Rhode Island ’ s precinctive beetles : Taxonomic lessons and the utility of distributional checklists

Th e sixty-two species of precinctive (reported from the state and from nowhere else) beetles known from Rhode Island, and originally described by Th omas L. Casey, are reviewed. Th ese 62 are reduced to 12 potential candidates, with a further 12 unrevised species awaiting investigation. In terms of the Rhode Island Coleoptera fauna, the present re-evaluation decreases the number of beetles known from the state by 12 to 2,243 species. Th is information is briefl y presented in the context of T.L. Casey’s contribution to our knowledge of the North American (and specifi cally Rhode Island) beetle fauna and the strengths and weaknesses of his taxonomic approach. Th e utility of distributional checklists, such as the one of the Rhode Island beetle fauna, are discussed. Th e resolution of taxonomic problems is of central importance to many spheres of biological investigation and accurate distributional checklists are vital in this process. Such checklists are useful in the context of determining biodiversity and environmental monitoring and for many practical purposes such as assisting in identifi cation and highlighting gaps in distribution. Furthermore, data from such checklists also have utility in a variety of zoogeographic investigations, such as contributing to an understanding of latitudinal gradients in species diversity and the geographic basis for proportionate faunal composition. All these are compelling arguments for developing and maintaining such resources.


Introduction
Th omas Lincoln Casey (1857Casey ( -1925) ) was an army engineer by profession.Commencing in 1884, and particularly after his retirement from the army in 1912, he devoted much of his time to the study of Coleoptera.He described over 9,200 species, subspecies, and varieties of beetles.Th e son of a general, a West Point graduate, a brilliant student, a successful theoretical and applied astronomer, and an avid conchologist and coleopterist, Casey undoubtedly made an enormous contribution to the study of Coleoptera in North America, particularly of the Staphylinidae.When he passed away in Washington, DC on 3 February 1925, he was buried along with his microscope.Intensely observant of detail, his keen intellect and powers of observation served him well in many respects, but his over-attention to small details and minor variations lead to signifi cant shortcomings that marred some of his work, and have bedeviled the work of taxonomists who have followed him.In the fi rst volume of Th e Coleopterists Bulletin, Ross H. Arnett, Jr. (1947: 67) wrote, "Casey was perhaps the most ardent student of beetles this country has yet produced.Also he was and is the subject of the greatest amount of criticism."Lindroth (1975) wrote that, "Casey's concept for this unit (species) was apparently considerably diff erent, not only from that of our time, but also from that of his contemporaries and most older authors."Of the 902 "types" of Carabidae studied by Lindroth, he found only 81 that he regarded as valid species.Seevers (1951) is another author who studied Casey's collections extensively in the course of his revision of Gyrophaena and related genera (Casey described 47 new species in this group of genera within the Aleocharinae).Seevers (1951: 664) synonymized 20 of these and wrote that, "Casey's studies on the Gyrophaenae suff er from several weaknesses that characterize much of his work.One of the chief criticisms of his taxonomy is that he failed to take into account intraspecifi c variation.… He proposed many species for which no males were available, a practice that greatly complicated the taxonomy of the group.… Casey's practice of including the descriptions in a key to the species did not prove to be satisfactory; diagnostic characters were often obscured by non-essential details. … Casey's papers give no indication that he attempted to relate the American species … to their European allies." Rhode Island was a particular benefi ciary of Th omas Casey's attention in that his family farm was located in Saunderstown (currently owned and operated by the Society for the Preservation of New England Antiquities); he collected extensively in the state, and described many beetles from there.As such, as Sikes (1998Sikes ( , 2004) ) has pointed out, part of this legacy are a large number of apparently precinctive (native to the area specifi ed and occurring nowhere else) Rhode Island beetles.Sikes (2004) listed 62 species known from Rhode Island and not reported elsewhere.All of these were species described by Th omas Casey.As Sikes (1998Sikes ( , 2004) ) pointed out, there is no real reason to suppose that these species are actually endemic (i.e., restricted) to Rhode Island since adjacent states share the same environments, and there are no obvious natural barriers to dispersal that would isolate populations in Rhode Island.Nonetheless, until the taxonomic status of these species is resolved, the possibility of endemism cannot be excluded.Block Island in the state, continues to support a population of Nicrophorus americanus Olivier, a burying beetle that has otherwise disappeared from all of its eastern range, and Bibloplectus sobrinus (Casey), is a pselaphine rove beetle that is known from only Connecticut and Rhode Island (Sikes 1998).
Bearing this in mind we decided to further investigate this potential suite of precinctive Rhode Island beetles.

