One hundred and one new species of Trigonopterus weevils from New Guinea

Abstract A species discovery and description pipeline to accelerate and improve taxonomy is outlined, relying on concise expert descriptions, combined with DNA sequencing, digital imaging, and automated wiki species page creation from the journal. One hundred and one new species of Trigonopterus Fauvel, 1862 are described to demonstrate the feasibility of this approach: Trigonopterus aeneipennis sp. n., Trigonopterus aeneus sp. n., Trigonopterus agathis sp. n., Trigonopterus agilis sp. n., Trigonopterus amplipennis sp. n., Trigonopterus ancoruncus sp. n., Trigonopterus angulatus sp. n., Trigonopterus angustus sp. n., Trigonopterus apicalis sp. n., Trigonopterus armatus sp. n., Trigonopterus ascendens sp. n., Trigonopterus augur sp. n., Trigonopterus balimensis sp. n., Trigonopterus basalis sp. n., Trigonopterus conformis sp. n., Trigonopterus constrictus sp. n., Trigonopterus costatus sp. n., Trigonopterus costicollis sp. n., Trigonopterus crassicornis sp. n., Trigonopterus cuneipennis sp. n., Trigonopterus cyclopensis sp. n., Trigonopterus dentirostris sp. n., Trigonopterus discoidalis sp. n., Trigonopterus dromedarius sp. n., Trigonopterus durus sp. n., Trigonopterus echinus sp. n., Trigonopterus edaphus sp. n., Trigonopterus eremitus sp. n., Trigonopterus euops sp. n., Trigonopterus ferrugineus sp. n., Trigonopterus fusiformis sp. n., Trigonopterus glaber sp. n., Trigonopterus gonatoceros sp. n., Trigonopterus granum sp. n., Trigonopterus helios sp. n., Trigonopterus hitoloorum sp. n., Trigonopterus imitatus sp. n., Trigonopterus inflatus sp. n., Trigonopterus insularis sp. n., Trigonopterus irregularis sp. n., Trigonopterus ixodiformis sp. n., Trigonopterus kanawiorum sp. n., Trigonopterus katayoi sp. n., Trigonopterus koveorum sp. n., Trigonopterus kurulu sp. n., Trigonopterus lekiorum sp. n., Trigonopterus lineatus sp. n., Trigonopterus lineellus sp. n., Trigonopterus maculatus sp. n., Trigonopterus mimicus sp. n., Trigonopterus monticola sp. n., Trigonopterus montivagus sp. n., Trigonopterus moreaorum sp. n., Trigonopterus myops sp. n., Trigonopterus nangiorum sp. n., Trigonopterus nothofagorum sp. n., Trigonopterus ovatus sp. n., Trigonopterus oviformis sp. n., Trigonopterus parumsquamosus sp. n., Trigonopterus parvulus sp. n., Trigonopterus phoenix sp. n., Trigonopterus plicicollis sp. n., Trigonopterus politoides sp. n., Trigonopterus pseudogranum sp. n., Trigonopterus pseudonasutus sp. n., Trigonopterus ptolycoides sp. n., Trigonopterus punctulatus sp. n., Trigonopterus ragaorum sp. n., Trigonopterus rhinoceros sp. n., Trigonopterus rhomboidalis sp. n., Trigonopterus rubiginosus sp. n., Trigonopterus rubripennis sp. n., Trigonopterus rufibasis sp. n., Trigonopterus scabrosus sp. n., Trigonopterus scissops sp. n., Trigonopterus scharfi sp. n., Trigonopterus signicollis sp. n., Trigonopterus simulans sp. n., Trigonopterus soiorum sp. n., T sordidus sp. n., Trigonopterus squamirostris sp. n., Trigonopterus striatus sp. n., Trigonopterus strigatus sp. n., Trigonopterus strombosceroides sp. n., Trigonopterus subglabratus sp. n., Trigonopterus sulcatus sp. n., Trigonopterus taenzleri sp. n., Trigonopterus talpa sp. n., Trigonopterus taurekaorum sp. n., Trigonopterus tialeorum sp. n., Trigonopterus tibialis sp. n., Trigonopterus tridentatus sp. n., Trigonopterus uniformis sp. n., Trigonopterus variabilis sp. n., Trigonopterus velaris sp. n., Trigonopterus verrucosus sp. n., Trigonopterus violaceus sp. n., Trigonopterus viridescens sp. n., Trigonopterus wamenaensis sp. n., Trigonopterus wariorum sp. n., Trigonopterus zygops sp. n.. All new species are authored by the taxonomist-in-charge, Alexander Riedel.


