope to find the stridulatoryorgans in the abdomen of the pupae and then the segmentsinvolved dissected for analysis. The dorsal region of the pupaeand the cast skins were mounted on stubs fixing them withsticky paper. Then samples were covered with a 102 A goldlayer with a Sputter Coater SC502 (Quorum Technologies,Laughton Lewes, South Sussex) and measured and photo-graphed at the Interdepartmental Research Service (SIDI) ofthe Universidad Aut�onoma of Madrid. We used a Philips XL30(Koninklijke Philips N.V., Amsterdam) and Hitachi S-3000NHitachi, Chiyoda, Tokio) for scanning electron microscopy.

Photographs, Morphological analysis, andMeasurements

The fourth-fifth, fifth-sixth, and sixth-seventh abdominalintersegments of each specimen were studied to confirm thelocation of the stridulatory organs. Quantitative data on speci-alized cuticular structures were taken from each specimencounting the teeth, ridges, or tubercles from five quadrats of 503 50 mm (2,500 mm2) on the stridulatory plate and on the file.Counts were always made on areas where the plate was per-pendicular to the observing direction. These results were usedto compare the stridulatory organs from a qualitative andquantitative point of view at the species, genus or tribe level.The absence of specialized structures was also used in the com-parative analysis (see data in Supporting Information S2).

415COMPARATIVE STUDY OF STRIDULATORY ORGANS

Journal of Morphology

The studied specimens and sample preparations are storedwith curatorial purposes in the collection of the Biology Depart-ment of the Universidad Aut�onoma of Madrid. Nomenclaturefor the family, subfamily, and tribe levels followed Fiedler(1991) and the nomenclature used for the species follows Garc�ıaBarros et al. (2013) with some modifications from the recentarticle of Talavera et al. (2012). The taxonomic list of thestudied species is provided in Table 1.

A phylogenetic tree was constructed using the software Tree-view X 0.5.1 (Softonic, Barcelona) modifying the one in Car-nicer et al. (2013) to represent only the Iberian Lycaenidaegenera. To this tree, we added the presence or absence of strid-ulatory organs.

RESULTS

We describe and compare the stridulatoryorgans of the pupae of 36 species, which represent50% of the 72 Iberian Lycaenidae species covering19 of the 26 Iberian genera. All the Iberian lycae-nid species belong to the subfamily Lycaeninae.Regarding the tribes, we studied one member ofthe Theclini, seven of the Eumaeini, two of theLycaenini, and 25 of the Polyommatini. Of thestudied lycaenids, 26 species (74%) show stridula-tory organs (one Theclini, six Eumaeini, two

TABLE 1. Number of specimens, localities, geographic coordinates, and provinces from which samples were taken for the study ofstridulatory organs of Iberian Lycaenidae

Species No. of samples Localities

THECLINIFavonius quercus (Linnaeus, 1758) 2 Miraflores 40�4805100N 3�4505900W (Madrid)

La Granja 40�5400500N 4�0002400W (Segovia)EUMAEINISatyrium spini (Denis and Schifferm€uller, 1775) 3 Valsa�ın 40�5204000N 4�0104000W (Segovia)

Manzanares el Real 40�4303300N 3�5105400W (Madrid)Satyrium ilicis (Esper, 1778) 3 Miraflores 40�4805100N 3�4505900W (Madrid)

La Granja 40�5400500N 4�0002400W (Segovia)Satyrium acaciae (Fabricius, 1787) 1 Miraflores 40�4805100N 3�4505900W (Madrid)Satyrium w-album (Knoch, 1782) 1 Ans�o 42�4502100N 0�4904600W (Huesca)Tomares ballus (Fabricius, 1787) 1 Oliva de M�erida 38�4503700N 6�0703700W (Badajoz)Callophrys rubi (Linnaeus, 1758) 1 Monte Oroel, Jaca 42�3104200N 0�3105600W (Huesca)Callophrys avis (Chapman, 1909) 2 Candeleda 40�0902800N 5�1403600W (�Avila)LYCAENINILycaena phlaeas (Linnaeus, 1761) 3 Lozoya 40�5702000N 3�4704900W (Madrid)Lycaena bleusei (Oberth€ur, 1884) 2 Oteruelo del Valle 40�5302400N 3�5005900W (Madrid)POLYOMMATINILampides boeticus (Linnaeus, 1767) 4 Campo Real 40�2201000N 3�2200500W (Madrid)

Puebla de Bele~na 40�5301000N 3�1405800W (Guadalajara)Cacyreus marshalli (Butler, 1898) 3 Colmenar Viejo 40�3904900N 3�4601700W (Madrid)Leptotes pirithous (Linnaeus, 1767) 2 Cantoblanco 40�3205600N 3�4103500W (Madrid)

