FNZ 47 - Erotylidae (Insecta: Coleoptera: Cucujoidea) - Introduction
Leschen, RAB 2003. Erotylidae (Insecta: Coleoptera: Cucujoidea): phylogeny and review. Fauna of New Zealand 47, 108 pages.
(
ISSN 0111-5383 (print),
;
no.
47.
ISBN 0-478-09350-0 (print),
).
Published 05 Jun 2003
ZooBank: http://zoobank.org/References/F7BD20B2-D210-4D51-9CB2-1BC51DF75F2C
Introduction
Groups of microcoleoptera, especially those in the superfamily Cucujoidea, are taxonomically challenging because similar body forms exist among completely unrelated lineages and they require careful dissection to confirm taxonomic placement. Confusion about the family limits of Cryptophagidae, Erotylidae, Languriidae and other cucujoids has resulted in a complex history of paraphyletic groupings requiring detailed phylogenetic study. The objectives of this study were three-fold:
- reconstruct the phylogeny of Languriidae and Erotylidae using adult characters,
- provide a higher classification for these families, and
- review the species of Languriidae sensu lato occurring in New Zealand.
Leschen & Wegrzynowicz (1998) reviewed many of the problems with the higher classification of Languriidae and their treatment serves as an appropriate introduction to this work. The phylogenetic study of Languriidae presented here grew from previous work on the systematics of Cryptophagidae because many languriid taxa were described as cryptophagids, and members of both families are usually misidentified together in insect collections. While Cryptophagidae appears to be a monophyletic group (Leschen 1996), a two-century long controversy remains about the status of Languriidae: this family is either monophyletic or should be included in a broadly defined family Erotylidae (see references in Leschen & Wegrzynowicz 1998). A cladistic analysis based on genera and adult characters forms a major portion of this study to support the monophyly of Erotylidae plus Languriidae as well as determine the monophyly of the higher taxa within these two families. An arrangement based on the phylogenetic relationships shown in this study will, in turn, provide a better classification for placing the New Zealand species in a global taxonomic context. In light of the new phylogenetic information I broaden the taxonomic concept of Erotylidae to include all Languriidae.
In this paper the New Zealand species that were included in the Languriidae sensu lato are reviewed and a future publication to be co-authored with P. Skelley will treat the species of erotylids currently placed in Cryptodacne and Thallis.
SYNOPSIS OF THE SYSTEMATIC PROBLEMS
Leschen & Wegrzynowicz (1998), in their review of Languriidae classification, included comments on salient characters of Languriidae and the discussion here is limited to major points relevant to questions to be addressed by the cladistic analysis. The major issues concern the monophyly of the higher taxa at the familial, subfamilial, and tribal levels (current arrangement of Languriidae is provided in Table 1).
The family Erotylidae, which originally included the large-bodied, colourful, and plant-feeding members of Languriidae (Crotch 1876; Gorham 1887a, b; Fowler 1908), was considered separate from languriids by Crotch (1873) and this classification was followed by others (Arrow 1925; Crowson 1952; Schenkling 1923, 1928; Sen Gupta & Crowson 1971; Lawrence & Newton 1982, 1995; Pakaluk et al. 1995). These two families were separated largely on the basis of different biologies (Lewis 1884), with Erotylidae being mycophagous and Languriidae being phytophagous. This two-family system spans two centuries (Lawrence et al. 1995), and when the classification of Languriidae was re-examined in light of taxa transferred to it from Cryptophagidae, the morphological grade between Languriidae and Erotylidae remained (Leschen & Wegrzynowicz 1998). Some authors (e.g., Rymer Roberts 1939, 1958; Lawrence 1991; Sen Gupta & Crowson 1971; Leschen & Wegyrnowicz 1998) have questioned the separation of Erotylidae and Languriidae, though the two-family system is followed presently. While there are unambiguous characters that support the monophyly of Erotylidae (Leschen & Wegrzynowicz 1998, Lawrence 1991), Languriidae is not a monophyletic taxon, as shown below.
Several problems exist with the monophyly of the family-group taxa of Languriidae (Leschen & Wegrzynowicz 1998). A key problem is the monophyly of the heterogenous subfamily Xenoscelinae: synapomorphies have not been identified for the subfamily, the monophyly of all of the three tribes is questionable, the pharaxonothine genus Loberopsyllus may be a member of Cryptophilinae (Leschen & Ashe 1999), and the widespread and diverse genus Loberus may be paraphyletic with respect to similar genera that were described as distinct (Leschen & Wegrzynowicz 1998). The two tribes of Languriinae (Languriini and Cladoxenini) which contain the highest diversity in Languriidae are thought to be paraphyletic (Crowson 1955, Sen Gupta & Crowson 1971).
While the emphasis of this study is adult morphology, larval characters have been used to clarify the relationships among some of the higher taxa (e.g., Rymer Roberts 1939, 1958; Sen Gupta & Crowson 1971) and information about larval work is contained in Leschen & Wegrzynowicz (1998) and is not repeated here. Meanwhile, a phylogenetic study using larval characters is in progress by Joseph McHugh.
TAXONOMIC HISTORY OF NEW ZEALAND LANGURIIDAE (now Erotylidae)
The Languriidae fauna of New Zealand is relatively small and consists of only 8 endemic species (Watt 1982, Klimaszewski & Watt 1997). It is, however, an important group historically and taxonomically, and from two scientific perspectives led by two pairs of contemporaries. David Sharp and Thomas Broun (late 1800's and early 1900's) described the New Zealand species, and later Roy Crowson and Tapan Sen Gupta (mid 1960's and early 1970's) placed the fauna in a world classification of Cucujoidea.