introduction
The number of undescribed species on Earth is immense (Scheffers et al. 2012). Large scale studies on morphology, functional biology, community ecology, and phylogeny lead to the discovery of large numbers of new species, but suffer from the lack of a sound taxonomic foundation. DNA barcoding and molecular biodiversity assessment studies do indeed suffer from the same issue. The reason is apparent -it is comparably easy to collect many species and create large datasets, but it is not so easy to identify them, especially if numerous samples from tropical localities are involved. It is easy to obtain hundreds or thousands of DNA sequences or insect samples for a beta diversity study, even for a student project. Identification of samples from moderately to poorly studied regions, and more specifically the recognition and formal scientific description of new species however require taxonomic expertise.
Here, we will not review the significant body of literature addressing the various suggestions how to overcome the "taxonomic impediment". Rather we report a species discovery and description pipeline (Riedel et al. 2013) that accelerates and improves the way taxonomy flanks research in related disciplines such as biogeography, phylogenetics and not the least community ecology. The term "turbo-taxonomy" was coined for a similar procedure describing 178 new species of parasitic wasps (Butcher et al. 2012) and is discussed below. When faced with a large number of morphologically similar, undescribed species, it is not an option to carry on "business as usual" and prepare very detailed descriptions with an output of only few species per year. Such a strategy will not achieve a sustained success within this century.
The first step is to select a suitable study group (see also Riedel et al. 2013 for a process chart). After an initial taxon screening, we have selected the hyperdiverse weevil genus Trigonopterus Fauvel for our research on biodiversity patterns and biogeography across the Indomalayan Archipelago and Melanesia. Trigonopterus are flightless weevils placed in the subfamily Cryptorhynchinae of Curculionidae (Alonso-Zarazaga and Lyal 1999). It contains 91 described species ranging from Sumatra to Samoa, and from the Philippines to New Caledonia. To date, 50 species of Trigonopterus have formally been described from New Guinea, the center of its diversity. The majority of these species were described from the Papuan peninsula (Faust 1898(Faust , 1899 and from the Sattelberg area of the Huon peninsula (Voss 1960), both in present day Papua New Guinea. We have previously established that Trigonopterus are suitable for accelerated taxonomic study combining morphology and the DNA barcoding approach using mitochondrial cox1 data (Riedel et al. 2010;Tänzler et al. 2012). Trigonopterus species were clearly delineated by both molecular data (nuclear as well as mitochondrial sequences) and morphology, and both data sets were fully compatible. These preliminary surveys already resulted in the recognition of 279 Trigonopterus species from seven localities across New Guinea. Most of these were undescribed. DNA barcoding is recommended as an identification tool for Trigonopterus since the sequence data in a dynamic identification engine represent an efficient substitute for a traditional species-level key. Considering the high proportion of unknown and usually morphologically similar species both traditional dichotomous keys and computer-based interactive keys would be of very limited use. In the following we provide short diagnostic descriptions with photographs of habitus and male genitalia. In keeping these descriptions concise, it is possible to increase the number of species treated dramatically. This study demonstrates that the taxonomy of hyperdiverse groups can be tackled with the combination of DNA-barcoding and taxonomic expertise. Such work does neither proceed at lightning speed, nor can it be fully automated. However, fully embracing technological development, work can be sped up and results are more sustainable. Significant workloads can be trusted to technicians and students, while the taxonomist can focus on the actual comparative taxonomic work.
Some of the historic Trigonopterus species from New Guinea were revised by Riedel (2011). Types of all relevant Papuan species have been examined and additional revisions of the previously described species are in preparation. In the following, we concentrate on species which are not closely related to the ones already known to science. A selection of 101 species covering the morphological diversity of Papuan Trigonopterus is described below providing a scaffold for ongoing, future work on this genus.