Aliseda 39�2503900N 6�4102800W (C�aceres)Tarucus theophrastus (Fabricius, 1793) 2 El Egido 36�4603600N 2�4805600W (Almera)

Guardias Viejas 36�4200300N 2�5100700W (Almer�ıa)Zizeeria knysna (Trimen, 1862) 2 Granada 37�1004100N 3�3505800W (Granada)Glaucopsyche alexis (Poda, 1761) 2 Campo Real 40�2201000N 3�2200500W (Madrid)Glaucopsyche melanops (Boisduval, 1828) 1 Campo Real 40�2201000N 3�2200500W (Madrid)Iolana debilitata (Schultz, 1905) 2 Campo Real 40�2201000N 3�2200500W (Madrid)Phengaris alcon (Denis and Schifferm€uller, 1775) 2 Sonabia 43�2405100N 3�2000400W (Cantabria)Phengaris alcon rebeli (Hirschke, 1904) 1 Panticosa 42�4302500N 0�1700100W (Huesca)Phengaris nausithous (Bergstr€asser, 1779) 2 Oteruelo del Valle 40�5302400N 3�5005900W (Madrid)Scolitantides panoptes (H€ubner, 1813) 1 Dur�on 40�3704100N 2�4303300W (Guadalajara)Scolitantides abencerragus (Pierret, 1837) 3 Campo Real 40�2201000N 3�2200500W (Madrid)

Aranjuez 40�0100800N 3�3505700W (Madrid)Scolitantides orion (Pallas, 1771) 2 Santa Casilda, Briviesca 42�3301100N 3�2400500W (Burgos)Agriades zullichi (Hemming, 1933) 2 Aldeire 37�0504000N 3�0604800W (Granada)Agriades pyrenaicus (Boisduval, 1840) 2 �Aliva 43�0905300N 4�4705300W (Cantabria)Kretania hesperica (Rambur, 1839) 2 Campo Real 40�2201000N 3�22’0500W (Madrid)Aricia cramera (Eschscholtz, 1821) 2 Cantoblanco 40�3205600N 3�4103500W (Madrid)Aricia morronensis (Ribbe, 1910) 3 Cazorla 37�5401400N 2�5704200W (Ja�en)

Veleta, Sierra Nevada 37�0504100N 3�2302000W (Granada)Polyommatus ripartii (Freyer, 1830) 1 Bolta~na 42�2400600N 0�0104700W (Huesca)Polyommatus fabressei (Oberth€ur,1910) 1 Albarrac�ın 40�2403000N 1�2603800W (Teruel)Polyommatus golgus (H€ubner,1813) 1 Veleta, Sierra Nevada 37�0504100N 3�2302000W (Granada)Polyommatus dorylas (Dennis and Schifferm€uller, 1775) 2 Vivanco de Mena 43�0503800N 3�2101200W (Burgos)Polyommatus thersites (Cantener, 1834) 3 Campo Real 40�2201000N 3�2200500W (Madrid)Polyommatus icarus (Rottemburg, 1775) 1 Cantoblanco 40�3205600N 3�4103500W (Madrid)Plebejus argus (Linnaeus, 1758) 3 Ans�o 42�4502100N 0�4904600W (Huesca)

Valle de Ordesa 42�3900800N 0�4904600W (Huesca)Renanu�e 42�2901000N 0�3202900E (Huesca)

416 M. �ALVAREZ ET AL.


Lycaenini, and 17 Polyommatini) and nine lackthese structures (one Eumaeini and eight Polyom-matini), showing that a relevant proportion of theregional fauna have these structures.

Description of Stridulatory Organs

The stridulatory organs consist of a stridulatoryplate and a file. The plate appears on all fourtribes at the end of the dorsal area of the fifthabdominal segment (Fig. 1A) and shows morpho-logical differences between the Theclini, Eumaeini,

Lycaenini, and Polyommatini. It presentstubercles, teeth, ridges, or variable unspecializedstructures. The file appears on the anterior part ofthe sixth abdominal segment. In the four tribesreferred to above, it is formed by teeth, but withdifferences in number, form, and size. The study ofthe fourth-fifth and sixth-seventh intersegmentsconfirmed that the specialized structures for strid-ulation, only appear in the studied species on thefifth and sixth abdominal segments. In 21 of thestudied lycaenid species, there are teeth on theanterior regions of the fifth and seventh segments,

Fig. 1. A: Scanning electron micrographs of the location of Lycaenid pupae stridulatory organs on the fifth and sixth abdominalsegments of Agriades pyrenaicus. B: Pupa of Tomares ballus with no organs on this region. C: Stridulatory organs of Favonius quer-cus on the fifth and sixth abdominal segments. D: Teeth on the anterior part of the fifth abdominal segment of F. quercus. E and F:General view and detail of the fifth and sixth abdominal segments of the pupa of T. ballus.



but no specialized stridulatory structures are pres-ent at the end of the fourth and sixth segments.This character appears in all four Lycaenidaetribes. The species lacking the organs have no spe-cialized structures on the mentioned segments(Fig. 1B).

