Taxonomic Revision of Hispaniola Tiger Beetles in the Genus Brasiella Rivalier 1954 (Coleoptera, Carabidae, Cicindelinae) Author Acciavatti, Robert E. text ZooKeys 2011 147 99 182 http://dx.doi.org/10.3897/zookeys.147.2012 journal article http://dx.doi.org/10.3897/zookeys.147.2012 1313-2970-147-99 Brasiella youngi sp. n. Figs 1617 Holotype. Male! labeled "DOMIN.REP.:Prov.Pedernales / 13.5km N Cabo Rojo, 140m / 21AUG-10SEP1988, flight / intercept trap, M.A.Ivie / T.K.Philips & K.A.Johnson" [typeset black on white label]; "HOLOTYPE / Brasiella / youngi / Acciavatti" [typeset black on red label]. [Genitalia in glycerin in a microvial pinned beneath specimen.] Allotype. Female! labeled "DOMINICAN REPUBLIC: / Pedernales. 26 km N / Cabo Rojo, 730 m. / 18-06N, 71-38W" [typeset black on white label]; "25-27 September 1991. / R.Davidson, C.Young / S. Thompson, J.Rawlins / Wet deciduous forest" [typeset black on white label]; "Carnegie Museum / Specimen Number / CMNH-132,618" [typeset black on white label]; "ALLOTYPE / Brasiella / youngi / Acciavatti" [typeset black on red label]. Type Depositories. Holotype at WIBP; allotype at CMNH, [CMNH Unique Number stored in data files at CMNH]. Type Locality. DOMINICAN REPUBLIC: Pedernales Province, 13.5 km N Cabo Rojo, 18°02'04"N , 71°38'38"W at 140 m, in the Sierra de Baoruco. Aerial view taken at lower elevations on the southern slopes of the Sierra de Baoruco in Fig. 21E. Notes on the Type Locality. No coordinates are on the holotype labels; however, an approximation of the coordinates shown above was obtained by measuring 13.5 km to 140 m along the only road north of Cabo Rojo using Google Earth©. Deciduous woodlands adjoining desert scrub typically are found at this elevation on the lower southern slopes of the Sierra de Baoruco (Figs 21E, 21F). The allotype was collected at a site called Haitian Hut on the CMNH 1991 expedition (J.E. Rawlins, 2010, personal communication). The site was given this name because of the Haitian squatters who have gained easy access to the lower southern slopes of the Sierra de Baoruco by using the highway built by Alcoa, Inc., to reach bauxite mines higher in the Sierra de Baoruco ( Woodruff 2004 ). The Haitian Hut site (Figs 21C, 21D), located in the deciduous forest accessible from the Alcoa highway, has coordinates of 18°06'42"N , 71°37'21"W at 730 m at 26 km N of Cabo Rojo; these slightly more precise coordinates, than originally obtained in the field, were based on using Google Earth©. Diagnosis . Distinguished from Brasiella species on Hispaniola by the following combination of characters: 1) head small, eyes proportionally large, prominent and distinctly bulging laterally; 2) subglobose to subarcuate pronotum; 3) female legs and antennal scape primarily testaceous, male appendages dominated by dark metallic reflections; 4) female mesepisternal coupling sulcus a shallow, concentric depression with a small, central pit; 5) faint, nearly translucent elytral markings, forming a complete, but reduced pattern with humeral lunule divided into a humeral spot and a discal dot; 6) female 5th abdominal sternum with a small, membranous wedge along the posterior margin and a wide, membranous band extending along midline only to middle; 7) female 8th sternum median notch shallowly incised. Description. General.Figs 16A, 17A. Body. Formslender; head small, eyes proportionally large, prominent and distinctly bulging laterally; pronotum slender, subglobose to subarcuate, wider than long; elytra uniformly narrow, apices separately rounded in female, conjointly rounded in male; appendages primarily testaceous in female, mainly dark metallic in male. Size.Male, length 6.0 mm, width 1.7 mm; female, length 6.2 mm, width 1.8 mm. Head. Figs 16B, 16D, 17D, 17F. Shiny dark copper brown dorsally and blue green ventrally; entire surface glabrous except for two pairs of supraorbital sensory setae. Frons finely and longitudinally rugose. Vertex more coarsely rugose, transverse rugae along anterior margin narrow and irregularly arranged, 15-18 more or less complete longitudinal rugae between eyes and middle, in female rugae mainly parallel only slightly converge, in male rugae converge into an arcuate pattern; rugae transition abruptly