The New Zealand languriid species were originally described as members of the families Cryptophagidae or Cucujidae. David Sharp (1876, 1886) described two species (now placed Loberus) in the holarctic genus Telmatophilus (Cryptophagidae), the only species of Hapalips in the mediterranean genus Xenoscelis (Cucujidae), and erected a genus Cathartocryptus for the species C. obscurus (Cucujidae), a species that was described previously by Broun in Paramecosoma. Thomas Broun (1881, 1893) was busy describing the entire beetle fauna and placed his languriid species into holarctic cryptophagid genera (Cryptophagus anthracinus, Paramecosoma maculosa, Telmatophilus vestitus, and T. olivascens).
The incorrect placement of New Zealand taxa may not have been a case of messy taxonomy for the industrious Major Broun or the distinguished and world-renowned Sharp, but rather a conservative decision to accept the premodern concepts of cucujoid classification. Sharp influenced Broun's career in entomology (Zimmerman 1993) and the classifications accepted by these coleopterists reflected the uncertain status Cryptophagidae, Cucujidae, and Languriidae and their subordinate genera. Concepts of these taxa were questionable at the time and in some respects continue to be so.
While Broun and Sharp were working, Telmatophilus was a paraphyletic genus consisting of members of Cryptophagidae (the true Holarctic Telmatophilus is in Cryptophagidae) and many taxa now included in Loberus. Polyphyletic groups like Telmatophilus existed because the concept of homology in Victorian England and elsewhere was very different from that accepted today, especially in connection with classification. The distinction between homology and analogy was being discussed fervently, typically in light of Darwinian theory (Russell 1916), and there was no adequate empirical method to distinguish between these two classes of similarity. For example Sharp thought that Xenoscelis prolixus (now in Hapalips) resembled Xenoscelis deplanatus (Wollaston), a taxon once included in Cucujidae. These two species are dorsoventrally compressed (as is the genus Cathartocryptus, which was also once included in Cucujidae) and the morphological connection between similar and unrelated taxa made by Sharp resulted in an error in classification recognised later using other methods that differentiate homology and analogy. Also, the cultural goals of Victorian taxonomists were different from those of modern systematists, and Broun and Sharp were trying to describe as quickly and as adequately as possible the new taxa brought to them by local and foreign naturalists. Accusing these predecessors of their taxonomic errors is easy, but we have the luxury of modern approaches, training, and skills that did not exist over 100 years ago.
In the 1900's the use of light microscopy and dissection increased, and entomologists began examining characters in more detail and testing existing taxonomies. Roy Crowson, whose influence on modern Coleoptera classification was unprecedented, came to New Zealand in the 1950's to collect its rich relictual fauna (Leschen 2000) and together with, or in an advisory role to, his student Tapan Sen Gupta provided the first modern attempts to classify Languriidae into monophyletic taxa. (Actually, Sen Gupta was supposed to be working on a world review of Cryptophagidae for his Ph.D. thesis at the University of Glasgow, but spent most of his time transferring cryptophagid species to Languriidae and working on the classification of this group.) Sen Gupta (1968a) transferred Sharp's Xenoscelis prolixus to Hapalips (subgenus Xenosceloides), and with Crowson (1969), erected Loberonotha for Broun's Telmatophilus species, and placed this genus in the tribe Loberonothini. The work of Crowson and Sen Gupta is considered in more detail elsewhere (Leschen & Wegrzynowicz 1998).
Crowson and Sen Gupta are credited for being the first modern systematists to study Languriidae, but Crowson was philosophically opposed to cladistics, a "postmodern" method developed in the mid to late 1900's and credited to Willi Hennig (see the English translation of Hennig 1966). Cladistics provides systematists with a rigorous method to distinguish between similarity (or analogy) and homology (synapomorphy), the latter of which is taken as evidence for the monophyly of a group. The resultant tree is a cladogram that is produced by grouping taxa by synapomorphies and minimising the number of character changes over the entire tree. Though he did recognise the need to differentiate between primitive and derived characters, one reason why Crowson was opposed to cladistics was that character weighting could not be applied explicitly and he preferred a detailed narrative approach to homology and classification (Crowson 1991a). Crowson also acknowledged that a character could be derived in one group and primitive in another group, a pattern that "sometimes leads to false conclusions" (Sen Gupta & Crowson 1971: 29). Though Sen Gupta & Crowson (1971) did not produce a phylogenetic tree that represented languriid relationships, they provided three box or chart diagrams. Each chart was a sort of periodic table of selected Cucujoidea that tabulated larval and adult characters. The cladistic analysis that follows is a third phase in the development of a classification for Languriidae.
An important and somewhat neglected addition to this short history is the contribution made by Charles Watt, one of my predecessors at NZAC. Watt was New Zealand's specialist on tenebrionoids (among other groups) and was very active in collection management and curation. He made many unpublished observations on NZAC specimens and indicated syntypes that he recognised were original specimens contained in Broun's type series. These labels are easily identified and have handwritten labels with locality (usually one word) and Broun's unique species numbers followed by a full stop (.) (T. K. Crosby, pers. comm.). Watt (1982) transferred two of David Sharp's species of Telmatophilus to Loberus (unaware of Bruce 1952b) and, based on his curation labels, he thought that C. anthracinus should be a new genus of Languriidae. However, C. anthracinus is actually a member of Loberus and is similar to other species, especially to one montane South American form that is also apterous.