An unusual and well-preserved fossil staphylinid is described and figured from a single specimen in Upper Cretaceous Burmese amber. Gollandia planata gen. et sp. n. is tentatively placed in the extant oxyteline tribe Coprophilini, although it lacks a few characteristic features of present-day members of the group, likely indicating it to be either a stem group of the tribe or prove to be distinct pending future discoveries. The discovery of this genus suggests that early oxytelines were more morphologically diverse during the Cretaceous and their evolutionary history was more complicated than previously documented. Tribal placement as regards fossil oxyteline taxa is discussed.

Burmese amber, Cenomanian, Mesozoic, Myanmar, new genus, new species

The staphylinid subfamily Oxytelinae Fleming, 1821 is a relatively large group with over 2049 valid extant species placed in 40 genera (Thayer 2016). Oxytelines are distributed worldwide and have proven to be remarkably diverse. This diversity is so far not fully known as in several geographical areas the species are incompletely known, in many cases a vast portion (up to 80%) still await formal description. The generic revision of Herman (1970), which was also a first attempt at a phylogenetic analysis, set the foundation for a better understanding of relationships within the subfamily and for all subsequent revisionary and evolutionary research on oxytelines.

Despite Herman’s (1970) monumental study, the classification of tribes within Oxytelinae remains unsettled. The most primitive oxytelines (formerly as tribe Deleasterini, sensu Makranczy 2006, based on abdominal segments with only one pair of laterosclerites) are now often split into three separate tribes: Deleasterini Reitter, 1909, Euphaniini Reitter, 1909, and Syntomiini Böving & Craighead, 1931. The higher oxytelines, comprising 98% of the described valid species in the subfamily, are variously classified in the more primitive Coprophilini Heer, 1839, as well as the now widely accepted Blediini Ádám, 2001 and Planeustomini Jacquelin du Val, 1857. The more derived clades seem to belong into a once again unified Oxytelini (sensu Makranczy 2006), although this remains debated and awaits support from molecular studies.

The tribe Coprophilini is currently without identified synapomorphies, and is instead defined by a lack of features of the more derived lineages (i.e., it is presently circumscribed by putative plesiomorphies, may be paraphyletic, and is in need of considerable revisionary and phylogenetic exploration). The extant coprophilines are characterised by the following combination of traits: mesocoxae narrowly separated by mesosternal process or contiguous, tarsal formula 5-5-5, abdominal segments with two pairs of laterosclerites, and with only six sternites visible. According to Herman (2001), the tribe contains three extant genera, Coprophilus Latreille, 1829, Coprostygnus Sharp, 1886, and Homalotrichus Solier, 1849, although a poorly described genus, Coprotrichus Hayashi, 2005 (based on a single species from Japan) was later added and is presently considered as valid but requires detailed study. The number of species in these genera is not great; Coprophilus stands in the last catalogue with 30 valid species (Herman 2001), Homalotrichus and Coprostygnus are being revised by the first author, standing with at least 13 and five species, respectively (Makranczy, unpubl. data). Coprophilus is widespread in the Northern Temperate zone, Homalotrichus is known from Australia (including Tasmania) and South America, while Coprostygnus is confined to New Zealand. It is remarkable that all of these species prefer cold climates, often occurring at the highest elevation where one can find oxytelines, up to 4000–4200 m.

The various records of fossil Oxytelinae and its related subfamilies were summarized by Cai et al. (2017). Hitherto, the only definitive fossil Coprophilini described is a rather poorly preserved compression fossil (without counterpart), Mesocoprophilus clavatus Cai & Huang, 2013. A further fossil genus, Sinoxytelus Yue, Zhao, & Ren, 2010 was subsequently transferred to Coprophilini, but this placement is tentative as the abdomen has basolateral ridges, a trait of Oxytelini. The genus originally contained three extinct species (Sinoxytelus euglypheus Yue, Zhao, & Ren, 2010, S. breviventer Yue, Zhao, & Ren, 2010, and S. longisetosus Yue, Zhao, & Ren, 2010) from the Yixian Formation (Lower Cretaceous, ca. 126 Mya, 41°36’44”N, 120°49’48”E), Liaoning, China (Yue et al. 2010), and a fourth taxon, S. transbaicalicus Cai, Yan, & Vasilenko, 2013 was later described from the Urey beds (Lower Cretaceous, 50°38.730’N, 112°50.338’E), Transbaikalia, Russia (Cai et al. 2013). Here we describe a new genus and species from the Upper Cretaceous amber of northern Myanmar.

