The Distribution and Evolution of Exocrine Compound

This article was downloaded by: [79.238.118.44] On: 17 July 2013, At: 23:24 Publisher: Taylor & Francis Informa Ltd Registered in England and Wales Registered Number: 1072954 Registered office: Mortimer House, 37-41 Mortimer Street, London W1T 3JH, UK

Annales de la Société entomologique de France (N.S.): International Journal of Entomology Publication details, including instructions for authors and subscription information: http://www.tandfonline.com/loi/tase20 The distribution and evolution of exocrine compound glands in Erotylinae (Insecta: Coleoptera: Erotylidae) Kai Drilling a b , Konrad Dettner a & Klaus-Dieter Klass b a Department for Animal Ecology II , University of Bayreuth, Universitätsstraße 30 , 95440 , Bayreuth , Germany b Senckenberg Natural History Collections Dresden , Museum of Zoology , Königsbrücker Landstraße 159, 01109 , Dresden , Germany Published online: 24 May 2013.

To cite this article: Kai Drilling , Konrad Dettner & Klaus-Dieter Klass (2013) The distribution and evolution of exocrine compound glands in Erotylinae (Insecta: Coleoptera: Erotylidae), Annales de la Société entomologique de France (N.S.): International Journal of Entomology, 49:1, 36-52, DOI: 10.1080/00379271.2013.763458 To link to this article: http://dx.doi.org/10.1080/00379271.2013.763458

PLEASE SCROLL DOWN FOR ARTICLE

Taylor & Francis makes every effort to ensure the accuracy of all the information (the “Content”) contained in the publications on our platform. However, Taylor & Francis, our agents, and our licensors make no representations or warranties whatsoever as to the accuracy, completeness, or suitability for any purpose of the Content. Any opinions and views expressed in this publication are the opinions and views of the authors, and are not the views of or endorsed by Taylor & Francis. The accuracy of the Content should not be relied upon and should be independently verified with primary sources of information. Taylor and Francis shall not be liable for any losses, actions, claims, proceedings, demands, costs, expenses, damages, and other liabilities whatsoever or howsoever caused arising directly or indirectly in connection with, in relation to or arising out of the use of the Content. This article may be used for research, teaching, and private study purposes. Any substantial or systematic reproduction, redistribution, reselling, loan, sub-licensing, systematic supply, or distribution in any form to anyone is expressly forbidden. Terms & Conditions of access and use can be found at http:// www.tandfonline.com/page/terms-and-conditions Annales de la Société entomologique de France (N.S.), 2013 Vol. 49, No. 1, 36–52, http://dx.doi.org/10.1080/00379271.2013.763458

The distribution and evolution of exocrine compound glands in Erotylinae (Insecta: Coleoptera: Erotylidae) Kai Drillinga,b*, Konrad Dettnera & Klaus-Dieter Klassb aDepartment for Animal Ecology II, University of Bayreuth, Universitätsstraße 30, 95440 Bayreuth, Germany; bSenckenberg Natural History Collections Dresden, Museum of Zoology, Königsbrücker Landstraße 159, 01109 Dresden, Germany (Accepté le 22 mars 2012)

Summary. Members of the family Erotylidae and especially of the subfamily Erotylinae possess a whole arsenal of compound integumentary glands. Their external pores are located in several parts of the body, mainly in the corners and along the lateral margins of the pronotum, beside the compound eyes (periocular glands), on the subgenal braces (subocular glands), on the abdominal ventrites, and more rarely on the prosternal and mesoventral intercoxal processes, the mentum and the mandibles. To evaluate the evolution of the various exocrine glands their occurrence is described for 55 species of Erotylidae (including the former Languriidae), and data from the literature are included for nine further species and two subfamilies. In some phylogenetically crucial cases, the glandular nature was veriﬁed by internal inspection (search for glandular ducts), and in some critical species or genera an extended sample was studied. Gland characters and their phylogenetic implications are discussed and mapped on a previously published erotylid phylogeny. Résumé. La distribution et l’évolution des glandes exocrines composées chez les Erotylinae (Insecta: Coleoptera: Erotylidae). Les membres de la famille des Erotylidae et spécialement de la sous-famille des Erotylinae ont un large arsenal de glandes tégumentaire composées. Leurs pores externes sont situés dans différentes parties du corps, surtout dans les coins et les bords des marges latérales du pronotum, à côté des yeux composés (glandes péri-oculaires), sur les ventrites abdominaux, et plus rarement sur les processus prosternaux intercoxal et mésoventral, le mentum et les mandibules. Pour évaluer l’évolution des différentes glandes exocrines, leur occurrence est décrite pour 55 espèces d’Erotylidae (y compris les anciens Languriidae), tandis que des données de la littérature sont incluses pour 9 espèces supplémentaires et 2 sous-familles. Dans certains cas phylogénétiquement cruciaux, la nature glandulaire a été vériﬁée par un examen interne (recherche de canaux glandulaires). L’échantillonnage a été étendu pour certaines espèces ou certains genres critiques. Les caractères des glandes et leurs implications phylogénétiques sont discutées et cartographiées à partir d’une phylogénie des Erotylidae précédemment parue. Keywords: languriidae; dermal glands; phylogeny; SEM

The cosmopolitan Erotylidae comprises about 3500 tionships in Cucujiformia are widely unresolved and described species in ca. 258 genera (Leschen et al. 2010). Cucujoidea are unlikely to be a monophyletic group (see They are classified into six subfamilies: Cryptophilinae, Leschen et al. 2005; Hunt et al. 2007; Buder et al. 2008; Downloaded by [79.238.118.44] at 23:24 17 July 2013 Pharaxonothinae, Languriinae, Xenoscelinae, Loberinae, and Lawrence et al. 2011). Moreover, the phylogenetic rela- Erotylinae. These subfamilies (excl. Erotylinae) previously tionships of Erotylidae to other cucujiform taxa are still formed a separate family, “Languriidae”; the family previously ambiguous. Various families were regarded as the closest known as Erotylidae is now ranked as the subfamily relatives of Erotylidae in the past: members of the cerylo- Erotylinae. This subfamily is further classified into five tribes: nid series of Cucujoidea (Alexiidae, Endomychidae; Dacnini, Tritomini, Erotylini, Encaustini, and Megalodacnini Crowson 1955; Sen Gupta & Crowson 1971), certain (Węgrzynowicz 2002; Leschen 2003; Leschen et al. 2010). “lower Cucujoidea” (Cryptophagidae, Propalticidae; Sen Erotylidae is a subgroup of the cucujiform beetles, Gupta & Crowson 1969, 1971; Leschen 1996, 2003; which are supported as monophyletic by a number of McHugh et al. 1997; Robertson et al. 2004) as well as autapomorphies (e.g., Lawrence & Newton 1982; Phloeostichidae, Lamingtoniidae (Leschen 2003), and Klausnitzer 2005; Leschen & Ślipiński 2010) and also by Biphyllidae (Sen Gupta & Crowson 1971; Leschen the extensive molecular study of Hunt et al. (2007). 2003; Leschen & Buckley 2007). The molecular studies However, the large-scale morphological analysis by by Hunt et al. (2007) suggest either Monotomidae, Lawrence et al. (2011) found Erotylidae to be paraphyletic. Helotidae, and Protocucujidae as the closest relatives Within Cucujiformia the Erotylidae are generally of Erotylidae (supporting fig. S1 therein: Bayesian analy- assigned to Cucujoidea. However, basal phylogenetic rela- sis), or a clade comprising Laemophloeidae, Phalacridae,

*Corresponding author. Email: [email protected]

Propalticidae, and Cucujidae (supporting fig. S4 therein: (1) In cases where our results differed from those in parsimony analysis). Węgrzynowicz (2002; Table 1), we examined several fi It is well known that Erotylidae show a particularly rich specimens from different localities to test for intraspeci c variability. This concerns six species: Toramus pilifer,3× equipment of compound exocrine glands. Glands can occur in Taraz (Kazakhstan), 2× Margilan (Uzbekistan); Dacne the corners and along the lateral margins of the pronotum, on bipustulata, 2× Saxon Switzerland (Germany), 2× the prosternal and mesoventral intercoxal processes, on the Trenčin (Slovakia), 2× from near Berlin (Germany); head anteromesal to the compound eyes, on the subgenal Triplax russica, 2× Magdeburg (Germany), 2× Saxon ň brace, on the abdominal ventrites, and rarely on the mentum Switzerland (Germany), 2× Kada (Czech Republic); Tritoma bipustulata, 4× Saxon Switzerland (Germany), and on the mandibles. Despite this manifold occurrence of 6× Bayreuth (Germany), 5× Trenčin (Slovakia); such glands over the beetles’ body, their consideration in the Pselaphacus nigropunctatus (species identification con- previous literature is quite sparse. The distribution of several firmed by P. Węgrzynowicz), 2× Peru, 3× São Paulo de gland pores across erotylid taxa was included in the phyloge- Olivença (Brazil), 2× Bogota (Colombia); Megalodacne netic studies of Węgrzynowicz (2002; mostly on the erotyline fasciata, 2× Texas (USA), 2× Chicago (USA), 1× New “ ” Orleans (USA). tribes) and Leschen (2003; mostly on the former languriid (2) In the single case where Węgrzynowicz’s (2002) and subfamilies). In both papers many taxa are scored with regard our results differ for species from the same genus we to the occurrence of glands, but the pictorial documentation is studied the gland pores concerned in an expanded sam- very limited. In addition, a morphological and histological ple of several congeneric species; this applies to study of the pronotal glands in one erotyline species, Tritoma Megischyrus brasiliensis versus M. undatus in Table 1: we studied M. brasiliensis, 2× Brasilia (Brazil), 2× bipustulata, was recently published by Drilling et al. (2010). Espírito Santo (Brazil); M. semipunctatus, 1× Brasilia The main scope of the present contribution is the (Brazil), 2× Santa Catharina (Brazil); M. decempunctatus, further exploration of the distribution of exocrine glands 3× Peru; M. discipennis, 1× Santa Catharina (Brazil), with a focus on “basal” Erotylinae. We specifically discuss 1× Chiriqui (Panama) (M. undatus was not available the evolution of the set of glands based on previously to us). (3) The same as in (2) was done in some cases crucial for published phylogenetic hypotheses. As compared to pre- phylogenetic conclusions; this applies to (a) Ischyrus vious studies, we specifically included additional “langur- quadripunctatus, 2× Manaos (Brazil); I. scriptus,2× iid” taxa and additional members of the Erotylinae; mainly Rio de Janeiro (Brazil), 1× Joinville (Brazil); I. femor- the “basal” tribes Dacnini and Tritomini. Especially in alis, 2× Rio de Janeiro (Brazil), 1× Peru; I. flavitarsis, these taxa our results show some striking differences to 3× Cuba; (b) Pselaphacus nigropunctatus, 4× São Paulo ę de Olivença (Brazil), 2× Peru, 1× Bogota (Colombia); those in the major foregoing studies of W grzynowicz P. giganteus, 3× Cayenne (French Guiana); P. rubrica- (2002) and Leschen (2003). tus, 2× Peru, 1× Amazonas (Brazil); P. puncticollis,2× Rio Grande do Sul (Brazilia), 1× Paraguay.