On the other hand, four species (Phengarisalcon, Lampides boeticus, Lycaena bleusei, and L.phlaeas) have a strip of teeth before the special-ized structures of the stridulatory plate of the fifthabdominal segment. It is interesting to note that

this feature appears in species belonging to thetribes Polyommatini and Lycaenini.

Tribe Theclini

The stridulatory organs of the pupae of Favo-nius quercus are shown in Figure 1C. The stridu-latory plate is located on the end of the fifthabdominal segment and is formed by three parts,one with tubercles, a second one with prominentridges, and a third again with tubercles, the latter

Fig. 2. Scanning electron micrographs of the stridulatory organs on the fifth and sixth abdominal segments of the pupae of Saty-rium w-album (A) and S. spini (B). C: Teeth on the anterior part of the seventh abdominal segment of the pupa of S. spini. D: Strid-ulatory organs on the fifth and sixth abdominal segments of the pupae of S. ilicis E: Detail of the undulations on the fifth abdominalsegment of the pupa of S. spini. F: Stridulatory organs of S. acaciae.



tubercles are in touch with the membrane thatconnects this segment with the next. The sixthsegment has the file that bears teeth. Other seg-ments of the pupae of F. quercus have been stud-ied and have teeth on their anterior regions, butno other cuticular specializations are observed onthe end of the segments (Fig. 1D). The number oftubercles of the fifth and of teeth on the sixth seg-ment is given in Supporting Information S2.

Tribe Eumaeini

Among the Eumaeini, we observed differencesamong species in the presence and morphology of

stridulatory organs. Tomares ballus lacks theseorgans and has no specialized structures (teeth,tubercles, or ridges) on the fifth and sixth abdomi-nal segments (Fig. 1E,F). In contrast, all species ofthe genus Satyrium show stridulatory organs.

Satyrium w-album (Fig. 2A) and S. ilicis (Fig.2D,E) have undulations on the fifth abdominalsegment. S. spini (Fig. 2B,C), S. acaciae (Figs. 2Fand 3A), Callophrys rubi (Fig. 3B), and C. avis(Fig. 3C), show ridges on the fifth abdominal seg-ment. In all these species, except in C. avis, theseridges cross each other to form a reticulum. Withthe exception of Tomares ballus, all the species of

Fig. 3. A: Detail of the stridulatory organs of the pupa of Satyrium acaciae. Stridulatory organs on the fifth and sixth abdominalsegments of the pupae of Callophrys rubi (B), C.avis (C) and Lycaena phlaeas (D). E: Teeth on the beginning of the seventh abdomi-nal segment of the pupa of L. phlaeas. F: Stridulatory organs of the pupa of L. bleusei. D and F show a strip of teeth before thetubercles of the fifth abdominal segment.



this tribe show teeth on the sixth abdominal seg-ment with their tips oriented towards the poste-rior part of the segment. Figure 2C shows theteeth in the anterior part of the seventh abdomi-nal segment. The number of undulations, ridges,and teeth of the stridulatory organs in a surface of2,500 mm2 is given in Supporting Information S2,except for S. w-album in which the images werenot useful to count undulations.

Tribe Lycaenini

The two studied species of the Lycaenini,Lycaena phlaeas and L. bleusei, have stridulatory

organs and both show the same specialized struc-tures: tubercles on the fifth abdominal segmentand teeth on the sixth segment (Fig. 3D). In bothspecies, a strip of teeth is present on the fifth seg-ment before the tubercles (Fig. 3D,F). As in otherspecies of the Lycaenidae, teeth are present on thedorsal area at the beginning of other abdominalsegments (Fig. 3E). The L. bleusei pupa (Fig. 4A)shows tubercles of different sizes, large and smallones, the latter appearing in groups of two orthree. Counts of the number of tubercles and teethof the stridulatory organs are given for a surfaceof 2,500 mm2 in Supporting Information S2.

Fig. 4. A: Stridulatory organs of the pupa of Lycaena bleusei. B-F: Details of the fifth and sixthabdominal segments without stridulatory organs of the pupa of Cacyreus marshalli (B), Tarucustheophrastus (C-D), Glaucopsyche alexis (E), and G. melanops (F). D: Teeth on the beginning ofthe sixth abdominal segment of the pupa of T. theophrastus.