Materials and methods
This study is based on a selection of 101 out of 279 species recognized by . The number of 101 species was chosen as large enough to cover a major portion of diversity and small enough to complete the task within the scheduled time frame in 2012. Species represented only by females were not included in this selection. Care was taken that all major groups are represented, based on our unpublished phylogenetic analysis. Moreover, we describe some clades of closely related species to demonstrate that our technique also works well for these. Four cryptic species (T. granum sp. n., T. imitatus sp. n., T. pseudogranum sp. n., and T. velaris sp. n.) are here described; they differ only in minor morphological characters but exhibit a marked genetic divergence (9.9-13.9 % uncorrected p-distance in our cox1 fragment). In all, 50 species of Trigonopterus were previously known from the Papuan region, and three of these species could be identified with confidence in our full dataset. Many of the other previously described species have type series of mixed species which requires additional taxonomic work. This will be done in the near future after the completion of ongoing field campaigns might reveal fresh specimens for study. Species resembling the historic described species were excluded to avoid the risk of creating synonyms. Therefore, species from the Papuan and the Huon peninsula are somewhat underrepresented here.
Holotypes were selected from the sequenced specimens of Tänzler et al. (2012); their DNA had been extracted nondestructively as described by Riedel et al. (2010) and in our laboratory wiki (http://zsm-entomology.de/wiki/The_Beetle_D_N_A_Lab). The genitalia of most specimens did not require maceration after DNA-extraction; they could be directly stained with an alcoholic Chlorazol Black solution and stored in glycerol in microvials attached to the pin of the specimens. Genitalia of collection specimens or specimens whose abdominal muscle tissue was not sufficiently digested after DNA extraction were macerated with 10% KOH and rinsed in diluted acetic acid before staining. Illustrations of habitus and genitalia were prepared from holotypes. Finally, type series were supplemented with specimens stored in ethanol and older material from the dry collection. As always the case in paratypes, there is a chance that some of these are incorrectly assigned; this is especially true for specimens without sequence-data as an identification based on external morphological characters is more prone to error than an identification based on a cox1 sequence . Altogether, the selection of 101 species herein is represented by 4,624 specimens. Type depositories are cited using the following codens: Morphological descriptions are limited to major diagnostic characters. For example, the aedeagus often bears characters suitable to separate closely related species and is therefore illustrated and briefly described. Tegmen and sternite VIII of males show peculiar characters in some species, but these are usually not species-specific. Therefore, they are omitted from the diagnostic descriptions. Measurements such as length / width ratio of elytra or pronotum are avoided and can be taken from the photographs if needed. Identification of females is difficult and is best done based on cox1-sequences. Illustrations of female genitalia would alleviate this situation only marginally and the time required to prepare the relevant illustrations did not appear justified. Negative character states (i.e. the absence of a character) are only mentioned explicitly where it appears appropriate. For example, there are few species with swollen or denticulate epistome. In these cases the character state is described, but for the majority of species with simple epistome it is not mentioned. Common practice would require to state explicitly "epistome simple". Although formally accurate, in groups comprising hundreds of species this leads to inflated descriptions that distract the reader from the important information by enumerating the absence of rare character states. Except in the case of cryptic species no mention is made of "closely related species", as their choice is highly subjective. The data provided by the cox1-sequences should be sufficient at the moment. At a later stage a phylogeny will be published based on several markers and then suitable subgroups may be formally named as subgenera.
Etymology. This epithet is based on the Latin adjective angustus (narrow) and refers to its habitus.
Etymology. This epithet is based on the Latin adjective apicalis (pertaining to the apex) and refers to the species´ contrasting elytral coloration.

Biology. Collected by beating foliage in montane forests.
Etymology. This epithet is based on the Latin participle armatus (armed) and refers to the teeth of the male meso-and metatibia.
Distribution. Jayawijaya Reg. (Poga). Elevation: ca. 2620-2715 m. Biology. Beaten from foliage of upper montane forests. Etymology. This epithet is based on the Latin participle ascendens (climbing up) and refers to its occurrence on higher elevations.
Distribution. Jayapura Reg. (Cyclops Mts). Elevation: 300-1200 m. Biology. Collected by beating foliage in montane forest. Etymology. This epithet is based on the Latin noun augur in apposition and refers to the large eyes that help the species to see birds, presumably important predators.
Distribution. Jayawijaya Reg. (Bokondini). Elevation: ca. 1705-1710 m. Biology. Beaten from foliage of montane forest. Etymology. This epithet is based on the Latin adjective conformis (like, similar) and refers to the similarity of this species, both to some closely related sibling species, and to others of only superficial resemblance.
Etymology. This epithet is based on a combination of the Latin nouns cuneus (wedge) and penna (wing, elytron) and refers to the shape of elytra.
Etymology. This epithet is based on the name of the dromedary camel (Camelus dromedaries L.) and refers to the body shape.
Etymology. This epithet is the latinized form of the Greek noun eremites (hermit) and refers to the species´ restricted occurrence in the montane forests of the Cyclops Mountains.