Biphyllidae were chosen as outgroup taxa based on the out- Material and methods group information of Wegrzynowicz (2002), Leschen (2003), Examined taxa Leschen et al. (2005), and Leschen and Buckley (2007). The generic and suprageneric classification of Erotylidae follows Leschen (2003) and Leschen et al. (2010). All subfamilies are Downloaded by [79.238.118.44] at 23:24 17 July 2013 Morphological studies, illustrations, and terminology represented in this study, except the Xenoscelinae and Loberinae (data taken from Leschen 2003). Of the erotyline tribes, all five – All morphological studies are based on dried specimens from the Dacnini, Tritomini, Erotylini, Encaustini, and Megalodacnini – Museum of Zoology Dresden; freshly killed specimens were are represented. additionally used in Triplax russica and Tritoma bipustulata. For the Erotylinae, species-level systematics follows the The head, pronotum and remaining body were separated and catalogs of Chûjô and Chûjô (1988, 1989, 1990: Palaearctic, macerated in hot 10% KOH. Afterwards the external cuticular Oriental, and Australian faunas, respectively) and Alvarenga surface was searched for gland openings in ca. 70% ethanol (1994: Neotropical fauna), and the surveys of Delkeskamp under a stereomicroscope, using magnifications up to 160× (1981: Ethiopian fauna) and Boyle (1956: Nearctic fauna). For (some inconsistencies between our observations and prior the “languriids” only a genus-level catalog (Leschen & papers, especially Leschen’s work, may be due to the different Węgrzynowicz 1998) and an issue about the Nearctic species taxon sampling and methods for observation; Leschen’s speci- (Lawrence & Vaurie 1983) are available. mens were slide-mounted; personal communication, 2011, via The 55 erotylid species here used are listed in Table 1. email). In some exemplary cases the presence of a compound Of these, 20 were also studied by Węgrzynowicz (2002) gland was checked for by an inspection of the internal face of (marked with “w” in Table 1, column 3) and nine species the cuticle of some pore area. For scanning electron microscope were exclusively studied by Węgrzynowicz (2002) (marked (SEM) studies of structural details of gland pores the respective with “W” in Table 1, column 4); data for two subfamilies parts of the exoskeleton were critical point dried (with CO2 as (Xenoscelinae and Loberinae) are also included in Table 1 transitional medium), coated with gold (Edwards S150B), and (Leschen 2003; marked with “W” or “L” in column 4) and examined with a Philips/FEI XL 30 ESEM. Pores confirmed in our discussions. An expanded sample of one particular as glandular in this way are marked with an asterisk in species or of several congeneric species was examined in the Table 1. Drawings and plates were completed using the following cases: graphic computer programmes CorelPhotoPaint version 12 and 38 K. Drilling et al.

Table 1. List of taxa studied (column 2) with systematic assignment (column 1), occurrence of various glands (as a character matrix; see chapter 3 for character definitions) and number of specimens examined (column 4; “W” = data exclusively from Węgrzynowicz 2002, “L” = data exclusively from Leschen 2003). Taxa also studied by Węgrzynowicz (2002) marked with “w” in column 3 (Lit.). Entry of more than one number indicates differences between our results (first entry) and those of Węgrzynowicz (2002) (second entry) or Leschen (2003) (second entry, bold). ~ = character is polymorphic; * = the glandular nature of particular pores has been checked by internal examination, i.e. observation of a glandular duct. Genus and species names according to Chûjô & Chûjô (1988, 1989, 1990), Delkeskamp (1981) and Alvarenga (1994), higher-level systematic assignment according to Leschen and Węgrzynowicz (1998) and Leschen (2003). Systematic Assignment Species Lit. No. C1 C2 C3 C4 C5 C6 C7 Biphyllidae Biphyllus lunulatus Fabricius 1787 w 3 0 0 0 0 0 0 0 Erotylidae, Xenoscelinae L 01? 000? Erotylidae, Pharaxonothinae Setariola sericea Mulsant 1863 w 3/L 0 0/1 00/1 0/1 0/1 0/? Erotylidae, Loberinae L 01? 1~1? Erotylidae, Cryptophilinae Cryptophilus integer Heer 1841 w W/L 0 0/1 00/1 00/1 0 Erotylidae, Cryptophilinae Toramus pilifer Reitter 1885 w 5/L 1 1/0 0 0/~ 000 Erotylidae, Languriinae Languria bicolor Fabricius 1798 w 2/L 1*/0 11 0 00/1 0/? Erotylidae, Languriinae Languria puncticollis Say 1823 2 1* 1 0 0 0 0 0 Erotylidae, Languriinae Pachylanguria elongata Fabricius 1801 2 1 1 0 0 0 0 0 Erotylidae, Languriinae Tetraphala (Tetralanguria) elongata Fabricius 1801 3 0 0 0 0 0 0 0 Erotylidae, Languriinae Camptocarpus longicollis Motschulsky 1860 2 0 0 0 0 0 0 0 Erotylidae, Languriinae Langurites lineata Laporte 1832 2 0 0 0 0 0 0 0 Erotylidae, Languriinae Pentelanguria elateroides Crotch 1876 4 0 1 0 0 0 0 0 Erotylidae, Erotylinae, Dacnini Endytus bizonatus Crotch 1876 2 1 1 0 2 1 0 0 Erotylidae, Erotylinae, Dacnini Triplatoma macleayi Lacordaire 1842 2 1 1 0 1 1 0 0 Erotylidae, Erotylinae, Dacnini Dacne bipustulata Thunberg 1781 w 6 0 1/0 0 0 0 0 0 Erotylidae, Erotylinae, Dacnini Dacne rufifrons Fabricius 1775 2 0 1 0 0 0 0 0 Erotylidae, Erotylinae, Dacnini Dacne notata Gmelin 1790 2 1 1 0 0 0 0 0 Erotylidae, Erotylinae, Dacnini Combocerus glaber Schaller 1783 w 2 1 1 0 1 0 0 0 Erotylidae, Erotylinae, Dacnini Euzostria aruensis Gorham 1888 2 1 1 0 1 0 0 0 Erotylidae, Erotylinae, Dacnini Nesitis celebesica Heller 1926 2 1 1 0 2* 1 0 0 Erotylidae, Erotylinae, Dacnini Nesitis sexnotata Wiedemann 1823 2 1 1 0 2 1 0 0 Erotylidae, Erotylinae, Dacnini Apteronesitis fulgurata Arrow 1928 2 1 1 0 2 1 0 0 Erotylidae, Erotylinae, Dacnini Cryptodacne synthetica Sharp 1878 w W 0 1 0 0 0 0 0 Erotylidae, Erotylinae, Dacnini Thallis nigroaenea Crotch 1876 2 0 1 0 0 0 0 0 Erotylidae, Erotylinae, Tritomini Cyrtomorphus pantherinus Lacordaire 1842 3 1 1 0 1 1 0 0 Erotylidae, Erotylinae, Tritomini Cyrtomorphoides albicornis Heller 1920 2 1 1 0 1 1 0 0 Erotylidae, Erotylinae, Tritomini Callischyrus venustus Lacordaire 1842 2 1 1 0 2 0 0 0 Erotylidae, Erotylinae, Tritomini Epytus violaceus Sturm 1826 2 1 1 1 1 0 0 0 Erotylidae, Erotylinae, Tritomini Lybas bicolor Guèrin-Mèneville 1841 2 1 1 1 1 0 0 0 Erotylidae, Erotylinae, Tritomini Palaeolybas coccinelloides Crotch 1876 2 1 1 0 1 1 0 0 Erotylidae, Erotylinae, Tritomini Triplax russica Linnaeus 1758 w 6 1 1 1/0 1* 0 0 0 Erotylidae, Erotylinae, Tritomini Mycotretus ornatus Duponchel 1825 w 2 1 1 1 1 0 0 0 Downloaded by [79.238.118.44] at 23:24 17 July 2013 Erotylidae, Erotylinae, Tritomini Triplacidea motschulskyi Bedel 1872 w W 1 1 1 1 1 0 0 Erotylidae, Erotylinae, Tritomini Tritoma bipustulata Fabricius 1775 w 15 1 1 1/0 1* 1* 0 0 Erotylidae, Erotylinae, Tritomini Tritoma atripennis Gorham 1885 2 1 1 1 1 1 0 0 Erotylidae, Erotylinae, Tritomini Amblyscelis kelleni Gorham 1888 w W 1 1 1 1 1 0 0 Erotylidae, Erotylinae, Tritomini Amblyopus vittatus Olivier 1807 w 2 1 1 1 1 1 0 0 Erotylidae, Erotylinae, Tritomini Zythonia fulva Westwood 1874 w W 1 1 1 1 0 0 0 Erotylidae, Erotylinae, Tritomini Apolybas bicolor Guèrin-Mèneville 1841 w W 1 1 0 1 0 0 1 Erotylidae, Erotylinae, Tritomini Megischyrus undatus Olivier 1792 w W 1 1 0 1 0 0 1 Erotylidae, Erotylinae, Tritomini Megischyrus brasiliensis Lacordaire 1842 4 1 1 1 2 0 0 1 Erotylidae, Erotylinae, Tritomini Ischyrus quadripunctatus Olivier 1792 w 2 1 1 1 1 0 0 1 Erotylidae, Erotylinae, Tritomini Pselaphacus nigropunctatus Percheron 1835 w 7 1* 1 1 2*/1 1 0 0 Erotylidae, Erotylinae, Tritomini Pselaphacus signatus Guérin-Menneville 1841 3 1 1 1 2 1 0 0 Erotylidae, Erotylinae, Erotylini Thonius pavonius Lacordaire 1842 2 1 1 0 1 1 0 0 Erotylidae, Erotylinae, Erotylini Aegithus clavicornis Linnaeus 1758 3 1 1 1 2 1 0 0 Erotylidae, Erotylinae, Erotylini Coccimorphus dichrous Lacordaire 1842 4 1 1 1 2 1 0 0 Erotylidae, Erotylinae, Erotylini Coccimorphus unicolor Olivier 1807 w 4 1 1 1 2 1 0 0 Erotylidae, Erotylinae, Erotylini Gibbifer vicinus Guèrin-Mèneville 1841 2 1 1 1 2 1 0 0 Erotylidae, Erotylinae, Erotylini Erotylus histrio Linnaeus 1758 2 1 1 1 2* 1 0 0 Erotylidae, Erotylinae, Erotylini Erotylus giganteus Linnaeus 1758 w 3 1 1 1 2 1 0 0 Erotylidae, Erotylinae, Erotylini Homoeotelus dorbignyi Guèrin-Mèneville 1841 4 1 1 1 1 1 0 0 (continued) Annales de la Société entomologique de France (N.S.) 39