Tribe Polyommatini

The tribe Polyommatini gathers the majority ofIberian lycaenids and also shows the greatest vari-ability of specialized structures within the stridula-tory organs. There are species without stridulatoryorgans, with tubercles on the fifth segment of theorgans and teeth on the sixth, with teeth on bothsegments, with a ridge on the fifth and teeth onsixth segment, and with nonspecialized structureson the fifth and teeth on the sixth segment.

Cacyreus marshalli (Fig. 4B), Tarucus theophras-tus (Fig. 4C,D), Glaucopsyche alexis (Fig. 4E), G.

melanops (Fig. 4F), Iolana debilitata (Fig. 5A), andScolitantides panoptes (Fig. 5B) have no stridula-tory organs. There are also two species that haveteeth on the sixth abdominal segment but with nospecialized structures on the stridulatory plate(fifth abdominal segment): Scolitantides orion (Fig.5C) and S. abencerragus (two of the three studiedspecimens of S. abencerragus, Fig. 5D).

Typical stridulatory organs with a stridulatoryplate formed by tubercles and a file formed byteeth are present in the following species: Lamp-ides boeticus (Fig. 5E,F), Leptotes pirithous (Fig.6A,B), Zizeeria knysna (Fig. 6C,D), Phengaris

Fig. 5. A–D: Details of the fifth and sixth abdominal segments without stridulatory organs ofthe pupa of Iolana debilitata (A), Scolitantides panoptes (B), S. orion (C), and S. abencerragus(D). E-F: Details of the tubercles of the fifth and the teeth on the sixth abdominal segments ofthe stridulatory organs of the pupa of Lampides boeticus.



alcon alcon (Fig. 6E,F), subspecies rebeli that wasconsidered a different species by many authors issimilar to the nominal species for this trait (Fig.7A), Agriades pyrenaicus (Fig. 7B–D), Kretaniahesperica (Fig. 7E,F), Aricia cramera (Fig. 8A,B),A. morronensis (Fig. 8C,D), Polyommatus ripartii(Figs. 8E,F and 9A–C,E), P. dorylas (Fig. 9D,F), P.golgus (Fig. 10A,B), P. thersites (Fig. 10C–E), andP. icarus (Fig. 10F). The average number oftubercles and teeth of the stridulatory organs in2,500 mm2 squares is given for all the species inSupporting Information S2.

Agriades zullichi has stridulatory organs withthe stridulatory plate and the file formed bothby teeth (Fig. 11A,B) and Phengaris nausithousshows a ridge in the stridulatory plate (Fig.11C). The latter species also lacks teeth at thebeginning of abdominal segments other than thestridulatory segments. It is interesting to notethat P. nausithous is different from its close rel-ative P. alcon that shows tubercles on the strid-ulatory plate. Finally, Plebejus argus (Fig. 11D–F) has stridulatory organs in which the stridu-latory plate is a lipped structure with no

Fig. 6. A: Stridulatory organs of a Leptotes pirithous pupa. B: Teeth on the beginning of the fifth abdominal segment of the pupaof L. pirithous. C: Stridulatory organs of the pupa of Zizeeria knysna. D: Detail of the tubercles of the fifth abdominal segment of thestridulatory organsof Z. knysna. E: Stridulatory organs of the pupa of Phengaris alcon with teeth before the tubercles of the fifthabdominal segment. F: Detail of the stridulatory organs of P. alcon.



cuticular outcrops and has consequently asmooth surface.

Measurements and Phylogeny ofStridulatory Organs

Supporting Information S2 gives the averagenumber of structures that form the stridulatoryorgans for each species in a surface of 2,500 mm2.Specimens with zero values lack cuticular special-izations on the segments hosting stridulatory

organs, and the others without values, are notpresent (n.p).

Within the tribes Theclini and Eumaeini, thestridulatory plate on the fifth abdominal segmenthas ridges or undulations, a character that differsfrom the Lycaenini and Polyommatini where onlyPhengaris nausithous has a single transversalridge. Ridges, present in species such as Satyriumspini, S. acaciae, Callophrys rubi, and C. avis,have a minimum average value of 3.8 ridges per2,500 mm2 (S. acaciae) and a maximum of 23.6.

Fig. 7. Stridulatory organs of the pupa of Phengaris alcon rebeli (A) and Agriades pyrenaicus (B). Detail of the tubercles on thefifth abdominal segment (C) and of teeth on the anterior part of the fifth abdominal segment (D) of the pupa of A. pyrenaicus. E:Stridulatory organs of the pupa of Kretania hesperica. F: Detail of the tubercles on the fifth and teeth on the sixth abdominal seg-ments of K. hesperica.



In the tribe Lycaenini, the average number oftubercles on the fifth abdominal segment isgreater in Lycaena bleusei (76 tubercles per 2,500mm2) than in L. phlaeas (43 per 2,500 mm2). Thenumber of teeth on the sixth segment is alsogreater for the former species: 12 teeth per 2,500mm2 in L. phlaeas and 17 in L. bleusei.