Results
Upon further examination the 62 species reported by Sikes (2004) as known only from Rhode Island can now be placed in the following categories: 1) Synonymized species: these are species names that have been synonymized by authors who have reviewed the respective taxa (Table 1).In all cases these have proved to be species known to occur in jurisdictions beyond Rhode Island.2) Valid species not confi ned to Rhode Island: these are species that have been determined to be valid by authors who have reviewed the respective taxa, but that represent species known to occur beyond Rhode Island (Table 1).3) Valid species known only from Rhode Island: these are species that have (to varying degrees) been treated as valid in recent taxonomic publications, and that have not been reported from outside of Rhode Island (Table 2).Th ey could potentially qualify as precinctive Rhode Island beetles.In most cases, however, they belong to relatively poorly known and largely unrevised genera.In some instances the type specimens have not been critically, comparatively examined to ascertain if they are valid and distinct, and/or collections of these groups at other institutions have not been examined with a view of ascertaining if these species have a wider distribution.4) Species in unrevised groups: these are species names of unrevised North American taxa that have received little or no critical attention or examination since their description by Casey (Table 2).Th ey could represent any of the above three categories, but, pending revisionary studies, little can be said with respect to their status.Specifi c notes with respect to a few of these species:

Olisthopus fi licornis Casey, 1913
Th is species is treated as valid by Bousquet and Larochelle (1993) and has only been recorded from Rhode Island.Th ere is, however, but a single specimen, a female, making it diffi cult to determine (without a comprehensive review of the genus) if it is actually diff erent from the widely distributed Olithopus micans LeConte, 1848 (Y.Bousquet, pers. comm.).

Olisthopus innuens Casey, 1913
Th is species is treated as valid by Bousquet and Larochelle (1993) and has only been recorded from Rhode Island.Th ere is, however, but a single specimen, a tenereal male, making it diffi cult to determine (without a comprehensive review of the genus) if it is a distinct species.Y. Bousquet (pers. comm.)feels that the specimen may be conspecifi c with the widely distributed Olisthopus parmatus (Say, 1823).Th ere are no consistent external structural diff erences between the holotype of O. innuens and O. parmatus except that the specimen of O. innuens is smaller.

Ptinella ochracea (Casey, 1924)
Casey described this species in the genus Ptiliodina Casey, 1924 which is listed by Hall (2001) as a junior synonym of Ptinella Motschulsky, 1872.Poole and Gentilli (1996) treated it in Ptilodina, Downie and Arnett (1996) included it in Pteryx Matthews, 1859, while Sikes (2004) placed it in Ptinella.Since this genus is completely unrevised in North America, the generic assignment of this species is subject to considerable uncertainty.

Staphylinidae, Aleocharinae
Acrotona avia (Casey, 1910) Th is species was synonymized with Acrotona subpygmaea Bernhauer, 1909 by Gusarov (2003c), however, the two type specimens of A. subpygmaea have the median lobe of the aedeagus missing in the male and a collapsed spermatheca in the female, complicating taxonomic evaluation.Majka et al. (2008), however, revalidated the name A. avia and reported the species from Nova Scotia.

Atheta (Dimetrota) limulina Casey, 1911
Th is is a valid species, however, it also now includes a junior synonym, Atheta rusticula Casey, 1911, described by Casey from specimens collected in New York (Gusarov 2003c).
Atheta dadopora Th omson, 1867 Gusarov (2003b) treated this species as an introduced Palaearctic one, however, Majka andKlimaszewski (2008a, 2008b) regarded it as probably being a widely distributed Holarctic species.Casey, 1906Seevers (1951) revised Gyrophaena in North America and wrote that, "G.rhodeana is probably a valid species of the illiana group" (pp.686).It has only been recorded from Rhode Island.

Stethusa dichroa (Gravenhorst, 1802)
Th is widely distributed species is found in the United States and south through the West Indies to Paraguay and the Galapagos Islands of South America (Gusarov 2003a).

Carpelimus egregius (Casey, 1889)
Th is species is treated by Herman (2001Herman ( , 1655) ) as a valid species, known only from its type locality of Newport, Rhode Island.Th e genus Carpelimus, however, is badly in need of revision in North America (Newton et al. 2001) and C. egregius was listed as a valid name only because no synonymy for it has been published (L.Herman, pers. comm).

Carpelimus facetus (Casey, 1889)
Th is species is treated by Herman (2001Herman ( , 1655) ) as a valid species, apparently known only from its type locality of Newport, Rhode Island.Th e genus Carpelimus, however, is badly in need of revision in North America (Newton et al. 2001) and C. facetus was listed as valid name only because no synonymy for it has been published (L.Herman, pers. comm.).In addition to Rhode Island, Moore and Legner (1975) list "Lake Superior?" as a possible locality for this species.Th erefore, even if valid, it may not be restricted to Rhode Island.