Table 1. (Continued). Systematic Assignment Species Lit. No. C1 C2 C3 C4 C5 C6 C7 Erotylidae, Erotylinae, Erotylini Homoeotelus testaceus Fabricius 1775 w 2 1 1 1 1 1 0 0 Erotylidae, Erotylinae, Erotylini Scaphidomorphus quinquepunctatus Fabricius 1775 w W 1 1 1 2 1 0 0 Erotylidae, Erotylinae, Encaustini Encaustes verticalis MacLeay 1825 w 3 1 1 0 2 1 0 0 Erotylidae, Erotylinae, Encaustini Aulacochilus quadrisignatus Guèrin-Mèneville 1841 w 2 1 1 0 2 0 0 0 Erotylidae, Erotylinae, Encaustini Micrencaustes lunulata MacLeay 1825 2 1 1 0 2 1 0 0 Erotylidae, Erotylinae, Megalodacnini Plagiopisthen paradoxus Thomson 1856 3 1 1 0 2 1 0 0 Erotylidae, Erotylinae, Megalodacnini Megalodacne heros Say 1823 5 1 1 0 1 1 1 0 Erotylidae, Erotylinae, Megalodacnini Megalodacne fasciata Fabricius 1777 w 5 1 1 0 1/2 1 1 0 Erotylidae, Erotylinae, Megalodacnini Coptengis sheppardi Pascoe 1860 w 4 1 1 0 1 1 0 0 Erotylidae, Erotylinae, Megalodacnini Episcapha quadrimacula Wiedemann 1823 w 4 1 1 0 2 0 0 0 Erotylidae, Erotylinae, Megalodacnini Linodesmus coecus Fabricius 1777 w 3 1 1 0 2 1 0 0 Erotylidae, Erotylinae, Megalodacnini Macrodacne luteoguttata Crotch 1876 2 1 1 0 2 1 0 0 Erotylidae, Erotylinae, Megalodacnini Scaphodacne rectesignata Crotch 1876 w W 1 1 0 2 1 0 0 Erotylidae, Erotylinae, Megalodacnini Scaphodacne rectesignata f. neglecta Heller 1918 2 1 1 0 2 1 0 0

CorelDraw version 12. Morphological terminology follows (6) Mental glands are located at the base of the men- Lawrence et al. (2010). tum, if present always as one pair (Figure 5K). The distribution of the previously reported abdominal glands (fig. 15 in Leschen 2003) was Results not examined in this study. Distribution and morphology of compound gland pores To consider the evolution of various glands in Erotylidae we used the following gland pores (mostly based on Nearly every examined specimen of Erotylidae was external inspection of the cuticle). found to possess gland pores in at least one of the above-mentioned body parts. There was a complete (1) Periocular glands are located on the frons antero- absence in the following taxa studied herein: Setariola, mesal to the compound eye and posterior to the Camptocarpus, Langurites, and Tetraphala. For the antenna base, if present always as one pair (Figures majority of the “languriid” Erotylidae several glands 4D, E, 5F; figs. 4, 8 in Węgrzynowicz 2002). were detected, but their distribution is ambiguous (2) Subgenal glands are located on the subgenal braces in most subfamilies. The non-erotylid outgroup species or inside the antennal groove on the subgenal here studied had no glands or gland pores at all process (ventrolaterally on the head; subocular (Table 1). pores), if present always as one pair (Figures 4B, Usually the gland pores in Erotylidae lack any special F, H, 5D, H, K; figs. 5, 9 in Węgrzynowicz 2002). modifications for the release or evaporation of the dis-

Downloaded by [79.238.118.44] at 23:24 17 July 2013 (3) Pronotal glands occur on the lateral margins of the charged secretion. In some exceptional cases, however, pronotum (Figure 3). If they are present, they the pores show a groove- or plateau-like extension, such always include one pair of pores each at the ante- as the pores at the lateral pronotal margin in Aulacochilus rior and posterior corners of the pronotum; addi- quadrisignatus (Figure 2A) and the subgenal pores of tional glands along the lateral pronotal edges in Encaustes verticalis (Figure 2B). These pores in E. verti- between can be present in varying numbers (1–19 calis (Figure 2B) and the periocular pores in Megischyrus per side). The gland pores at the anterior and brasiliensis (Figure 2C) show a pattern of near-circular posterior corners can be placed at different levels ridge-like elevations of the cuticle. In other cases the area with regard to the lateral edge of the pronotum; around the pore looks sponge-like; this is true for the accordingly, they are either visible from dorsally, subgenal pores of many taxa (e.g. Tritoma bipustulata, or not. Triplax russica and Pselaphacus nigropunctatus in (4) Prosternal glands are located on the ventral sur- Figures 2E, D, 5H). In many erotylids the sponge-like face of the prosternal process, if present always as gland pores are accompanied by groups of trichomes, one pair (Figures 5G, M; ﬁgs. 82, 83 in which might enhance the evaporation of the secretion; Węgrzynowicz 2002). such trichomes are frequently found associated with the (5) Mesoventral glands are located on the ventral pores on the subgenal brace. surface of the mesoventral process, if present Based on the occurrence of particular gland pores and always as one pair (Figure 5M). of trichomes associated with the subgenal brace pores we 40 K. Drilling et al.

Figure 1. Erotylid phylogeny by Robertson et al. (2004) based on 18S and 28S rDNA sequences. The two included “languriid” subgroups [Languria + Caenolanguria (Languriinae) and Toramus (Cryptophilinae)] are subordinate; the erotyline Dacne is sister to a clade comprising these “languriids” and the other Erotylinae. Downloaded by [79.238.118.44] at 23:24 17 July 2013

Figure 2. Modiﬁed gland pores in Erotylinae. A, Pore on left anterior corner of pronotum in Aulacochilus quadrisignatus, with groove- like extension (the smaller additional pore is one of those along the lateral pronotal edge; the anterior pronotal corner would follow at the left side). B, Pore on subgenal brace of Encaustes verticalis, with plateau-like adjacent area. C, Periocular pore of Megischyrus brasiliensis, with folded and wrinkled external part of outlet duct. D, Pore on subgenal brace of Triplax russica, associated with a group of trichomes. E, Pore on subgenal brace of Tritoma bipustulata, appearing sponge-like and associated with trichomes. Scales in µm. Annales de la Société entomologique de France (N.S.) 41 Downloaded by [79.238.118.44] at 23:24 17 July 2013 Figure 3. Distribution of gland openings on left lateral edge of pronotum in Erotylinae species; left margin of pronotum shown in lateral (C–F, N–P, T, U) or dorsal (A, B, G–M, Q–S, V) view, anterior at top. A–C, Dacnini: A, Dacne bipustulata, B,Nesitis celebesica, C, Combocerus glaber. D–F, Encaustini: D, Encaustes verticalis, E, Aulacochilus quadrisignatus, F, Micrencaustes lunulata. G–M, Tritomini: G, Tritoma bipustulata, H, Megischyrus brasiliensis, I, Amblyopus vittatus, J, Triplax russica, K, Pselaphacus nigropunctatus, L, Ischyrus quadripunctatus, M, Mycotretus ornatus. N–S, Megalodacnini: N, Megalodacne heros, O, Megalodacne fasciata, P, Coptengis sheppardi, Q, Episcapha quadrimaculata, R, Linodesmus coecus, S, Scaphodacne rectesignatus f. neglecta. T–V, Erotylini: T, Erotylus histrio, U, Ellipticus dorbignyi, V, Coccimorphus dichrous.

deﬁne the following characters C1–C7. The distribution of C3 Group of trichomes associated with pores on sub- character states across our sampling of taxa (and the genal brace: [0] absent, [1] present. erotylids additionally studied by Węgrzynowicz 2002 C4 Pores on lateral pronotal margins: [0] absent, [1] and Leschen 2003) is given in Table 1. only one in each anterior and posterior corner, [2] additional ones along the lateral margin. C1 Pores beside compound eyes: [0] absent, [1] pre- C5 Pores on prosternal process: [0] absent, [1] present sent as one pair. as one pair. C2 Pores on subgenal brace: [0] absent, [1] present as C6 Pores on mesoventral process: [0] absent, [1] pre- one pair. sent as one pair. 42 K. Drilling et al.