A phylogenetic tree of the Iberian Lycaenidaegenera is shown in Figure 12. The tree shows thepresence of stridulatory organs in all the lycaenidlineages except for the Glaucopsyche related gen-era and the genus Tomares.

DISCUSSION

The stridulatory organs of the pupae of lycaenidspecies from the study area are located on the fifthand sixth abdominal segments. This is consistentwith the data from Hinton (1948), Dumortier(1963), Downey (1966), and Scoble (1992), whilethe pupae of other Lepidoptera with stridulatoryorgans can have them in other locations. However,Downey (1973) and Pierce et al. (2002) mentionthe presence in some lycaenid species of stridula-tory structures on the abdominal segments 4/5and 6/7. For this reason, we explored the presence

Fig. 8. A: Stridulatory organs of the pupa of Aricia cramera. B: Teeth on the anterior part of the fifth abdominal segment of A. cra-mera. C: Stridulatory organs of the pupa of A. morronensis. D: Detail of the tubercles and teeth on the fifth and sixth abdominal seg-ments of the pupa of A. morronensis. E: Stridulatory organs of the pupa of Polyommatus ripartii. F: Teeth on the anterior part of theseventh abdominal segment of the pupa of P. ripartii.



of such structures in our species without positiveresults.

One of the species in which we did not findstridulatory organs is Glaucopsyche alexis. Thiscontrasts with the results of Downey (1973) whoreported the presence of such organs in this spe-cies. However, this author only found teeth in thesixth abdominal segment of the pupae, which notnecessarily represents a stridulatory function forthis structure. Some of the species where stridula-tory organs are missing are closely related: Iolanadebilitata, Glaucopsyche alexis, and G. melanops

(Als et al., 2004). Regarding the species in our study,the presence of stridulatory organs has been previ-ously stated by Downey (1973) for Satyrium ilicis,Phengaris nausithous, Scolitantides orion, Polyomma-tus dorylas, and P. icarus, but images were only pro-vided for the latter species. The specialized structuresof the stridulatory organs (teeth, tubercles, ridges)from our study are coincident with those described inthe monograph published by Downey (1966).

Most of the species studied had teeth on theanterior areas of every abdominal segment.Downey (1973) found them in the majority of

Fig. 9. A: Stridulatory organs of the pupa of Polyommatus faberssei. Details of the teeth (B)and the tubercles (C) on the of the sixth and fifth abdominal segments of the pupa of P. faberssei.D: Stridulatory organs of the pupa of P. dorylas. Teeth on the anterior part of the fifth abdominalsegment of the pupa of P. fabressei (E) and P. dorylas (F).



intersegmental areas of the pupal abdomen, fromthe 2/3 to the 7/8 intersegments. The function ofthese teeth is assigned by Downey (1973) to themoment in which the larval casts off its cuticle,when these teeth can keep apart adjacent seg-ments preventing adherences. Downey (1973) sug-gests that the presence of teeth must be anadvantage for the pupae, because otherwise theywould not be present in areas outside the stridula-tory organs. In relation to the stridulatory organs,these teeth are undoubtedly a preadaptation thatwould have favored the development of the com-plex stridulatory structures.

There are differences in the morphology of theorgans (ridges, tubercles, teeth) between groupswithin the Lycaenidae, and there are differencesin the stridulatory organs of closely related speciestoo (e.g., Phengaris alcon and P. nausithous). It isinteresting to note that Kretania hesperica (Fig.7E,F), that has recently been positioned withinthis genus and drawn out from Plebejus, showsevident affinities to Agriades that Talavera et al.(2012) consider a close relative of Kretania. Thementioned species is clearly distinct from Plebejusargus (Fig. 11D,F) that was previously regardedas belonging to the same genus. So, the present

Fig. 10. A: Stridulatory organs of the pupa of Polyommatus golgus. B: Detail of the tuberclesand teeth on the fifth and sixth abdominal segments of the pupa of P. golgus. C: Stridulatoryorgans of the pupa of P. thersites. Detail of the tubercles (D) and teeth (E) on the fifth and sixthabdominal segments of the pupa of P. thersites. F: Stridulatory organs of the pupa of P. icarus.



results support the new adscription of K.hesperica.

Some butterfly species produce cohorts of chemi-cal and/or acoustical signals, which are involved intheir interactions with ants (Pierce et al., 2002).Lycaenidae and Riodinidae butterfly larvae havebeen proved to employ vibrational signals in mutual-istic relationships with ants (DeVries, 1990; De Vri-es, 1991; Travassos and Pierce, 2000). Therefore,some authors suggest that sound production and thepresence of specialized stridulatory organs evolvedas a result of myrmecophilous interactions (Pierceet al., 2002), but our results do not support this view.