Staphylinidae: Paederinae
Astenus spectrum (Casey, 1905) Th is is an apparently valid species that has now been recorded from Florida (Frank 1986), New Hampshire (Chandler 2001) (Newton et al. 2001).Notwithstanding, there are specimens of this species from Oklahoma [Latimer Co., VII.1987, VI.1988, VII.1988, K. Stephan, 4 specimens] in the Snow Entomological Museum indicating that this species is not restricted to Rhode Island.

Discussion
Th e Rhode Island beetle fauna.Of the 62 species reported by Sikes (2004) as known only from Rhode Island: 1) 25 nominal species are now known to be synonyms of valid 15 species, none of which are restricted to Rhode Island; 2) 11 nominal species are now known to be valid species, none of which are restricted to Rhode Island; 3) 14 nominal species, representing 12 species, have (to varying degrees) been treated as taxonomically valid, although the taxonomic status of most of these is as yet far from clear; 4) 12 species belong to genera that are wholly or largely unrevised.Pending revisionary work the status of these cannot be determined.
While not having been synonymized, the taxonomic status of most of the 12 species in the third group -the remaining possible Rhode Island "precinctive" species -is as yet far from clear.Two species (Olisthopus fi licornis and O. innuens) are represented by single individuals, and while the genus Olisthopus has not been recently revised, preliminary observations indicate that both of these may be synonymous with other widely distributed species.Two others (Carpelimus egregius and C. facetus) have not been subject to taxonomic examination per se and have only been treated as valid species because no synonymies have been published.One species (Gyrophaena rhodeana) has been listed as "probably" a valid species.Th e other aleocharines (Atheta longiclava, A. nugator, A. temporalis, Dilacra lineata, Philhygra fi nitima, and Trichiusa pilosa) are members of unrevised genera and are presently taxonomically valid largely by reason of inattention.It is not improbable that in the future, as taxonomic attention is devoted to these species and groups, these species may also fall to synonymy or, if found to be taxonomically valid, may be shown to be distributed beyond the confi nes of Rhode Island.
Similarly the 12 species in the fourth group, which are completely unrevised and are members of very poorly known groups, may with further investigation also fall to synonymy, or may be shown to be distributed beyond the confi nes of Rhode Island.In summary the 62 Rhode Island precinctive Coleoptera treated by Sikes (2004) have now been reduced to 12 potential candidates, with a further 12 unrevised species awaiting investigation.In terms of the Rhode Island Coleoptera fauna, the present reevaluation decreases the number of beetles known from the state by 12, reducing it to 2,243 species (Sikes 2004;Sikes and Webster 2005).
Th e value and utility of distributional checklists.What is the objective of pursuing such questions?Several useful purposes are served.Taxonomy is, of course, the bedrock of many areas of biology.Resolving questions of systematics and nomenclature lies at the heart of successful scientifi c communications in relation to many spheres of biological investigation, and the proper characterisation of species lies at the heart of this pursuit.If not correctly resolved, "phantom species" can continue to drift through systematic literature, potentially obscuring legitimate phylogenies.Consequently it is an important goal of taxonomy to investigate un-reviewed taxa and resolve their status.
Dependent on the resolution of such taxonomic studies, are questions of biodiversity.Knowing the faunal composition of a habitat, park, reserve, county, province, state, or country is valuable information.Faunal composition is a refl ection of underlying ecological dynamics, environmental circumstances, and historical processes.Th us, a correct understanding of the fauna opens up possible avenues of understanding and can suggest profi table courses of investigation and research.To monitor for change (whether as a result of natural processes, or by anthropogenic factors such as climate change, pollution, habitat loss, introduction of species, etc.) one needs to establish baselines from which to measure.Sikes (2004: 1) wrote, "If, a century from now, the checklist were to be revised again, the Rhode Island beetle fauna could become part of a valuable long-term study, with updates and monitoring at 100-year intervals."Monitoring for ecological change can be useful over intervals of decades as well as of centuries.All these objectives are served by accurate representation of the species composition.
Conservation eff orts are also assisted by the development of distributional checklists, that can help to delineate present ranges and hence where conservation eff orts should be focused.Sikes (2004: 14) wrote, "For lack of adequate knowledge, globally rare species whose habitat requirements are entirely unknown may be slipping towards extinction in the 'backyards' of New England's most capable naturalists and conservationists.Th is is an often-cited concern for conservation eff orts in the understudied tropics but it is, unfortunately, equally relevant to regions of the world that should already be far better known." Distributional checklists have a plethora of practical uses.Th ey are of utility to investigators in correlating data sets (i.e., faunistic and fl oristic) in searching for instructive patterns at their intersections.Th ey can assist in identifi cation by narrowing down the set of likely candidates.A sometimes-neglected point is that distributional checklists can be as informative for what they lack as well as for what they contain.For instance, the lack of records of a species in an area can lead to questions as to whether such an absence represents a real gap, or simply an artifact of to insuffi cient collecting eff ort.Th is can then serve as a springboard for further research.
