Phylogeny of the Yponomeuta species (Lepidoptera, Yponomeutidae) and the history of their host plant associations Sandrine A. Ulenberg This study presents the results of cladistic analyses of the morphology of 1) the subfamilies of the Yponomeutidae, 2) the genera of the Yponomeutinae, and 3) the species of Yponomeuta Latreille, 1796. The sequential steps in the evolution of the host plant associations, the presumed key factor in the processes of speciation within Yponomeuta, as extrapolated from the cladograms are discussed. The hypothesis that the present-day host plant associations evolved from an ancestral relation with Celastraceae through speciation in allopatry mostly on Euonymus is supported by the underlying study. The biogeographical patterns suggest speciation through dispersion from Australia, the Oriental region, Africa to western Europe. Sandrine A. Ulenberg, Zoological Museum, University of Amsterdam, Plantage Middenlaan 64, 1018 DH Amsterdam, The Netherlands. [email protected] Introduction (1998) “These studies were based on the working The study of speciation mechanisms as a means to hypothesis that the present-day host plant associa- gain a better understanding of the origins of bio- tions evolved from an ancestral relation with Celas- logical diversity has always been a key area in bio- traceae through speciation in allopatry mostly on logical research. Recently, the role of ecological spe- Euonymus and through host shifts in sympatry or cialisation has received considerable attention. Good allopatry to mainly Rosaceae, following the scenario theoretical (Dieckmann & Doebeli 1999; Doebeli of host race formation (Bush 1975)”. In the present & Dieckmann 2000; Janz & Nylin 2008; Percy et paper the validity of this working hypothesis is tested al. 2004; Via 2001) as well as empirical progress is by analysing the most probable sequence of the evo- being made, in particular in our understanding of lutionary changes in the host plant associations of the evolution of insect–plant relations and speciation Yponomeuta based on their species phylogeny. in phytophagous insects (Kölsch & Pedersen 2008; Via 2002; Via & Hawthorne 2002). Yponomeuta’s distribution and host plant One of the model systems developed for this pur- associations pose is the genus Yponomeuta Latreille, 1796 and its The species of the genus Yponomeuta are phytopha- host relations. Since the 1970s many multidiscipli- gous and have a wide, mainly palaearctic, distribu- nary studies have been dedicated to the associations tion. The genus is represented in all major eco regions between Yponomeuta species and their host plants. except South America and Antarctica. Host-associa- The main objectives of studying this model were to tions are known for 39 of the 76 species. Of the 39 obtain insights into the evolution of the host rela- Yponomeuta species with a known host-association, tions, and the speciation processes that have led to 32 are mono- or oligophagous within one genus the present-day associations (Bakker et al. 2008; of trees or shrubs. Of those 32, 27 species feed on Hora & Roessingh 1999; Menken 1996; Menken et Celastraceae genera, of which 22 exclusively feed on al. 1992; Menken & Roessingh 1998; Raijmann & species of the genus Euonymus. The Yponomeuta spe- Menken 2000). According to Menken & Roessingh cies that do not use Celastraceae as their host plant Tijdschrift voor Entomologie 152: 187–207, Figs 1–5, Tables 1–5. [ISSN 0040–7496]. http://www.nev.nl/tve © 2009 Nederlandse Entomologische Vereniging. Published 1 December 2009. Downloaded from Brill.com09/29/2021 07:19:23AM via free access 188 Tijdschrift voor Entomologie, volume 152, 2009 feed on Crassulaceae, Rosaceae, Rhamnaceae, or plesiomorphic state, was incorporated based on Salicaceae (see Gershenson & Ulenberg 1998). Kyrki’s opinion on plesio- and apomorphies of the subfamilies. This hypothetical ancestor is chosen Classification of the Yponomeutidae as outgroup because the group most related to the The systematic position of the (sub)families within Yponomeutidae is unknown. In this and the fol- the superfamily Yponomeutoidea was open to dis- lowing analyses, PAUP* 4.0b10 (Swofford 2003) cussion until recently and differently treated by was used, all characters were unordered and of equal systematists, according to the historical review weight. An exhaustive search was made for the most in Gershenson & Ulenberg’s (1998) book on the parsimonious tree. Yponomeutinae. No general consensus exists regarding the ranks of The genera of the Yponomeutinae the higher taxa in Yponomeutidae. Some authors The phylogenetic relationships of the yponomeutine (Kyrki 1990; Moriuti 1977) lowered the position of genera was examined utilizing a maximum parsi- what was formerly regarded