Among the 36 studied taxa, the stridulatoryorgans appear in myrmecophilous and amyrmeco-philous species. Myrmecophily in most lycaenidspecies only takes place at the larval stage. In thecase of obligate myrmecophilous relationships, thepupae can also have interactions with ants and inthe case of the genus Phengaris, the sounds pro-duced by the pupae have proved to be relevant inthe relationships (Barbero et al., 2009). In ourstudy, we found 27 species with stridulatoryorgans and of these, 20 species are myrmecophi-lous and seven are not (�Alvarez et al., 2012).Regarding the nine species without stridulatory

Fig. 11. A: Stridulatory organs of the pupa of Agriades zullichi. B: Detail of the teeth on the fifth abdominal segment of A. zullichi.C: Stridulatory organs of the pupa of Phengaris nausithous with a ridge on the stridulatory plate. D: Stridulatory organs of the pupaof Plebejus argus. E-F: Details of the thickening and of the teeth of the fifth and sixth abdominal segments of the pupa of P. argus.



organs, seven are myrmecophilous and two arenot. The ants that are related with the myrmeco-philous lycaenids, may have stridulatory organs ornot. However, the acoustic signals can be producedby knocking body parts onto the substrate (also

known as drumming), and can elicit variousbehavioral responses (H€olldobler, 1999; Billen,2006).

DeVries (1990, 1991) found that only myrmeco-philous lycaenids produce calls, and several non-myrmecophilous members of the Eumaeini aresilent. However, since then, several nonmyrmeco-philous lycaenids have been observed to producesound. Fiedler (1992) suggested that the ability toproduce calls may be universal in the Lycaenidae:whereas nonmyrmecophilous lycaenids producesimple calls in response to a disturbance, myrme-cophilous lycaenids have calls of greater complex-ity that are produced more often, and Travassosand Pierce (2000) showed that pupal calling is alsoinvolved in ant recruitment.

Some myrmecophilous species lack these organsin contrast with the opinions of Pierce et al. (2002)that state that the organs have been found in allthe analyzed species, both myrmecophilous andnonmyrmecophilous. This supports the idea thatstridulatory organs and sound production are notnecessarily related with myrmecophily within thegroup. Once this trait is present in many lycaenidspecies, they can use it for communication withants and/or as a defensive device (Pierce andNash, 1999; Travassos and Pierce, 2000). However,Schild (1876) states that undisturbed pupae canproduce sounds when kept in a container. Fromour results, it is clear that the studied amyrmeco-philous species have stridulatory organs, (someSatyrium and the genera Callophrys, Lycaena,and Agriades) and therefore, we can infer that thepresence of these organs is not necessarily relatedwith myrmecophily. One of our species (Cacyreusmarshalli) has not been proved to have relationswith ants and also lacks stridulatory organs, butthe life cycle of this species is peculiar because thelarvae is endophagous for most of its life and canthus hardly have relationships with some kinds ofants. Assuming that the stridulatory organs arerelated to sound production, there is a species(Glaucopsyche alexis) without these structuresthat is able to produce sounds (�Alvarez, 2009) butwe do not know the mechanism involved. Whilespecies with specialized organs readily producesounds after disturbance, a clear support for theirdefensive function, species lacking these struc-tures show less elaborated and more casual soundemissions. A study of the sounds produced by abroad sample of Iberian Lycaenidae can be foundin �Alvarez (2009), and the data in this study isbeing compiled for further publication.

The pupae produce signals with both vibrationaland airborne components (Hoegh-Guldberg, 1972;Downey and Allyn, 1978). Calls are induced whenpupae are disturbed, and Downey and Allyn(1978) concluded that calls act primarily as adeterrent to predators and parasites. Sounds pro-duced by lycaenid pupae originate from tooth-and-

Fig. 12. Phylogenetic tree of the Iberian genera of the familyLycaenidae adapted from Carnicer et al. (2013) showing the spe-cies with (red) and without (black) stridulatory organs. The gen-era studied for this article are surrounded by a rectangle. [Colorfigure can be viewed in the online issue, which is available atwileyonlinelibrary.com.]



wileyonlinelibrary.com

comb stridulatory organs between the fifth andsixth segments (Downey, 1966; Downey and Allyn,1973, 1978; Pierce et al., 2002; Barbero et al.,2012).