Figure 4. SEM pictures of typically gland-pore-bearing areas in “languriid” Erotylidae (with or without glandular pores). A, Setariola

Downloaded by [79.238.118.44] at 23:24 17 July 2013 sericea. The subgenal brace bears no glandular pore. B–D, Toramus pilifer: B, gland pore on subgenal brace (arrow); C, detail of lateral pronotal margin with gland pore (arrow); D, periocular pores (arrow) near the compound eye (at the left); the area around the pore appears ﬂuted. E-G, Languria bicolor: E, head with periocular pores (arrow); F, subgenal pore (arrow) beneath a compound eye; G, periocular pore continues internally into an elongated compound gland (arrow). H, Pentelanguria elateroides, subgenal pore (arrow). ce – compound eye; m – mentum; ma – mandible; sb – subgenal brace. Scales in µm.

C7 Pores on mentum: [0] absent, [1] present as one Cryptophilus integer. No gland pores are present (data pair. from Węgrzynowicz 2002). However, Leschen (2003) reported subgenal pores, pronotal pores, and mesoventral pores for this species. In Figure 6 we adopt data from External observations Leschen (2003). Setariola sericea. No gland pores are present (Figure Toramus pilifer. One pair each of periocular pores, 4A; same in Węgrzynowicz 2002). Leschen (2003) subgenal pores, and pronotal pores (Figures 4D, B, C). reported subgenal pores, pronotal pores, prosternal and Our ﬁndings differ from Węgrzynowicz (2002), who mesoventral pores for this species; these pores were reported no gland pores for this species. Leschen (2003) not detected in our analyses (due to the observation tech- reported periocular pores as absent. We examined ﬁve nique). In Figure 6A we adopt data from Leschen specimens by SEM and always found the same arrange- (2003). ment of gland pores (Table 1). Annales de la Société entomologique de France (N.S.) 43 Downloaded by [79.238.118.44] at 23:24 17 July 2013

Figure 5. SEM pictures and photographs of typically gland-pore-bearing areas in erotyline Erotylidae (with or without glandular pores). A–D, Dacne bipustulata exhibits neither A, periocular pores nor B, pores in the pronotal corners or C, at the prosternal process, but there are D, subgenal pores beneath the compound eye. E–G, Nesitis celebesica: E, numerous pores along the lateral pronotal margin (arrows; anterior to the right); F, a periocular pore; G, a pair of prosternal pores (arrows). H–J, Pselaphacus nigropunctatus: H, detail of a subgenal pore; within the depression numerous tiny pores are placed beside a group of trichomes; insert: overview over the subgenal brace; I, three pores (arrows) along the lateral pronotal margin (anterior to the right); J, the inner surface of the pronotum (photo taken through stereomicroscope); one branched gland each originates from the pronotal corner and from the pronotal margin (arrows; compare Figure 3K). K–L, Megischyrus brasiliensis: K, mental pores (arrows) and a subgenal pore with trichomes (arrowhead); L, the inner surface of the pronotum (photo taken through stereomicroscope); one branched gland each originates from the pronotal corner and from the pronotal margin (arrows; compare Figure 3H). M, Glandular pores on the prosternal and mesoventral processes in Megalodacne fasciata (arrows). N, The closely related Episcapha quadrimacula lacks glandular pores on the prosternal and mesoventral processes. c – coxa; ce – compound eye; m – mentum; ma – mandible; mv – mesoventrite; ps – prosternum; sb – subgenal brace. Scales in µm. 44 K. Drilling et al.

Tetraphala elongata. No gland pores are present. Cyrtomorphus pantherinus. Pronotal glands are repre- Camptocarpus longicollis. No gland pores are present. sented by a single pore in each anterior and posterior Langurites lineata. No gland pores are present. corner. One pair each of periocular pores, subgenal Pentelanguria elateroides. One pair of subgenal pores. pores, and prosternal pores. Languria bicolor. One pair each of periocular pores Cyrtomorphoides albicornis. Pronotal glands are (Figures 4E, G) and subgenal pores (Figure 4F); the area represented by a single pore in each anterior and posterior around the latter bears a tuft of trichomes (same in corner. One pair each of periocular pores, subgenal pores, Węgrzynowicz 2002). Leschen (2003) reported periocular and prosternal pores. pores as absent as well as pronotal pores and mesoventral Palaeolybas coccinelloides. Pronotal glands are repre- pores as present; these pores were not detected in our sented by a single pore in each anterior and posterior analyses. corner. One pair each of periocular pores, subgenal Languria puncticollis. One pair each of periocular pores, and prosternal pores. pores and subgenal pores. Callischyrus venustus. Pronotal glands are repre- Pachylanguria elongata. One pair each of periocular sented by a single pore in each anterior and posterior pores and subgenal pores (Figure 4H). corner plus a single one at the lateral margin. One pair Dacne bipustulata. One pair of subgenal pores (Figure each of periocular pores and subgenal pores. 5D). Our findings differ from Węgrzynowicz (2002), who Epytus violaceus. Pronotal glands are represented by a reported no gland pores for this species. We examined six single pore in each anterior and posterior corner. One pair specimens by SEM and found always the same arrange- each of periocular pores and subgenal pores. ment of gland pores (Table 1). Lybas bicolor. Pronotal glands are represented by a Dacne rufifrons. One pair of subgenal pores. single pore in each anterior and posterior corner. One pair Dacne notata. One pair each of periocular pores and each of periocular pores and subgenal pores. subgenal pores. Triplax russica. Pronotal glands are represented by a Combocerus glaber. Pronotal glands are represented single pore in each anterior and posterior corner (Figure by a single pore in each anterior and posterior corner 3J). One pair each of periocular pores and subgenal pores. (Figure 3C), anterior ones not visible from dorsal. One The area around the subgenal pores appears sponge-like pair each of periocular pores and subgenal pores (same in and bears additional trichomes (Figure 2D). Our finding of Węgrzynowicz 2002). additional trichomes differs from Węgrzynowicz (2002), Euzostria aruensis. Pronotal glands are represented by who coded such trichomes as absent in this species. We a single pore in each anterior and posterior corner, pores examined six specimens by SEM and always found tri- visible from dorsal. One pair each of periocular pores and chomes (Table 1). subgenal pores. Mycotretus ornatus. Pronotal glands are represented Triplatoma macleayi. Pronotal glands are represented by a single pore in each anterior and posterior corner by a single pore in each anterior and posterior corner, (Figure 3M). One pair of periocular pores and subgenal pores visible from dorsal. One pair each of periocular pores; the respective area bears a tuft of trichomes (same pores, subgenal pores, and prosternal pores. in Węgrzynowicz 2002).

Downloaded by [79.238.118.44] at 23:24 17 July 2013 Endytus bizonatus. Pronotal glands are represented by Triplacidea motschulskyi. Pronotal glands are repre- three pores per side. One pair each of periocular pores, sented by a single pore in each anterior and posterior subgenal pores, and prosternal pores. corner. One pair each of periocular pores, prosternal Nesitis celebesica. Pronotal glands are represented by pores, and subgenal pores; the respective area bears a 12 pores per side, distributed over the whole lateral margin tuft of trichomes pores (data from Węgrzynowicz of the pronotum (Figures 3B; 5E). One pair each of 2002). periocular pores (Figure 5F), prosternal pores (Figure Tritoma bipustulata. Pronotal glands are represented 5G), and subgenal pores. by a single pore in each anterior and posterior corner Nesitis sexnotata. Pronotal glands are represented by (Figure 3G). One pair each of periocular pores, subgenal seven pores per side, distributed over the whole lateral pores, and prosternal pores. The area around the subgenal margin of the pronotum. One pair each of periocular pores appears sponge-like and bears additional trichomes pores, prosternal pores (Figure 5G), and subgenal pores. (Figure 2E). Our ﬁnding of additional trichomes differs Apteronesitis fulgurata. Pronotal glands are repre- from Węgrzynowicz (2002), who coded such trichomes as sented by four pores per side, distributed over the lateral absent in this species. We examined 15 specimens by margin of the pronotum. One pair each of periocular SEM and always found trichomes (Table 1). pores, prosternal pores, and subgenal pores. Tritoma atripennis. Pronotal glands are represented by Cryptodacne synthetica. One pair of subgenal pores a single pore in each anterior and posterior corner. One (data from Węgrzynowicz 2002). pair each of periocular pores, subgenal pores, and proster- Thallis nigroaenea. One pair of subgenal pores. nal pores. The area around the subgenal pores appears Annales de la Société entomologique de France (N.S.) 45 Downloaded by [79.238.118.44] at 23:24 17 July 2013