Trigonopterus glaber
Notes. Trigonopterus glaber Riedel, sp. n. was coded as "Trigonopterus sp. 164" by Tänzler et al. (2012). Etymology. This epithet is based on a combination of the Greek nouns gonatos (knee) and ceros (horn) in apposition and refers to the peculiar extensions of the tibial base.
Notes. Trigonopterus granum Riedel, sp. n. was coded as "Trigonopterus sp. 15" by Riedel et al. (2010) and Tänzler et al. (2012), respectively "Trigonopterus spo" in the EMBL/GenBank/DDBJ databases. It is closely related to T. pseudogranum sp. n., T. velaris sp. n., and T. imitatus sp. n.; from the latter two it can be distinguished by its sparsely punctate body and the structure of its male abdominal ventrite 5. The externally very similar T. pseudogranum sp. n. is best separated by the cox1-sequence which diverges 12.1 %.

Trigonopterus imitatus
Notes. Trigonopterus imitatus Riedel, sp. n. was coded as "Trigonopterus sp. 273" by Tänzler et al. (2012). It is closely related to T. granum sp. n., T. pseudogranum sp. n., and T. velaris sp. n. from which it can be distinguished by the male venter with long setae. Despite its close morphological similarity its cox1-sequence diverges 9.9-12.3 % from the other species.

Etymology. This epithet is a combination of the name Ixodidae and the Latin suffix
Etymology. This species is dedicated to the people of Papua New Guinea. The epithet is based on the family name Kanawi, found on page 236 of the Papua New Guinea Telephone Directory of 2010 and treated in genitive plural.
Etymology. This species is named in honour of our colleague Katayo Sagata (Goroka) who greatly supported our field-work in PNG and who collected some of the specimens.

Biology. Beaten from foliage of montane forests.
Etymology. This epithet is based on a combination of the Latin noun macula (mark, spot) and refers to the conspicuous orange spot on the elytra.
Etymology. This epithet is based on the Latin adjective mimicus (acting, imitating) and refers to the resemblance to other species with ferruginous elytra.
Etymology. This epithet is based on a combination of the Latin noun mons (mountain) and the participle vagus (wandering) and refers to the relatively wide distribution in the highlands of Papua New Guinea.

Biology. Collected by beating foliage in primary forests.
Etymology. This epithet is based on the Greek prefix pseudo (false) and the name of Trigonopterus nasutus (Pascoe) which is superficially very similar and occurs sympatrically.
Distribution. Eastern Highlands Prov. (Mt. Michael). Elevation: ca. 2179-2800 m. Biology. Sifted from leaf litter in primary forest. Etymology. This epithet is a combination of the genus name Ptolycus and the Latin suffix -oides (having the form of) and refers to the species´ resemblance in habitus.
Etymology. This epithet is based on the Latin participle punctulatus (provided with little punctures) and refers to the species´ surface scattered with small punctures.
Etymology. This epithet is based on the combination of the Latin adjective rufus (reddish) and the noun basis (base) and refers to the elytral coloration.
Material examined. Holotype (SMNK): ARC1858 (EMBL # HE616135), PAP-UA NEW GUINEA, Eastern Highlands Prov., Aiyura, S06°21.033', E145°54.597', Biology. Sifted from leaf litter in montane forest. Etymology. This epithet is based on the Latin adjective sordidus (dirty) and refers both to the occurrence of incrustations and the species´ general appearance making it hard to distinguish from a grain of dirt.
Etymology. This epithet is based on the Latin participle striatus (provided with furrows) and refers to the species´ body-sculpture.
Etymology. This epithet is based on a combination of the Latin prefix sub-(less than; almost) and the participle glabratus (smoothened). It refers to its smooth body surface.
Etymology. This epithet is based on the Latin participle sulcatus (furrowed, grooved) and refers to the elytral sculpture.
Etymology. This species is named in honor of Rene Tänzler (Munich), who has spent years working on Trigonopterus weevils.
Etymology. This epithet is composed of the Latin prefix tri-(three) and the participle dentatus (toothed) and refers to the three apical teeth of the male rostrum.

