Zootaxa 4097 (1): 041–058 ISSN 1175-5326 (print edition) http://www.mapress.com/j/zt/ Article ZOOTAXA Copyright © 2016 Magnolia Press ISSN 1175-5334 (online edition) http://dx.doi.org/10.11646/zootaxa.4097.1.2 http://zoobank.org/urn:lsid:zoobank.org:pub:6189C0A9-0BDA-4A8E-83B4-717C7A6EDA2B Bryophyte-feeding of Litoleptis (Diptera: Rhagionidae) with descriptions of new species from Japan YUME IMADA1 & MAKOTO KATO Graduate School of Human and Environmental Studies, Kyoto University, Yoshida-nihonmatsu-cho, Sakyo-ku, Kyoto, 606-8501, Japan 1Corresponding author. E-mail: [email protected] Abstract Here we report the larval phytophagous habit of Litoleptis for the first time, and describe six new species of Litoleptis in Japan; L. japonica n. sp., L. kiiensis n. sp., L. niyodoensis n. sp., L. himukaensis n. sp., L. izuensis n. sp., and L. asterel- laphile n. sp. All the species described here are thallus-miners of liverworts belonging to Aytoniaceae and Conocephal- aceae (Marchantiopsida: Marchantiophyta). Each fly species mined thalli of only one of the following genera: Conocephalum, Reboulia, and Asterella. The descriptions of the Japanese Litoleptis species here expand the concept of this genus. The female genital morphology of Litoleptis strengthened the current placement of Litoleptis as a member of Spaniinae. Key words: herbivory, Brachycera, Hilarimorpha, bryophytivore, leaf-mining, keys Introduction Diptera accounts for one of the most speciose clades of organisms and displays extremely broad ecological breadth (Grimaldi & Engel 2005; Wiegmann et al. 2011). A major expansion of dipteran families took place in mid- Cretaceous (Yeates 2002), which coincides with the age of angiosperm radiation (Grimaldi 1999). Association of Brachycera and flowering plants date back at least to the mid-Cretaceous (Ren 1998). About half of 13 major clades of brachycera are known to contain important pollinators (Friis et al. 2011) and some brachyceran families have diversified as angiosperm-feeders (Winkler et al. 2009; Novotny et al. 2005; Labandeira 2005), even though the number of species associating with plants is inferred to rather be underestimated (Friis et al. 2001). Lower Brachycera became rich in some assemblages of the Middle and Late Triassic (Krzemiński & Krzemińska 2003), although the origin and phylogenetic relationships of this group are scarcely understood. In particular, Rhagionidae sensu lato (i.e. Rhagionidae, Austroleptidae, and Bolbomyiidae) have experienced the greatest diversification during Middle and Late Jurassic (Mostovski 2000) and have lessened their impact in the modern fauna. The natural history of Rhagionidae s. l. has historically attracted much attention, although the life histories of most species are largely unknown (Oldroyd 1964). As adults, at least some rhagionids are zoophages. Spaniopsis and Symphoromyia have blood-sucking species (Colless & McAlpine 1991; Ferguson 1915; Turner & William 1979). Female flies of Rhagio are believed to be insect predators (Kellogg 1908; Leonard 1930; Narchuk 1988; Paramonov 1962), albeit not proven (Kerr 2010). Two genera, Arthroteles and Austroleptis are reported as flower- visitors (Colless & McAlpine 1991; Stuckenberg 1956). Larval feeding habits of many of rhagionid flies are poorly known. Rhagio and Chrysopilus are mostly terrestrial and are known as predatory (Paramonov 1962; Roberts 1969). Chrysopilus contains terrestrial, rotten wood feeders, and some aquatic species (Roberts 1969; Tsacas 1962; Nagatomi 1958). Some species of Austroleptis are thought to feed on rotten wood or detritus (Colless & McAlpine 1991). Some Symphoromyia species are inferred to be detritivores or partly associated with decayed plants (Sommerman 1962). Spania nigra and Ptiolina obscura are reported as herbivores of bryophytes (Mik 1896; Brindle 1959; Brauer 1883; Lane & Anderson 1982; Nartshuk 1995). Accepted by C. Lamas: 12 Feb. 2016; published: 30 Mar. 2016 41 Considering the variety of natural history of Rhagionidae s. l., revealing species diversity, systematics and feeding ecology of this group is important in understanding the evolutionary trajectory of flies. Nevertheless, the systematic position and arrangement of Rhagionidae s. l. have been unsettled throughout the history of study since Latreille (1802). Litoleptis is a small genus, which comprises at least four species. It was erected as a monotypic genus, which proposed for L. alaskensis Chillcott, 1963. Hennig (1972) added a new species L. chilensis Hennig and transferred Hilarimorpha orientalis Frey to Litoleptis. Lastly, Arillo et al. (2009) described L. fossilis, which is the first fossil species of this genus preserved in the Lower Cretaceous amber. In addition, three undescribed species of Litoleptis were