as a well-defined fam- mony analysis of the type species of the genera in the ily, the Yponomeutidae, to the level of subfamily. subfamily. The search was heuristic. Kyrki (1984) discussed the different classifications of The characters of the type species were taken from earlier authors and in his paper published in 1990 the literature, viz., Diakonoff (1967), Friese (1960), tentatively reclassified the Holarctic Yponomeutoi- Gates Clarke (1965), Gershenson (1974, 1990), dea, its families and subfamilies based on what he Gershenson & Ulenberg (1998), Huemer & Tar- regarded as apomorphies. Kyrki’s reclassification was mann (1991), Koster (1990), Koster & Schreurs followed by Scoble (1992) and Dugdale et al. (1998) (1992), Moriuti (1971, 1977) and Zagulajev (1990). in their treatment of the Yponomeutoidea. Heppner Most of the characters were checked on collection (1998), in his general classification of Lepidoptera, material. Of four genera no specimens of the type summarized the characters defining Yponomeutidae. species could be studied, viz., Epinomeuta trun- However, though he recognized three subfamilies catipennella Rebel, 1936 (fossil), Opsiclines leuco- (viz. Yponomeutinae, Saridoscelinae and Cedesti- morpha (Lower, 1900), Paradoxus osyridellus Millière, nae), he did not treat these in detail. Neither of them 1869, Parahyponomeuta egregiellus (Duponchel, published on the phylogeny of the groups within the [1839]). Yponomeutidae. As the literature did not give sufficient insight in their characters, neither of other species classified in these four genera, they were not incorporated in the Material and Methods analysis. The subfamilies of the Yponomeutidae and the gen- Saridoscelis sphenias Meyrick, 1894, the type species era of the Yponomeutinae were analysed cladistically of the only genus classified in the Saridoscelinae, the to determine the sister group of Yponomeuta and Yponomeutinae’s presumed sister group, served as the history of the host plant associations within the outgroup. Yponomeutinae leading to those in Yponomeuta. For a description of the characters and the matrix, see Table 2a and 2b. The subfamilies of the Yponomeutidae A heuristic search for the most parsimonious tree was The relationship between the subfamilies of the performed with PAUP* 4.0b10 (Swofford 2003), Yponomeutidae has been investigated largely based all characters were unordered and of equal weight. on the characters given by Kyrki (1990). One char- Starting trees were obtained via stepwise addition acter state was changed, viz., the number of segments with random sequence. The number of replicates of the maxillary palp in the Yponomeutinae. Kyrki was one million. The branch-swapping algorithm gives the number as four. According to the observa- was tree-bisection-reconnection (TBR). tions of Gershenson & Ulenberg (1998) the number of segments in this subfamily is less than four, as they The species of Yponomeuta did not count the protuberance of the stipes as basal The phylogeny of the Yponomeuta species treated in maxillary segment (see also Moriuti 1977; Dugdale Gershenson & Ulenberg (1998) was investigated, et al. 1998). Heppner (1998: 51) mentioned 1–2 for viz., the Yponomeuta species of the world exclusive the family Yponomeutidae (Yponomeutinae, Sarido- of the Americas. One of the six Nearctic Yponomeuta scelinae, Cedestinae). species, Y. multipunctellus Clemens, has been incor- For a description of the characters and the matrix, porated in the cladistic analyses. The other five spe- see Table 1. cies have been left out because of the author’s insuf- In the analysis an all-zero outgroup, with zero as ficient insight in the American fauna. Downloaded from Brill.com09/29/2021 07:19:23AM via free access Ulenberg: Phylogeny of Yponomeuta 189 A B Sc R1 R2 R3 R4 R5 M1 M2 A1 M3 A2+3 Cu1 Cu2 termen maxillary palpus labial palpus C Sc+R1 R M1 M2 Cu1 A2 A1 galea (tongue) Cu2 D E F socius uncus papillae analis apophysis posterioris valva cornuti ductus bursae lamella postvaginalis tegumen sacculus antrum gnathos vinculum aedeagus saccus scape ductus bursae apophysis anterioris signum Fig. 1A-C. Yponomeuta malinellus. – A, head; B, fore wing; C, hind wing. D-E. Yponomeuta padellus (L.) D, male genitalia; E, female genitalia. F. Yponomeuta morbillosus Zel., bursa copulatrix. Eumonopyta unicornis Moriuti, 1977 (male), Pti- favillaceus Meyrick, 1923 loteina melanaster (Meyrick, 1907) and Teinoptila grossipunctellus Guenée, 1879 interruptella Sauber, 1902 served as outgroups. leucothorax Meyrick, 1913 (American) The following species have not been incorporated in liberalis Meyrick, 1913 the analyses because no specimens could be investi- madagascariensis Gershenson, 2003 gated and thus no characters checked. The five not martinellus