Figure 6. A, Characters of pores mapped on a phylogeny of the Erotylidae based on 85 morphological characters (after Węgrzynowicz 2002; species added herein indicated by dashed lines). B, Hypothesis for the basal splitting events in Erotylidae derived from gland characters. Cr – Cryptophilinae; Da – Dacnini; En – Encaustini; Er – Erotylini; La – Languriinae; Me – Megalodacnini; Ph – Pharaxonothinae; T+ – Tritomini + remaining Erotylinae. 46 K. Drilling et al.

sponge-like and bears additional trichomes like in the in each anterior and posterior corner of the pronotum. We foregoing species. examined seven specimens by SEM and always found the Amblyscelis kelleni. Pronotal glands are represented additional pore in between (Table 1). Because of the phylo- by a single pore in each anterior and posterior corner. One genetic position of Pselaphacus (sister group to pair each of periocular pores, prosternal pores, and sub- Erotylini + Encasutini + Megalodacnini) we examined genal pores; the latter are associated with a tuft of tri- further representatives of this genus (P. giganteus, P. rubri- chomes (data from Węgrzynowicz 2002). culus, P. puncticollis). In all cases we found the same Amblyopus vittatus. Pronotal glands are represented arrangement of pronotal pores as in P. nigropunctatus. by a single pore in each anterior and posterior corner Pselaphacus signatus. Pronotal glands are represented (Figure 3I). One pair each of periocular pores, prosternal by a single pore in each anterior and posterior corner plus pores, and subgenal pores; the latter are associated with a a single one at the lateral margin (the same arrangement as tuft of trichomes (same in Węgrzynowicz 2002). in P. nigropunctatus). One pair each of periocular pores, Zythonia fulva. Pronotal glands are represented by a prosternal pores, and subgenal pores; the latter are asso- single pore in each anterior and posterior corner. One pair of ciated with a tuft of trichomes. periocular pores, and subgenal pores; the latter are associated Thonius pavoninus. Pronotal glands are represented with a tuft of trichomes (data from Węgrzynowicz 2002). by a single pore in each anterior and posterior corner. One Apolybas bicolor. Pronotal glands are represented by a pair each of periocular pores, prosternal pores, and sub- single pore in each anterior and posterior corner. One pair genal pores. each of periocular pores, mental pores, and subgenal pores Aegithus clavicornis. Pronotal glands are represented (data from Węgrzynowicz 2002). by eight pores per side, distributed over the whole lateral Megischyrus undatus. Pronotal glands are represented margin of the pronotum. One pair each of periocular by a single pore in each anterior and posterior corner. One pores, prosternal pores, and subgenal pores; the latter are pair each of periocular pores, mental pores, and subgenal associated with a tuft of trichomes. pores (data from Węgrzynowicz 2002). Coccimorphus dichrous. Pronotal glands are repre- Megischyrus brasiliensis. Pronotal glands are represented by seven pores per side, distributed over the whole sented by a single pore in each anterior and posterior lateral margin of the pronotum (Figure 3V). One pair each corner plus a single one at the lateral margin (Figures of periocular pores, prosternal pores, and subgenal pores; 3H, 5L). One pair each of periocular pores, mental the latter are associated with a tuft of trichomes. pores, and subgenal pores (Figure 5K); the latter are Coccimorphus unicolor. Pronotal glands are repre- associated with a tuft of trichomes. Periocular pores with sented by seven pores per side, distributed over the folded and wrinkled cuticular sculpture (Figure 2C). whole lateral margin of the pronotum. One pair each of Since the results differed between the two Megischyrus periocular pores, prosternal pores, and subgenal pores; the species (Figure 6A) we examined further representatives latter are associated with a tuft of trichomes (same in of this genus (M. semipunctatus, M. decempunctatus, Węgrzynowicz 2002). and M. discipennis). In all cases we always found the Gibbifer vicinus. Pronotal glands are represented by same arrangement of pronotal pores as in M. brasiliensis. 12 pores per side, distributed over the whole lateral margin

Downloaded by [79.238.118.44] at 23:24 17 July 2013 Ischyrus quadripunctatus. Pronotal glands are repre- of the pronotum. One pair each of periocular pores, pros- sented by a single pore in each anterior and posterior ternal pores, and subgenal pores; the latter are associated corner (Figure 3L). One pair each of periocular pores, with a tuft of trichomes. mental pores, and subgenal pores; the latter are associated Erotylus giganteus. Pronotal glands are represented with a tuft of trichomes (same in Węgrzynowicz 2002). by nine pores per side, distributed over the whole lateral Because of the phylogenetic position of Ischyrus and the margin of the pronotum. One pair each of periocular presence of an additional pronotal pore in the closely pores, prosternal pores, and subgenal pores; the latter are related M. brasiliensis (also in other species of associated with a tuft of trichomes (same in Megischyrus) we examined further representatives of this Węgrzynowicz 2002). genus (I. scriptus, I. femoralis, and I. ﬂavitarsis). In all Erotylus histrio. Pronotal glands are represented by 13 cases we always found the same arrangement of pronotal pores per side, distributed over the whole lateral margin of pores as in I. quadripunctatus. the pronotum (Figure 3T). One pair each of periocular Pselaphacus nigropunctatus. Pronotal glands are pores, prosternal pores, and subgenal pores; the latter are represented by a single pore in each anterior and posterior associated with a tuft of trichomes. corner plus a single one at the lateral margin (Figures 3K, 5I, Ellipticus dorbignyi. Pronotal glands are represented J). One pair each of periocular pores, prosternal pores, and by a single pore in each anterior and posterior corner subgenal pores; the latter are associated with a (Figure 3U). One pair each of periocular pores, prosternal tuft of trichomes (Figure 5H). Our ﬁndings differ from pores, and subgenal pores; the latter are associated with a Węgrzynowicz (2002), who reported merely a single pore tuft of trichomes. Annales de la Société entomologique de France (N.S.) 47

Ellipticus testaceus. Pronotal glands are represented Macrodacne luteoguttata. Pronotal glands are repre- by a single pore in each anterior and posterior corner. One sented by seven pores per side, distributed over the whole pair each of periocular pores, prosternal pores, and sub- lateral margin of the pronotum. One pair each of periocu- genal pores; the latter are associated with a tuft of tri- lar pores, subgenal pores, and prosternal pores. chomes (same in Węgrzynowicz 2002). Plagiopisthen paradoxus. Pronotal glands are repre- Scaphidomorphus quinquepunctatus. Pronotal glands sented by 11 pores per side, distributed over the whole are represented by seven pores per side, distributed over lateral margin of the pronotum. One pair each of periocular the lateral margin of the pronotum. One pair each of pores, prosternal pores (Figure 5G), and subgenal pores. periocular pores, prosternal pores, and subgenal pores; Scaphodacne rectesignata. Pronotal glands are repre- the latter are associated with a tuft of trichomes (data sented by seven pores per side, distributed over the whole from Węgrzynowicz 2002). lateral margin of the pronotum. One pair each of periocu- Encaustes verticalis. Pronotal glands are represented lar pores, subgenal pores, and prosternal pores (data from by nine pores per side, distributed over the whole lateral Węgrzynowicz 2002). margin of the pronotum (Figure 3D). One pair each of Scaphodacne rectesignata f. neglecta. Pronotal glands periocular pores, subgenal pores, and prosternal pores are represented by 11 pores per side, distributed over the (same in Węgrzynowicz 2002). Subgenal pores have a whole lateral margin of the pronotum (Figure 3S). One pair plateau-like adjacent area (Figure 2B). each of periocular pores, subgenal pores, and prosternal pores. Aulacochilus quadrisignatus. Pronotal glands are represented by 11 pores per side, distributed over the Internal observations whole lateral margin of the pronotum (Figure 3E). ﬁ Pronotal pores within the anterior corner have a groove- For species listed below pores were con rmed as glandu- like extension (Figure 2A). One pair each of periocular lar in the way described in section before. pores and subgenal pores (same in Węgrzynowicz 2002). Languria bicolor. Periocular pores continue internally Micrencaustes lunulata. Pronotal glands are repre- into elongated compound glands (Figure 4G). sented by two larger pores within the anterior and the Nesitis celebesica. Pores along the lateral margin of posterior corner and 19 tiny pores distributed over the the pronotum continue internally into compound glands whole lateral margin (Figure 3F). One pair each of perio- with a brush-like structure. cular pores, subgenal pores, and prosternal pores. Triplax russica. Pores in the anterior and posterior Megalodacne heros. Pronotal glands are represented corners of the pronotum continue internally into compound by a single pore in each anterior and posterior corner glands with a brush-like structure (Drilling et al. 2010). (Figure 3N). One pair each of periocular pores, subgenal Tritoma bipustulata. Pores in the anterior and poster- pores, prosternal pores, and mesoventral pores. ior corners of the pronotum as well as of the prosternal Megalodacne fasciata. Pronotal glands are represented process continue internally into compound glands with a by a single pore in each anterior and posterior corner (Figure brush-like structure (glands studied in detail in Drilling 3O). One pair each of periocular pores, subgenal pores, et al. 2010). prosternal pores, and mesoventral pores (Figure 5M). Our Megischyrus brasiliensis. Both the pore in the poster-

Downloaded by [79.238.118.44] at 23:24 17 July 2013 ﬁndings differ from Węgrzynowicz (2002), who reported ior pronotal corner and that on the posterior part of the numerous pores along the lateral margin of the pronotum. lateral pronotal margin continue internally into compound We examined ﬁve specimens by SEM and found always glands with a brush-like structure (Figure 5L, arrow). only a single pore in each pronotal corner (Table 1). Pselaphacus nigropunctatus. Same as for foregoing Coptengis sheppardi. Pronotal glands are represented species (Figure 5J, arrow); both glands are branched. by a single pore in each anterior and posterior corner (Figure Erotylus histrio. Pores along the lateral margin of the 3P). One pair of periocular pores, subgenal pores, and pros- pronotum continue internally into small compound glands ternal pores (same in Węgrzynowicz 2002). with a brush-like structure. Episcapha quadrimacula. Pronotal glands are represented by three pores per side, one each in the anterior and Discussion posterior corners and one in the posterior half of the lateral margin (Figure 3Q). One pair of periocular pores and sub- Previous hypotheses on the phylogeny of Erotylidae genal pores (same in Węgrzynowicz 2002; Figure 5N demon- The currently most elaborate hypotheses on Erotylidae strates the absence of the prosternal and mesoventral pores). phylogeny are those of Robertson et al. (2004) based on Linodesmus coecus. Pronotal glands are represented DNA-sequences (18S and 28S rDNA sequences; Figure 1), by 11 pores per side, distributed over the whole lateral and Węgrzynowicz (2002; Figure 6A) and Leschen (2003) margin of the pronotum (Figure 3R). One pair each of based on morphology; the latter treat mainly the former periocular pores, subgenal pores, and prosternal pores languriid subfamilies. However, there are some pivotal (same in Węgrzynowicz 2002). differences between these hypotheses: 48 K. Drilling et al.