In conclusion, we support the hypothesis thatthe stridulatory organs are generalized charactersand thus the absence of such organs may beregarded as an apomorphic trait. Some specializedstructures on the stridulatory plate, like thestrong ridge of Phengaris nausithous or the swol-len lips of Plebejus argus, could be intermediatecharacters. It is interesting to note that mostLycaenidae lineages have stridulatory organs like,for example, most of the Polyommatus relatedgenera and the Lycaenini tribe. On the otherhand, the genera related to Glaucopsyche lackthese organs except for Phengaris that is a veryspecialized obligate myrmecophilous genus inwhich the presence of these devices could be nec-essary for the closer relationship with the ants(Fig. 12). Recent studies (Barbero et al., 2009)about the association among Phengaris rebeli andit host ant, demonstrated this specialization, beingthe first case of acoustical mimicry in an antsocial parasite. They demonstrated that P. rebelilarvae and pupae are able to mimic the soundsproduced by the ant queens, thus obtaining a highstatus in the host colony hierarchy (Barbero et al.,2012).

ACKNOWLEDGMENTS

The authors are grateful to Jos�e Mar�ıa Hern�an-dez, Eduardo Ruiz, Enrique Garc�ıa-Barros, Jos�eMart�ın-Cano, and Roger Vila who significantlycontributed to our study with valuable commentsand suggestions. Enrique Garc�ıa-Barros helpedwith the construction of the phylogenetic tree.Esperanza Salvador, Marta Furi�o, and IsidoroPoveda of the Servicio Interdepartamental deInvestigaci�on (SIDI) of the Universidad Aut�onomaof Madrid collaborated with the Scanning ElectronMicroscope images. They also acknowledge thecontribution of Fernando Pardos, Michael Schmitt,and an anonymous referee for their significantcontribution to the improvement of a previous ver-sion of this manuscript.

LITERATURE CITED

Akino T, Knapp JJ, Thomas JA, Elmes GW. 1999. Chemicalmimicry and host specificity in the butterfly Maculinea rebeli,a social parasite of Myrmica ant colonies. Proc R Soc LondonB 266:1419–1426.

Als TD, Vila R, Kandul NP, Nash DR, Yen SH, Hsu YF,Mignault AA, Boomsma JJ, Pierce NE. 2004. The evolution ofalternative parasitic life histories in large blue butterflies.Nature 432:386–390.

�Alvarez M. 2009. Estudio de la comunicaci�on ac�ustica enlic�enidos (Lepidoptera, Lycaenidae) y form�ıcidos (Hymenop-tera, Formicidae). PhD Thesis. Universidad Complutense deMadrid. 264 p. unpublished.

�Alvarez M, Munguira ML, Ib�a~nez MDM. 2012. Nuevos Datos yRecopilaci�on de las Relaciones entre Lic�enidos y Form�ıcidosen la Pen�ınsula Ib�erica (Lepidoptera, Lycaenidae; Hymenop-tera, Formicidae). Shilap Revta Lepid 30:45–59.

Barbero F, Thomas J, Bonelli S, Balleto E, Sch€onrogge K. 2009.Queen ants make distinctive sounds that are mimicked by abutterfly social parasite. Science 323:782–785.

Barbero F, Patricelli D, Witek M, Balletto E, Casacci LP, SalaM, Bonelli S. 2012. Myrmica ants and their butterfly para-sites with special focus on the acoustic communication. Psy-che 2012:1–11.

Billen J. 2006. Signal Variety and Communication in SocialInsects. In: Proceedings of the Netherlands EntomologicalSociety Meeting, Vol. 17. pp 9–25. Amsterdam: NederlandseEntomologische Vereniging.

Carnicer J, Stefanescu C, Vila R, Dinc�a V, Font X, Pe~nuelas J.2013. A unified framework for diversity gradients: the adapt-ive trait continuum. Global Ecol Biogeogr 22:6–18.

DeVries PJ. 1990. Enhancement of symbiosis between butterflycaterpillars and ants by vibrational communication. Science248:1104–1106.

DeVries PJ. 1991. Call production by myrmecophilous riodinidand lycaenid butterfly caterpillars (Lepidoptera): Morphologi-cal, acoustical, functional, and evolutionary patterns. AmMus Novit 3025:1–23.

Downey JC. 1966. Sound production in pupae of Lycaenidae(Lep). J Lepid Soc 20:129–155.

Downey JC, Allyn AC. 1973. Butterfly ultrastructure. Soundproduction and associated abdominal structures in pupae ofLycaenidae and Riodinidae. Bull Allyn Mus 14:1–47.

Downey JC, Allyn AC. 1978. Sounds produced in pupae ofLycaenidae. Bull Allyn Mus 48:1–14.

Downey JC, Allyn AC. 1979. Morphology and biology of theimmature stages of Leptotes cassius theonus (Lucas) (Lepid.:Lycaenidae). Bull Allyn Mus 55:1–27.

Dumortier B. 1963. Morphology of sound emission apparatus inArthropoda. In: Busnel RG, editor. Acoustic Behaviour of Ani-mals. Amsterdam: Elsevier Publishing Company. pp 277–345.