This new taxon represents the second fossil occurrence of Oxytelinae documented from Mesozoic amber, and is also the oldest amber inclusion presently recorded for the subfamily along with Prajna tianmiaoae Lü et al., 2017, a species of Thinobiini (a more derived tribe) described from the same deposit.

Specimen photography was done with the amber piece mounted on a small plastic plate surface with a drop of viscous glycerine covered with a glass cover slip to eliminate distortions from the otherwise rounded amber surface. The habitus photographs were made with a Canon 5D Mark III camera and C Canon MP-E 65mm f/2.8 1–5× macro lens with two Canon Macro Twin Lite MT-24EX flash units and a Canon Speedlite 430EX III-RT flash unit standing on the shoe foot directly in front of the specimen (shooting directly into the head of the specimen). The light was diffused by a single ring of mylar. Raw photograph files were imported to Adobe Lightroom 5.7.1, and layers stacked with ZereneStacker (Richland, Washington, USA). Details of the specimen were photographed with a Canon EOS 6D camera attached to a Leica M205 C stereomicroscope with the help of a Canon EOS Utility 3.4.30.0 software, before being stacked using the same software as previously mentioned. Abbreviations for measurements are defined as follows:

HW head width with compound eyes;

TW head width at temples;

PW maximum width of pronotum;

SW approximate width of shoulders;

MW maximum width of elytra;

AW maximum width of abdomen;

HL head length at the middle-line from front margin of clypeus to the beginning of neck;

EL compound eye length;

TL length of temple;

PL length of pronotum at the middle-line;

SL length of elytra from shoulder;

SC length of elytra from hind apex of mesoscutellum;

FB fore-body length (combined length of head, pronotum, and elytra);

BL approximate body length.

The specimen is exceptionally well preserved but in a rather unfortunate position within the amber piece. Under UV-light examination it can be seen that the specimen is sitting within a dip in the internal flow of the amber (the amber flowed in layers when it was originally exuded from the tree and the beetle is in a dip within these flows). The result of its placement within the dip in the flow means ideal, clear images cannot be produced at high magnification unless some of the flows can be polished away. Whoever made the original preparation (probably local workers in Myanmar) polished the specimen too close to the amber surface and cut the piece poorly. The result is that it is now impossible to cut and polish the piece closer to minimize the optical impact of the flow lines.

The recent commercial amber mines are located in the Hukawng Valley (26°16.5’N, 96°34.0’E), Kachin, northern Myanmar. The minimum age of the amber is estimated to be 98.79 ± 0.62 Mya (by radioisotope dating of zircon crystals obtained from the volcanoclastic matrix, Shi et al. 2012), and so just slightly into the base of the Cenomanian. Cruickshank and Ko (2003), Ross et al. (2010), and Grimaldi and Ross (2017) provide a geological account of the deposits, and these authors also note that many of the amber pieces appear to have at some point eroded from sediments and been redeposited, at least suggesting the possibility that the inclusions could be slightly older, perhaps from the latest Albian. Recent studies speculate that the West Burma Plate/Block (west of the Sagaing fault line in Myanmar) was of Gondwanan origin (Poinar 2018), and the resin-producing tree was hypothesized to be of Agathis Salisb. (Araucariaceae) (Poinar et al. 2007). The paleoclimate of Burmese amber producing forests were suggested to be tropical with an average temperature range of 32–55 °C (Grimaldi et al. 2002; Ross et al. 2010).