(1) Robertson et al. (2004) find the taxon Particular morphological differentiations of similarly Cryptophilinae (Toramus) subordinate in the located pores are viewed as confirming the assumption Erotylinae–Tritomini, but no morphological char- of homology but are quite rare. acters have been proposed in support of this rela- In the various Erotylidae glands were found on the lateral tionship (Węgrzynowicz 2002; Robertson et al. edges of the pronotum (either only in the anterior and poster- 2004). In Węgrzynowicz (2002) as well as in ior corners, or additionally in between), beside the eyes, on Leschen (2003) Cryptophilinae are placed outside the subgenal braces, on the mentum, and on the prosternal of the Erotylinae, associated with other “langur- and mesoventral processes. However, in all Erotylidae only iid” lineages. Apart from that difference, basal part of this gland set is present, while some lack these glands relationships are Dacnini + (Languria bico- altogether (e.g., some “languriid” taxa). lor + remaining Erotylinae) in both trees For several cucujiform subgroups compound glands (Węgrzynowicz 2002; Robertson et al. 2004; have been reported (e.g., Discolomatidae, while Dacne is the only true dacnine taxon repre- Cryptophagidae, Endomychidae or Boganiidae (Leschen sented in both trees). In Leschen (2003) the 2003)), but these mostly open in positions completely Languriinae are also placed outside of different to Erotylidae (e.g., behind seventh visible Erotylinae, but nested within the Loberinae (fig. abdominal sternite in many Tenebrionidae (Doyen 1966; 109 in Leschen 2003). Tschinkel 1975), mesothoracic ventral midline in (2) The Erotylinae–Tritomini are paraphyletic in all Meloidae and Anthicidae (Morgan 1968; Berrioz-Oritz hypotheses, but while in Robertson et al. (2004) 1985; Hemp & Dettner 1997), on the labium (the cuticle they are paraphyletic only with regard to Erotylini of the labium shows numerous openings of glands, but no (and Cryptophilinae), the Encaustini and further details are provided) and the maxillary palps in Megalodacnini are additionally nested in Tritomini Semiadalia (Coccinellidae, Barbier et al. 1992), and the in the hypothesis of Węgrzynowicz (2002) and middle of first visible abdominal sternite in Ciidae (Buder Leschen (2003), where Erotylini, Encaustini, and et al. 2008; also in a few Erotylinae, but not studied Megalodacnini together form a clade. herein)). Glands that open in a similar position (e.g. the (3) Coptengis, formerly assigned to Dacnini, is placed pronotal ones) occur in taxa which are not closely in the Encaustini according to Robertson et al. 2004; related to Erotylidae (Cryptophagidae, Discolomatidae, sister to Aulacochilus), but in the Megalodacnini Endomychidae; Leschen 2003). This implies that not a according to Węgrzynowicz (2002). single homologue is known for erotylid glands in a closely (4) Each of the tribes Erotylini, Encaustini, and related cucujiform group. Consequently, the possession of Megalodacnini is monophyletic in all hypotheses these glands would appear as (a set of) autapomorphies (except for the case of Coptengis), but only in evolved within Erotylidae. Węgrzynowicz (2002) and Leschen (2003) do the three tribes form a clade, with Pselaphacus nigropunctatus (Tritomini) forming the sister taxon to Glands on lateral edge of pronotum (C4) this clade in Węgrzynowicz (2002). Nearly all examined Erotylinae possess these glands;

Downloaded by [79.238.118.44] at 23:24 17 July 2013 only the Dacnini species Dacne bipustulata (Figure 3A), The taxon sample used herein shows much overlap D. ruﬁfrons, D. notata, Thallis nigroaenea and Cryptodacne with that in Węgrzynowicz (2002) (see Table 1). In the synthetica (Figure 6A), as well as Languria bicolor (placed discussion of our gland characters and gland evolution we within Erotylinae), lack them (state [0] of character). thus mainly refer to his phylogenetic hypothesis (Figure Regarding the presence of pronotal gland pores, two cases 6A, where characters are mapped on that tree but addi- can be distinguished: either there is only one pore present in tional taxa studied herein are supplemented). The taxon each pronotal corner (state [1]), or additional pores are pre- sampling of Robertson et al. (2004) differs strongly; hence sent along the lateral pronotal edge (state [2]). Since the it is difﬁcult to devolve the gland characters to their examined outgroup species as well as most members of the phylogeny (Figure 1). basal erotylid subfamily Xenoscelinae lack these glands (present in Protoloberus according to Leschen 2003), this suggests that the absence of pronotal glands represents the Homology and systematic distribution of compound ancestral state. Within the Pharaxonothinae, Languriinae, glands and Cryptophilinae this character is polymorphic (Leschen Erotylidae compared with other cucujiform beetles 2003), but pores should be present in all examined species of Gland pores present on corresponding parts of the body Loberinae (Leschen 2003). Since the character is unfortu- are here considered homologous as long as there is no nately coded as present or absent (Leschen 2003) it is not contradictory evidence from phylogenetic analyses (pri- possible to estimate the arrangement and quantity for parti- mary homology hypotheses sensu De Pinna 1991). cular species. Annales de la Société entomologique de France (N.S.) 49

In Dacnini all three states of this character occur due to several independent losses have apparently occurred the likely incorrect placement of some species in Dacnini (Figure 6A). For the tribe Encaustini Leschen (2003) (Euzostria, Combocerus, Nesitis, Apteronesitis; Figure coded this character as absent since only Aulacochilus 6A). However, absence of pores is the predominant con- was included in his analysis. dition within Dacnini. One pore in each pronotal corner was mostly found in the likely paraphyletic Tritomini; only Callischyrus venustus, Megischyrus brasiliensis (as Glands on mesoventral process (C6) well as further species of this genus; see Results: Among Erotylidae we found such glands only in the two Distribution and morphology of compound gland pores) examined Megalodacne species from Megalodacnini. and both Pselaphacus species (sister group to the clade Leschen (2003), however, reported mesoventral pores for comprising Erotylini, Encaustini, and Megalodacnini) all tribes of Erotylinae (with the exception of Dacnini) and have a single additional pore in the posterior part of the absence of mesoventral pores for the erotylid subfamilies lateral edge (Figures 3H, K, 5I, J, L; in contrast to our Xenoscelinae and Cryptophilinae. In contrast to our results ę result W grzynowicz (2002) found in Pselaphacus nigro- (Figure 6A), the distribution of mesoventral pores within punctatus just one pore in each pronotal corner (Table 1). Languriinae, Pharaxonothinae and Loberinae was stated as Such conditions could possibly represent a transition polymorphic (Leschen 2003). from pores only in the pronotal corners (state [1]) to numerous pores along the pronotal edge (state [2]). Most species of the derived tribes Erotylini, Glands beside compound eyes – periocular glands (C1) Encaustini, and Megalodacnini possess numerous pores along the pronotal edge (Figure 6A) – with only few The periocular glands are found in nearly all examined exceptions: both examined Ellipticus species and Erotylinae as well as in species of Cryptophilinae Thonius among the Erotylini, as well as both (Toramus; Figure 5B) and Languriinae (both Languria Megalodacne species (in contrast to our result, in species and Pachylanguria; Figure 5F; we found these M. fasciata Węgrzynowicz (2002) found numerous pores pores in both examined species of Languria; Leschen along the pronotal edge; Table 1) and Coptengis sheppardi (2003) coded this character as absent for this genus). ﬁ among the Megalodacnini have pores only in the pronotal The Dacnini species Dacne bipustulata, D. ru frons and corners (Figure 6A); these cases likely represent second- Cryptodacne as well as the other erotylid subfamilies ary reductions. (Xenoscelinae, Pharaxonotinae and Loberinae) lack them (Leschen 2003).