Discussion
For the last few decades it has been common practice for taxonomists maintaining a good reputation to revise monophyletic groups with all the known species, to prepare elaborate descriptions bearing in mind the potential value of every tiny character for a phylogenetic analysis, and to make great efforts on the preparation of elaborate identification keys based on these characters. As a consequence, hyperdiverse taxa such as the genus Trigonopterus are usually avoided, simply because it appears impossible to get this task completed during a lifetime. Such considerations are a lesser issue for a minority of taxonomists with a lower quality-standard, the so called "mass-describers", usually publishing their works in journals without peer-review. Often, they do not refrain from proposing new names based on specimens without sufficient diagnostic characters, such as unique females, relying on the community to later sort out the resulting identification problems. This exacerbates the deterrence of such "difficult" taxa which are then prime examples and causes of a "taxonomic impediment", in this case not only an impediment to the end-users of taxonomy, but also to the taxonomists themselves (Ebach et al. 2011, Godfray 2002. In fact, at this stage taxonomic information becomes rather a burden to science than a useful tool. We believe that technologies developed within the past decade enable us to stop this vicious circle. Two components are of fundamental importance, i.e. online wiki databases and molecular systematics (Riedel et al. 2013).
Online wiki databases such as the Species-Id portal (http://species-id.net/wiki/ Main_Page) are not recognized as means of publication by the International Code of Zoological Nomenclature (1999), so their significance needs some explanation. Journals such as "ZooKeys" make a new name available with a traditional paper publication, simultaneously creating a wiki with the same content. This wiki can be updated later anytime with additional data, be it an elaborate 3D-model or a "quantum contribution" (Maddison et al. 2012) such as fixing a typo of the original description or adding a simple collecting record. At the time the species becomes formally named there is no urgency to provide the description with all possible data. It should contain a reasonable basis, so that its diagnosis is guaranteed. We expect that most users will later rather consult the online working description, gradually being supplemented with additional data. Thus, the formal species description is like a healthy newborn which is expected to grow into an adult with the help of its environment. In the case of Trigonopterus, characters such as the functional morphology of thanatosis or the morphology of the metendosternite, surely of great interest but of little diagnostic value can be added at a later stage without compromising their visibility. This approach does in fact bundle useful features of numerous other initiatives such as the Encyclopedia of Life (eol.org).
The impact of molecular systematics on species-descriptions is twofold and can be divided into reconstruction of species relationships and attempts to diagnose species. But let us start one step earlier, with the advent of phylogenetic systematics (Hennig 1966) and phenetics (Sneath and Sokal 1973). Both had a profound but little-noticed effect on the preparation of species descriptions. Since more and more taxonomic revisions incorporated phylogenetic analyses, it was attempted to maximize the number of informative characters. Thus, even characters of little value for species diagnosis were included in the descriptions. Another consequence was that species descriptions within a study were sought to be standardized, best illustrated by the program Delta (Partridge et al. 1993). Negative character states (i.e. the absence of a character) were often explicitly stated. Often enough, all this time-consuming procedure did not increase the usability of descriptions for the purpose of diagnosis, but rather inflated them. And after all, standardization among different authors was never achieved not to mention failure to introduce an urgently needed minimum standard.
Although in some taxa, phylogenies based on morphological data are still needed, in recent years the trend clearly goes towards purely molecular phylogenies. In the case of Trigonopterus we feel that our molecular data set is strong enough to produce a stable phylogeny without morphological characters included. So, for us it is time to ask -do we really need to describe every character with the hope that this might be a valuable addition to our character-matrix? For us, the answer is "no"! With changing needs on a species description, taxonomists should reflect if they want to carry on like during the past decades, or if it is time to adjust procedures and streamline descriptions to the purpose of diagnosis.
The potential of using a standard DNA marker for species identification, also known as "DNA barcoding", was recognized almost ten years ago (Hebert et al. 2003). Despite some initial criticism it proved to be a powerful tool. In many taxa, the cox1 sequence will pinpoint the correct species without additional information. In others it may not delineate species unambiguously (Hendrich et al. 2010), but even then it is possible to safely pinpoint a group of e.g. 5-10 species. For many taxa, be it nematodes, moss mites or rove beetles, a non-expert would hardly achieve this within reasonable time using traditional keys. After all, in combination with a few morphological characters the species can be safely identified in most cases. One of the great advantages of sequence data is that they can be databased, searched and accessed anytime from anywhere. The situation with type specimens is quite different: often they are not accessible, or if so, it is very time-consuming both for museum curators and active researchers to send them around the globe. Often enough, they give the only clue what species an insufficient description is referring to, or if the species is placed in the correct genus at all. Such issues could be much faster solved using "DNA barcodes". We strongly believe that the ICZN should make the publication of genetic data obligatory for the description of new extant species. Until such a decision is made, the contest between descriptions containing DNA barcodes and the ones without may give an answer of what is really needed. The combination of short expert morphological descriptions provided with a few high-resolution photographs and DNA-sequences appears to us as the way to proceed. We predict that similar works are to be expected in the near future on various taxa of hyperdiverse organisms.