recorded: two species from Japan and Nepal by Nagatomi (1982), and one from Yunnan, China by Yang et al. (1997). Collection records of Litoleptis are thus sparse and its biology has not been reported so far. The systematic position of Litoleptis has not yet been clarified due to the anomalous morphology and the scarcity of the specimens. This genus was originally included in Rhagionidae by Chillcott (1963). Nevertheless Litoleptis resembles Hilarimorpha Schiner (Hilarimorphidae) in wing venation and in lacking tibial spurs, it is differentiated by the short-styled antenna and the presence of a pulvilliform empodium (Chillcott 1963). Grimaldi & Cumming (1999) noted that Litoleptis is in close proximity to Bolbomyia and Austroleptis based on the lack of wing vein M3 (i.e. the lack of dm cell). It was placed in ‘Spania group’ by Hennig (1973) and subsequently was included in the subfamily Spaniinae (Rhagionidae) by Nagatomi (1982), along with Ptiolina, Spania, and Spaniopsis. Spaniinae was defined primarily based on the characters in the female genitalia besides the antennal form: cerci widely separated each other, tergum 10 absent or short (Nagatomi & Iwata 1976; Nagatomi 1982). Rhagionidae was a heterogenous group, of which sufficient autapomorphies were not identified (Stuckenberg 2001), with subfamilies separated by the shape of female genitalia (Nagatomi 1982). Each subfamily of Rhagionidae (i.e. Rhagioninae, Spaniinae, Austroleptinae) was raised to family rank so as to constitute the respective monophyly (Stuckenberg 2001). The concept of Spaniidae as a family has been rejected by Kerr (2010) because it was incompatible with the concept of Rhagionidae in the result of the combined analyses of morphological and molecular data. This taxonomic group has been instead treated as a subfamily of Rhagionidae, Spaniinae, adding Omphalophora and Symphoromyia (Kerr 2010). Litoleptis has been consistently placed in Spaniinae since Nagatomi (1982), nevertheless its female genital morphology, the most important character for separating the subfamilies of Rhagionidae (Nagatomi 1982), has never been examined. Examining the female genital morphology of Litoleptis is therefore necessary to understand the relationship between Litoleptis and the peripheral genera. From our recent survey of liverwort-dwelling insects, we have found that the larvae of Litoleptis mining within thalli (i.e. leaf-like organ of thallose liverworts) of liverworts are prevalent in Japan. To access the regional species diversity and to reveal the larval diet of Litoleptis, we conducted field observation and larval rearing. All the species obtained were new to science, which were found to be thallus-miners that feed on only single genus of liverwort species. Here we describe six new species of Litoleptis, and discuss the systematic position and concept of Litoleptis with special reference to female genitalia. This study provides a unique example of bryophyte-mining flies, which has long been overlooked despite widespread in Japan. Materials and methods We collected and reared the bryophyte-mining larvae of Litoleptis in Japan, thereby obtaining adult flies. We collected liverworts from 24 localities in Japan (Fig. 1) and kept the plant material until the larvae exited the plant and emerged as adults. For larval rearing, the plant material was packed in small plastic cases and occasionally moistened, and kept under laboratory conditions. Emerged adults were completely dried in a refrigerator. We examined 223 adult specimens of Litoleptis for this study. For specimen preparation, we removed the abdomen, placed it in 10% potassium hydroxide warmed in a hot water bath (70~75°C) for approximately 30 minutes to remove adipose tissue, and then washed it with distilled water and subsequently 50% ethanol to clean out debris. After dissection of the genitalia by separating the epandrium from the gonocoxites, the specimen was mounted in glycerine on a glass slide and observed under a microscope with transillumination. 42 · Zootaxa 4097 (1) © 2016 Magnolia Press IMADA & KATO FIGURE 1. Map showing the sampling localities. Locality numbers correspond to those in the text. All the specimens examined in this study are deposited in the following collections: National Museum of Nature and Science (NMNS), Graduate School of Human and Environmental Studies, Kyoto University, Kyoto, Japan (KUHE). All specimens studied have a serial number label added in the following format: “Rh XXXX”. The terminology follows Kerr (2010) with minor modifications. The abbreviation “Pref” using below stands for “prefecture”. The author’s names are abbreviated: MK and YI stand for Makoto Kato and Yume Imada, respectively. Taxonomy Litoleptis Chillcott Type species: Litoleptis alaskensis Chillcott