Distributional checklists can also be of value in the context of zoogeographic investigations.Latitudinal gradients in species diversity are amongst the longest recognized, widely documented, hotly disputed, and least understood phenomena in biogeography.Explaining them is considered by some scientists to be one of the greatest challenges of contemporary ecology (Pianka 1966;Gaston and Blackburn 2000;Pimm and Brown 2004).A large number of closely related and sometime interdependent hypotheses have been advanced to account for this phenomenon such as the mid-domain eff ect, geographical area hypothesis, the species-energy hypothesis, the climate harshness hypothesis, the climate stability hypothesis, the historical perturbation hypothesis, the evolutionary rate hypothesis, and the eff ective evolutionary time hypothesis (Pianka 1966;Gaston and Blackburn 2000;Hillebrand 2004;Pimm and Brown 2004).Hillebrand (2004) conducted a meta-analysis of 581 studies on latitudinal gradients in species diversity.In the present context, it simply worth noting that a preponderance of studies have pertained to birds and molluscs, and relatively few have examined insects.Janzen (1981) focused on Nearctic Ichneumonidae (Hymenoptera) concluding that the peak of species diversity for this group was between 38° and 42° N, an exception to the generally observed rule that species diversity decreases with increasing latitude.Gotelli and Ellison (2002) examined species den-sity of ants in New England concluding that latitude was the single-most important determinate in species density, even over a limited latitudinal span.Th eir results gave support to species-energy relationships at regional spatial scales.Despite the species-richness of Coleoptera, the group has been very little investigated with regard to latitudinal gradients.
Figure 1 illustrates a gradient of alpha species richness over geographic latitude as revealed by data from state and provincial distributional checklists.A linear regression of these data yields the equation Y= -13.97*X + 834.23 which is highly signifi cant (F=53.26DF = 1.12 p=0.0000), thus quantifying the relationship of species diversity to geographic latitude for eastern North American Coleoptera.Th ese numbers are in general agreement with the strength of the latitudinal gradient relationship (i.e., the signifi cance of the correlation coeffi cient) and its slope (i.e., the measure of the latitudinal gradient), found in Hillebrand's ( 2004) meta-analysis.Although comparable informa-    Finally, it should be emphasized that hard-copy publications of checklists, although valuable, are being increasingly augmented by dynamic, database-driven, online resources.Good examples of such resources include the Nearctic Spider Database (Shorthouse 2008) and the Global Biodiversity Information Facility (GBIF 2008), a portal through which data from many institutions are served.Users interested in species occurrence data can extract regional checklists electronically from such sources.Th e data are, in most cases, specimen-based allowing for misidentifi cations to be corrected, maps to be generated dynamically, and data to be added and edited over time.Th ese solutions are leading the way towards a future in which distributional biodiversity data are freely and easily available to the public, which, in turn, can provide greater public input (e.g., Esch 2008) and concern with respect to biodiversity conservation issues.
Such patterns illustrate the nature of zoogeographic investigations that are made possible through the compilation of accurate distributional checklists.For all these reasons it is worthwhile to continue fi eldwork and taxonomic research that contribute to such resources at all spatial scales.
Coleoptera in Eastern North America tion from western North American jurisdictions is limited, what data exists for Alaska, British Columbia, and California shows a similar slope but at a higher absolute level.Another illustration of the utility of distributional checklists is illustrated by Figure2which shows the proportionate composition of the ten most abundant North American families of Coleoptera(Marske and Ivie 2003) of the North American beetle fauna as a whole, as well as that found in a number of Canadian provinces (Nova Scotia, New Brunswick, Prince Edward Island, Québec, and Ontario) and American states (Rhode Island, New Hampshire, Florida, and California).Although the proportionate composition of some families (Curculionidae, Cerambycidae, and Elateridae) is relatively constant across northern, southern, eastern, and western jurisdictions, the composition of other families (Staphylinidae and Carabidae) shows marked variations from jurisdiction to jurisdiction.Families such as the Scarabaeidae, Tenebrionidae, and Melyridae exhibit a much greater proportionate representation in southern and western regions than they do in the northeastern portions of North America.Distributional checklist data can therefore yield important insights with respect to the geographical basis of the composition of the North American beetle fauna.

Table 2 .
Valid species confi ned to Rhode Island and unrevised species