Glands on prosternal process (C5) – A pair of such glands was found in most examined mem- Glands on subgenal brace subocular glands (C2) bers of a clade comprising Erotylini, Encaustini, and A pair of subgenal glands is present in all examined Megalodacnini (absent in Aulacochilus and Episcapha; Erotylinae as well as in most taxa studied herein (Table clade C in Figure 6A). Within Tritomini this character is 1). Only in Setariola (Pharaxonothinae), Camptocarpus,

Downloaded by [79.238.118.44] at 23:24 17 July 2013 polymorphic since these glands are absent in Epytus, Langurites and Tetraphala (all Languriinae) are these Lybas, Callischyrus, Triplax, Mycotretus, Zythonia, pores absent. Leschen (2003) stated this character as pre- Apolybas, Megischyrus, and Ischyrus. Among the other sent for all erotylid subfamilies, excluding the erotyline subfamilies prosternal glands are absent in the basal tribes Tritomini, Encaustini and Erotylini (absent or poly- Xenoscelinae, Pharaxonothinae (in contrast to our results, morphic). The absence of these glands in a few species is Leschen (2003) reported these glands as present for likely due to secondary losses, since this character is Setariola, Pharaxonothinae), Cryptophilinae (present in already present in the basal subfamily Xenoscelinae. Loberoschema; Leschen 2003), and Erotylinae–Dacnini The subgenal glands or rather pores are occasionally (present in Nesitis (Figure 5G) and Apteronesitis). associated with a group of trichomes (C3) (Figures 2D, However, pores are sporadically present in a few species E, 5H, K). The trichomes are found in most Tritomini of Loberinae (except for Telmatoscius claviger Sharp and (except Apolybas, Cyrtomorphus, Cyrtomorphoides, Paphezia detritophila Zablotny & Leschen) and Palaeolybas, Callischyrus,andMegischyrus undatus) Languriinae (Leschen 2003). as well as all examined Erotylini. For the derived Based on the presence of prosternal glands and glands Encaustini and Megalodacnini trichomes are reduced on the lateral edge of the pronotum (previous character), (Figure6A).IncontrasttoWęgrzynowicz’s (2002) the species Nesitis celebesica, N. sexnotata, and results, we also found trichomes on the subgenal braces Aperonesitis fulgurata appear misplaced in Dacnini. of the Tritomini species Tritoma bipustulata and Triplax Within the clade Erotylini + Encaustini + Megalodacnini russica (Figure 2D, E, Table 1). This character is not (clade C; Figure 6A) as well as the polymorphic Tritomini treated in Leschen (2003). 50 K. Drilling et al.

Glands on mentum (C7) and various Dacnini, and therefore their assumed These glands are restricted to a Tritomini clade comprising phylogenetic position. A hypothetical relationship ﬁ Apolybas bicolor, Megischyrus undatus, M. brasiliensis, Pentelanguria + Dacne bipusutulata + D. ru frons + and Ischyrus quadripunctatus (Figure 6A). This character Cryptodacne + Thallis +(D. notata + Pachylanguria + L. is also not treated in Leschen (2003). puncticollis + ((Combocerus + (Toramus + Cryptophilus) + (L. bicolor + Tritomini))) derived from glandular characters (as shown in Figure 6B), would be incongruent with the Phylogenetic implications and evolution of glands monophyly of Dacnini (Languriinae and Cryptophilinae ę In Figure 6A gland characters are mapped on the phylo- inside Dacnini) proposed by W grzynowicz (2002). genetic tree of Węgrzynowicz (2002), with character Nonetheless, the hypothesis of monophyletic Dacnini is transformations placed on the tree in the most parsimo- poorly supported by apomorphies: the mesepisterna are nious way. Additional taxa are included. Characters are fused to the mesepimera far in front of the meso-metathor- here essentially interpreted along Węgrzynowicz’s (2002) acic suture inside the coxal cavities, and cross-veins r3 and hypothesis. r4 approach or touch one another. Thus, there are hardly any counter-arguments against the relationships suggested by gland characters. As mentioned above, both Węgrzynowicz Basal relationships in Erotylinae (2002) and Robertson et al. (2004) found that Coptengis has Since the used outgroup lacks glands on respective body to be removed from Dacnini. According to the gland parts, the various glands in Erotylidae can be seen as characters studied here, Nesitis, Apteronesitis, Endytus, autapomorphies. However, the different glands found in Euzostria,andTriplatoma arealsohighlyunlikelytobe the family originate at different phylogenetic levels. related to any of the other genera assigned to Dacnini Erotylidae in this broad sense is supported by the presence (Nesitis, Apteronesitis,andEndytus are instead placed in of subgenal glands (C2). The few cases of absence are the Encaustini + Megalodacnini clade; clade G), and even most parsimoniously explained as secondary reductions. if limited to Dacne and Cryptodacne, Dacnini may be Moreover, the large group of Erotylidae excluding the paraphyletic based on the presence of pronotal and proster- basal subfamily Xenoscelinae is supported by the presence nal glands (C4, 5). of large prosternal glands (C4), present in pronoral angles The proposed position of Languria bicolor (character state 1). However, these glands are reduced in (Languriinae) inside the Erotylinae (Węgrzynowicz several lineages, especially in the phytophagous Langiriinae 2002; Robertson et al. 2004) is conﬁrmed by the presence and basal Erotylinae (Figure 6A). Furthermore, the pre- of periocular glands (C1) and trichomes next to the sub- sence of periocular glands (C1) supports the monophyly genal pores (C3). Pachylanguria, L. puncticollis and of the mycophagous subfamilies Erotylinae and Toramus lack these trichomes; the other Languriinae Cryptophilinae including the phytophagous species (sensu Leschen 2003) examined here, Tetraphala, Languria bicolor (clade A in Figure 6A). Other languriine Langurites, and Camptocarpus, lack all glands. This species studied herein (Languria puncticollis and suggests that these taxa are placed outside Erotylinae, and Pachylanguria; both taxa were not included in previous Languriinae results as non-monophyletic. Unfortunately,

Downloaded by [79.238.118.44] at 23:24 17 July 2013 phylogenetic analyses of Węgrzynowicz (2002), Leschen Tetraphala, Langurites and Camptocarpus species were (2003), and Robertson et al. (2004)) also offered these not included in the phylogenetic analyses of glands, suggesting a position similar to that of L. bicolor Węgrzynowicz (2002), Leschen (2003), and Robertson et (see below) or at least inside the Dacnini (Figure 6A). al. (2004). Leschen’s (2003) arguments for monophyletic Finally, the monophyly of a clade Languia Languriinae are the presence of submesocoxal lines, an bicolor + (Tritomini + remaining erotyline tribes; clade B apical pit of the spermatheca, and a wedge cell in the in Figure 6A) is supported by the presence of trichomes wing venation. In sum, the monophyly of Languriinae beside the above-mentioned subgenal pores. These tri- appears at least debatable. chomes are, however, reduced in the highly derived Encaustini and Megalodacnini (clade G in Figure 6A). Such tufts, located on different parts of the body, are wide- Apical relationships in Erotylinae spread among fungivorous beetles and enable transfer of The clade comprising the paraphyletic Tritomini as spores or fragments of spawn (Węgrzynowicz 2002). For well as Encaustini, Erotylini, and Megalodacnini Erotylinae an enhancement of the evaporation of discharged (and Cryptophilinae in Robertson et al. 2004; clade secretions is also conceivable since these trichomes are C in Figure 6A) likely has a groundplan set of glands located near the openings of the subgenal glands. that includes glands in the corners of the pronotum However, there are some points of ambiguity at the (C4) as well as periocular (C1) and subgenal glands base of the tree that mainly concern the occurrence of (C2); the latter associated with a tuft of trichomes particular glands in species of the former Languriidae (C3; Figure 6A). Annales de la Société entomologique de France (N.S.) 51

One character conflict at the base of Tritomini con- origins of these glands should be assumed and the useful- cerns the distribution of prosternal glands (C5; Figure 6A). ness of this character for phylogenetic conclusions would Węgrzynowicz’s (2002) clade comprising Tritomini, be limited. Erotylini, Encaustini, and Megalodacnini is weakly sup- In conclusion, the found distributions of compound ported by the presence of these glands (clade C in Figure glands are very conflicting in several cases and are most 6A, where the “accelerated” version of character transfor- satisfyingly explainable by the present phylogenies mation is mapped). However, considering the basal (Węgrzynowicz 2002; Robertson et al. 2004) for the sub- dichotomies, prosternal glands are consistently absent in family of Erotylinae. There is much convergence in the Apolybas + Ischyrus + Megischyrus (clade E in Figure evolution of the glandular equipment, mainly concerning 6A), while they are present in the respective sister clades the distribution of exocrine glands within the basal sub- Mycotretus + Triplacidea + Tritoma + Amblyscelis + families. Within the Erotylinae several lineages may be Amblyopus + Zythonia (clade D in Figure 6A; absent in supported by some of the glandular characters. However, Mycotretus and Zythonia) and Pselaphacus + Erotylini + evolution of exocrine glands seems to be strongly affected Encaustini + Megalodacnini (clade F in Figure 6A; with a by ecological factors and conclusions on phylogenetic few absences appearing as secondary, e.g. Aulacochilus relationships may be constricted. Unfortunately, little is quadrisignatus, and Episcapha quadrimaculata). Several known about the biology and ecology of many of the of the additional taxa lack these glands also (Epytus, erotylid genera and species, and virtually nothing about Lybas, and Callischyrus). Therefore, gland characters the contents and functional role of their various glands. would rather suggest that the two latter clades (clades D This presently hampers estimations of how plausibly sec- and F) form a monophyletic group. ondary losses of glands could be explained as resulting The clade Apolybas + Ischyrus + Megischyrus (clade from changes in life history. Indeed, gland characters E in Figure 6A) is supported by a gland apomorphy: the alone are hardly qualified to resolve the phylogeny of presence of glands on the mentum (C7; Figures 5K, 6A). Erotylidae; at most they are applicable in some apical This character was found in all examined members of erotyline clades (see also Tschinkel 1975; Tschinkel & the clade and seems to be unique among the family Doyen 1980 and Steidle & Dettner 1993 for gland char- Erotylidae. Skelley (1998) also mentioned these pores in acters as a phylogenetic tool). Nonetheless, a large-scale his revision of the genus Ischyrus. Admittedly, however, analysis on the structure, ultra-structure and chemistry of their occurrence does not seem stable since some Ischyrus such glands would surely provide additional interesting and species seem to lack them (Węgrzynowicz 2002). crucial characters for resolving phylogenetic problems A clade comprising the derived Tritomini Pselaphacus within this family, and would help to explain possible path- nigropunctatus as well as the examined Erotylini, ways in the evolution of the different glandular systems. Encaustini, and Megalodacnini (clade F in Figure 6A) is supported by the presence of additional glands (at least one) along the lateral pronotal edge. However, our finding Acknowledgments for P. nigropunctatus differs from that of Węgrzynowicz We would like to thank Dr. C. Schmidt (Senckenberg Natural (2002), who found only glands in the pronotal corners; History Collections Dresden) for his critical reading of the manuscript. Downloaded by [79.238.118.44] at 23:24 17 July 2013 this may indicate intraspecific variability, which would limit the usefulness of this character for phylogenetic conclusions (we examined seven specimens and found References always the same arrangement of glands; Figures 3K, 5I, Alvarenga M. 1994. Catálogo dos Erotylidae (Coleoptera) J). McHugh et al. (1997) already discuss the variability of Neotropical. Revista Brasileira de Zoologia. 11:1–175. the number of pores along the pronotal margins of Barbier R, Ferran A, Le Lannic J, Allo M-R. 1992. Morphology and ultrastructure of integumentary glands of Semiadalia Megalodacne heros. The presence of glands only in the undecimnotata Schn. (Coleoptera: Coccinellidae). pronotal corners in both examined Ellipticus species International Journal of Insect Morphology and among Erotylini and Coptengis as well as both examined Embryology. 21;3:223–234. Megalodacne species among Megalodacnini is also con- Berrioz-Ortiz A. 1985. The presence of mesothoracic glands in flicting, yet this is likely explainable as secondary losses. Epicauta segmenta (Coleoptera: Meloidae) and other blister beetles. Journal of the Canadian Entomological Society. Finally, in our sample both Megalodacne species 58:179–181. (Megalodacnini) had a pair of mesoventral glands, which Boyle WW. 1956. A revision of the Erotylidae of America north show identical location in both taxa. In contrast, Leschen of Mexico (Coleoptera). Bulletin of the American Museum (2003) reported mesoventral glands also for a few species of Natural History. 110:61–172. of the other erotylid subfamilies (Pharaxonothinae, Buder G, Grossmann C, Hundsdoerfer A, Klass K-D. 2008. A contribution to the phylogeny of the Ciidae and its relation- Loberinae) as well as for all Erotylinae in his sample ships with other cucujoid and tenebrionoid beetles (see Discussion: Homology and systematic distribution (Coleoptera: Cucujiformia). Arthropod Systematics and of compound glands). In this case several independent Phylogeny. 66:165–190. 52 K. Drilling et al.