Ewing AW. 1984. Insect communication. In: Lewis T, editor.Insect Communication. London: Academic Press. pp 223–240.

Fiedler K. 1991. Systematic, evolutionary, and ecological impli-cations of myrmecophily within the Lycaenidae (Insecta: Lepi-doptera. Papilionoidea). Bonn Zool Monograph 31:1–210.

Fiedler K. 1992. Notes on the biology of Hypolycaena othona(Lepidoptera: Lycaenidae) in West Malaysia. Nachr EntomolVereins Apollo 13:65–92.

Fiedler K, H€olldobler B, Seufert P. 1996. Butterflies and ants:The communicative domain. Experientia 52:14–24.

Gaiger F, Vanin SA. 2006. The elytro-femoral stridulatory appa-ratus in Curculionidae (Coleoptera), with notes on the acous-tic behaviour of Arniticus hylobioides (Boheman 1843) andErodiscus proximus (Viana 1959), and thanatosis display inthe latter species. Ann Soc Entomol Fr (n.s.): Int J Entomol42:165–170.

Garc�ıa-Barros E, Munguira ML, Stefanescu C, Vives A. 2013.In: Ramos MA, et al., editors. Fauna Ib�erica, Lepidoptera:Rhopalocera, Vol. 37, 1213 pp. Madrid: Museo Nacional deCiencias Naturales, CSIC.

Haskell PT. 1961. Insects Sounds. Aspects of Zoology Series.London: H. F. and G. Witherby Ltd. 189 p.

Hinton HE. 1948. Sound production in Lepidopterous pupae.Entomologist 81:254–269.

Hoegh-Guldberg O. 1972. Pupal sound production of someLycaenidae. J Res Lepid 10:127–147.

H€olldobler B. 1999. Multimodal signals in ant communication.J Comp Physiol 184A:129–141.

Kaminski L, Rodrigues D, Freitas AVL. 2012. Immature stagesof Parrhasius polibetes (Lepidoptera: Lycaenidae): Hostplants, tending ants, natural enemies and morphology. J NatHist 46:645–667.

Kirchner WH. 1997. Acoustical communication in social insects.In: Lehrer M, editor. Orientation and Communication inArthropods. Basel: Birkh€auser Verlag. pp 273–300.



Kitching RL, Luke B. 1985. The Myrmecophilous organs of thelarvae of some British Lycaenidae (Lepidoptera): A compara-tive study. J Nat Hist 19:259–276.

Kitching RL, Sheermeyer E, Jones RE, Pierce NE. 1999. TheImperial Blue: Jalmenus evagoras (Lycaenidae). The Biologyof Australian Butterflies (Monographs on Australian Lepidop-tera), Vol. 6. Sidney: CSIRO Press. pp 279–315.

Mart�ın Cano J. 1982. La Biolog�ıa de los Lic�enidos Espa~noles(Lep. Rhopalocera). Miscelanea Conmemorativa. Fac CienciasUniv Aut�onoma de Madrid 1:1003–1020.

Pierce NP, Nash D. 1999. The Imperial Blue: Jalmenusevagoras (Lycaenidae). The Biology of Australian Butterflies.(Monographs on Australian Lepidoptera), Vol. 6. Sydney:CSIRO Press. pp 277–313.

Pierce NP, Braby MF, Heath A, Lohman DJ, Mathew J, RandDB, Travassos MA. 2002. The ecology and evolution of antassociation in the Lycaenidae (Lepidoptera). Annu Rev Ento-mol 47:733–771.

Prell H. 1913. €Uber zirpende Schmetterlingspuppen. Biol Zen-tralbl 33:496–501.

Schild FG. 1876. Miscellen. F€unf Entwickelungsst€ande gleich-zeitig––Zirpende Insektenpuppen—Darwinistische Erw€agun-gen. Stett Ent Zeit 38:85–97.

Scoble MJ. 1992. The Lepidoptera. Form, Function and Diver-sity. Oxford: Natural History Museum and Oxford UniversityPress. 404 p.

Talavera G, Lukhtanov VA, Pierce NE, Vila R. 2012. Establish-ing criteria for higher-level classification using moleculardata: The systematics of Polyommatus blue butterflies (Lepi-doptera, Lycaenidae). Cladistics 29:166–192.

Travassos MA, Pierce NE. 2000. Acoustics, context and functionof vibrational signalling in a Lycaenid butterfly-ant mutual-ism. Anim Behav 60:13–26.

Wessel A. 2006. Stridulation in the Coleoptera an overview. In:Drosopoulos Sy, Claridge MF, editors. Insect Sounds andCommunication. Boca Rat�on: Taylor and Francis Group. pp397–403.



Documents

Comparative study of the morphology of stridulatory organs of the Iberian lycaenid butterfly pupae (Lepidoptera)