There can be a great debate on how to place fossil species in higher taxa when the classification is otherwise based entirely on present-day species. Naturally, fossil species placed cladistically within the crown group are not difficult, but when potential stem groups are discovered some difficulties and strong differences of opinion arise. Does one include them within the formal group, necessitating a new circumscription of the taxon boundaries in order to accommodate the fossil, or, at another extreme, more radically establish a new group for the fossil as a potential sister to its extant relatives? The latter will inevitably lead to a proliferation of meaningless and often monobasic groups based on limited characters (e.g., often lacking critical data on genital traits) and collectively forming a pectinate stem to any lineage circumscribed solely on the basis of its extant constituents, the end result becoming a classification of little explanatory power. Ideally, there one desires a balance between maintenance of a good diagnostic power and keeping the classification simple while simultaneously reflecting the hierarchical relationships supported among the various taxa being classified and granting the system maximal explanatory and predictive significance. These goals are sometimes simultaneously achieved, but more often than not the former objective becomes muddied while attempting to adequately reflect the latter. This challenge is particularly great for ancient groups, such as the Oxytelinae whose history goes back at least 150 Mya, where there has been inevitably considerable extinction over the intervening epochs between the first divergence of a given clade and its modern fauna. Cretaceous deposits will undoubtedly supply for generations to come a continuous stream of unusual fossil species that will force us to rethink our estimates of relationship and concomitantly the classifications from which they are built. Descriptive science (Grimaldi and Engel 2007) has undergone an unprecedented boom in the past few years with new techniques (tomography, confocal microscopy) providing details that bring the examination of fossil morphology more comparable to that of extant species. The result of this increase in available character data and extinct taxa will be finer phylogenetic placement of peculiar fossil species and, hopefully, greater clarity on how best to tackle each classificatory alteration as they arise.

In the case of the presently known fossil oxytelines or putative oxytelines this challenge is acute as there are limited character data available. The four fossil species presently placed in Sinoxytelus possess a mixture of ancient and relatively modern traits. The basolateral ridges on the abdominal tergites are a trait of the tribe Oxytelini, while the somewhat reduced second abdominal sternite suggests placement in more primitive tribes. A solid age estimate is available for part (Lujiatun Bed) of the Yixian Formation and at approximately 126 Mya, or Aptian (Chang et al. 2017). If Sinoxytelus is truly either a stem group to Oxytelini or Oxytelini + other higher oxytelines, then the subfamily Oxytelinae as whole was apparently already diverse by the latter part of the Early Cretaceous. It seems apparent that there has been considerable extinction within the subfamily, resulting in only a few lineages surviving and diversifying to their present state. These fossils may be rather distantly related to the ancestors of the present Coprophilini or Oxytelini, with potentially numerous additional extinct stem groups (yet to be discovered) present before the divergence of the crown groups of these latter tribes. Characters that Cai et al. (2013) cite for placing Sinoxytelus within Coprophilini (e.g., metasternal process slightly protruding but not meeting mesosternal process) are more indicative of particular genera rather than the tribe, as there is considerable variation within the present day genera for these traits and are not strictly diagnostic for coprophilines (Herman 1970).

The current Burmese amber fossil species has a similarly conflicting combination of traits, on the one hand it can be placed in Coprophilini because of the plesiomorphic condition of the basal abdominal segments (although the tribe is presently defined on putative plesiomorphies and so this feature alone indicates nothing more than the potential for the fossil to belong therein or to belong to stem-group coprophilines, assuming the tribe is truly monophyletic). On the other hand, the apomorphically reduced number of tarsomeres, is a derived feature, but currently cannot be considered anything more than autapomorphic for the genus. Thus, given the lack of abundant evidence definitively demonstrating its phylogenetic placement relative to modern oxyteline genera, we prefer to tentatively consider the genus as a member (a stem group member) of Coprophilini. Hopefully, in time further material and other fossil species will be discovered that will ultimately permit a clarification of its relationships along with its broader implications toward affinities among lineages of Oxytelinae, and at which time a redefinition of oxyteline tribes may be attempted.

We are grateful to Al Newton (Field Museum of Natural History, Chicago) for providing an initial set of notes and comments on the specimen, and to Aslak Kappel Hansen and Janina Lisa Kypke (University of Copenhagen, Denmark) who aided the first two authors with taking photographs and experience with photographing amber specimens. A Visiting Scholarship from the Field Museum supported the first author (GyM) (12 March–5 April 2018) for portions of this research. The second author (SY) was supported by a Japan Society for the Promotion of Science (Tokyo, Japan) Overseas Research Fellowship (29-212).