Chûjô M, Chûjô MT. 1988. A catalog of the Erotylidae (Insecta, Leschen RAB, Lawrence JF, Slipinski A. 2005. Classification of Coleoptera) from the old world (excl. the Ethiopian region). basal Cucujoidea (Coleoptera: Polyphaga), cladistic analysis, Esakia. 26:139–185. keys and review of new families. Invertebrate Systematics. Chûjô M, Chûjô MT. 1989. A catalog of the Erotylidae (Insecta, 19:17–73. Coleoptera) from the old world (excl. the Ethiopian region) Leschen RAB, Skelley PA, McHugh JV. 2010. 10.7. Erotylidae II. Esakia. 28:75–96. Leach, 1815. In: Leschen, RAB. Beutel RG. Lawrence, JF, Chûjô M, Chûjô MT. 1990. A catalog of the Erotylidae (Insecta, editors. Coleoptera, Beetles, Vol. 2: Morphology and Coleoptera) from the old world (excl. the Ethiopian region) systematics (Elateroidea, Bostrichiformia, Cucujiformia III. Esakia. 29:1–67. partim). Handbuch der Zoologie/Handbook of Zoology Crowson RA. 1955. The natural classification of the families of IV/39. Berlin, New York: Walter de Gruyter; p. 311–319. Coleoptera. London: Nathaniel Lloyd and Co. 187 p. Leschen RAB, Ślipiński A. 2010. 7. Cucujiformia Lameere, Delkeskamp K. 1981. Erotylidae von Afrika und Madagascar, 65 1938: Introduction, Phylogeny. In: Leschen RAB. Beutel p. In: Wilcox JA, editor. Coleopterorum catalogus supple- RG. Lawrence JF, editors. Coleoptera, Beetles, vol. 2: menta.The Hague: Pars 34. Dr W. Junk Publishers; 65 p. Morphology and systematics (Elateroidea, Bostrichiformia, De Pinna M. 1991. Concepts and tests of homology in the Cucujiformia partim). Handbuch der Zoologie/Handbook of cladistic paradigm. Cladistics. 7:367–394. Zoology IV/39. Berlin, New York: Walter de Gruyter; Doyen JT. 1966. The skeletal anatomy of Tenebrio molitor p. 227–228. (Coleoptera: Tenebrionidae). Miscellaneous Publications of Leschen RAB, Węgrzynowicz P. 1998. Generic catalogue and the Entomological Society of America. 5;3:103–150. taxonomic status of Languriidae (Cucujoidea). Annales Drilling K, Dettner K, Klass K-D. 2010. Morphology of the prono- Zoologici. 48;3/4:221–243. tal compound glands in Tritoma bipustulata (Coleoptera: McHugh JV, Marshall CJ, Fawcett FL. 1997. A study of adult Erotylidae). Organisms, Diversity & Evolution. 10:205–214. morphology in Megalodacne heros (Say) (Coleoptera: Hemp C, Dettner K. 1997. Morphology and chemistry of Erotylidae). Transactions of the American Entomological mesothoracic glands in anthicid beetles (Coleoptera: Society. 123;4:167–223. Anthicidae). Entomologia Generalis. 22;2:97–108. Morgan A. 1968. The occurance of a pair of exocrine glands in Hunt T, Bergsten J, Levkanicova Z, Papadopoulou A, St John O, the metathorax of Lytta nutalli (Coleoptera: Meloidae). Wild R, Hammond PM, Ahrens D, Balke M, Caterino MS et al. Canadian Journal of Zoology. 46:915–917. 2007. A comprehensive phylogeny of beetles reveals the evo- Robertson JA, McHugh JV, Whiting MF. 2004. A molecular lutionary origins of a superradiation. Science. 318:1913–1916. phylogenetic analysis of the pleasing fungus beetles Klausnitzer B. 2005. 26. Ordnung: Coleoptera, Käfer; Part: (Coleoptera: Erotylidae): evolution of colour patterns, gre- Polyphaga. In: Dathe HH, editor. Lehrbuch der Speziellen gariousness and mycophagy. Systematic Entomology. Zoologie. Band I, 5. Teil: Insecta. 2nd ed. Heidelberg, 29:173–187. Berlin: Spektrum Akademischer Verlag; p. 452–526. Sen Gupta T, Crowson RA. 1969. On a new family of Lawrence J, Beutel RG, Leschen RAB, Ślipiński A. 2010. 2. Clavicornia (Coleoptera) and a new genus of Languriidae. Glossary of Morphological Terms. In: Leschen RAB. Beutel Proceedings of the Royal Entomological Society London RG. Lawrence JF, editors. Coleoptera, Beetles, Vol. 2: (B). 38:125–131. Morphology and systematics (Elateroidea, Bostrichiformia, Sen Gupta T, Crowson RA. 1971. A review of classification of Cucujiformia partim). Handbuch der Zoologie/Handbook of the family Languriidae (Coleoptera: Clavicornia) and the Zoology IV/39. Berlin, New York: Walter de Gruyter; p. 9–20. place of Languriidae in the natural system of Clavicornia. Lawrence JF, Newton AF. 1982. Evolution and classification of Memoirs of the Zoological Survey of India. 15:1–42. beetles. Annual Review of Ecology and Systematics. Skelley PE. 1998. Revision of the genus Ischyrus Lacordaire 13:261–290. (1842) of North and Central America (Coleoptera: Lawrence JF, Ślipiński A, Seago AE, Thayer MK, Newton AF, Erotylidae: Tritominae). Occasional Papers of the Florida

Downloaded by [79.238.118.44] at 23:24 17 July 2013 Marvaldi AE. 2011. Phylogeny of the Coleoptera based on State Collection of Arthropods. 9:134 p. morphological characters of adults and larvae. Annales Steidle JLM, Dettner K. 1993. Chemistry and morphology of the Zoologici. 61(1):1–217. tergal gland of freeliving adult Aleocharinae (Coleoptera: Lawrence JF, Vaurie P. 1983. A catalog of the Coleoptera of Staphylinidae) and its phylogenetic signiﬁcance. Systematic America north of Mexico. Family: Languriidae. Washington, Entomology. 18:149–168. DC: United States Department of Agriculture. Washington, Tschinkel WR. 1975. A comparative study of the chemical DC. defensive system of tenebrionid beetles. 3. Morphology of Leschen RAB. 1996. Phylogeny and revision of the genera of the glands. Journal of Morphology. 145:355–370. Cryptophagidae (Coleoptera: Cucujoidea). Kansas Science Tschinkel WR, Doyen JT. 1980. Comparative anatomy of the Bulletin. 55:549–634. defensive glands, ovipositors and female genital tubes of Leschen RAB. 2003. Erotylidae (Insecta: Coleoptera: Cucujoidea): tenebrionid beetles (Coleoptera). International Journal of phylogeny and review. Fauna of New Zealand. 47:108 p. Morphology and Embryology. 9:321–368. Leschen RAB, Buckley TR. 2007. Multistate characters and diet Węgrzynowicz P. 2002. Morphology, phylogeny and classiﬁca- shifts: evolution of Erotylidae (Coleoptera). Systematic tion of the family Erotylidae based on adult characters Biology. 56:97–112. (Coleoptera: Cucujoidea). Genus. 13;4:435–504.