into a posterior area with a finely and irregularly granulate surface. Eyes prominent and greatly bulging laterally in both sexes. Genae longitudinally rugose. Clypeus finely and irregularly granulate, narrowed mesad. Labrum testaceous with a dark brown margin, rectangular, width to length ratio 3 in holotype male, ratio 3 in allotype female; anterior margin nearly straight with a small tooth; posterior margin distinctly arcuate mesad; medial carina broadly raised; 8 setae in an irregular row near middle symmetrically arranged. Maxillae and labium mainly testaceous, only distal palpal segments dark brown with metallic green black reflections. Mandibles sexually dimorphic; in male, surface mainly testaceous, only teeth metallic green; in female, surface only testaceous in basal half, apical half and teeth shiny brown; mandibles symmetrical, four teeth distad of molar, apical tooth longest, first and third tooth coequal in length, second tooth shortest; gaps between three intermediate teeth narrow in male, wide in female. Antennae 11 segmented; scape in female entirely testaceous, in male dorsally shiny green, ventrally testaceous; scape with a single subapical sensory seta; antennomeres 2-4 shiny copper green, glabrous except for a few, short erect setae along their length and distally; antennomeres 5-11 dull brown, sheathed with dense short sensory setae. Prothorax.Figs 16C, 16D, 17C, 17D. Pronotum shiny, dark copper brown. Proepisterna shiny, dark copper brown, surface wrinkled dorsad. Pronotum glabrous except for short, decumbent, white setae distributed in several, irregular rows medially directed, originating close to, and lying in a narrow band inside and slightly impinging on lateral suture, in a sparse narrow band transversely and anteriorly oriented on anterior margin, and in a sparse narrow band laterally oriented on each side of midline extending nearly to the narrow posterior margin; transverse submarginal sulci distinct, anterior sulcus shallow, posterior sulcus deeper and deepest at posterior angles; transverse rugae within broad anterior margin irregular and shallow, interrupted at middle by an irregularly arranged pattern, within posterior margin more distinctly and deeply engraved especially medially and extending onto midline; surface sculptured by fine, transverse rugae angled on disc and interrupted by a finely engraved longitudinal midline, and more finely and irregularly sculptured elsewhere. Proepisterna glabrous except for white, erect and appressed setae arising from small setigerous punctures scattered over most of the surface in male, only ventrally and along anterior margin in female. Prosternum glabrous, surface slightly roughened. Pterothorax.Figs 16C, 17C. Mesepisterna glabrous except for appressed setae near ventral margin; female mesepisternal coupling sulcus a shallow, concentric depression with a small, central pit, indistinct groove extends only dorsally from pit, surface smooth below pit. Mesepimeron with a few appressed setae. Metepisterna with scattered appressed setae, more abundant in male than female. Prosternum and mesosternum glabrous, smooth to slightly wrinkled; metasternum glabrous except for long, dense white appressed setae laterad, surface smooth mesad and coarsely sculpted laterad where setae originate. Scutellum triangular, cupreous. Legs.Figs 16A, 17B. Appendages in female primarily testaceous and translucent except for slightly darker coxae and tarsomeres; appendages in male dominated by dark metallic reflections. Coxae shiny metallic brown green; tarsomeres shiny violet to green; white, appressed setae on front and middle coxae, and laterally on hind coxae; erect setae and suberect closely spaced in several regular and irregular rows on all femora; setae widely spaced in a few rows on all tibiae; middle tibiae with patch of appressed setae dorsally along distal half; tarsomeres with short scattered setae mainly on ventral surface; distal tarsomeres with two asymmetrical rows each with a few to several small, erect setae; an erect subapical seta present only on front trochanter, absent on middle and hind trochanters; males with dense pad of erect setae ventrally on proximal three tarsal segments; tarsal claws small. Elytra.Figs 16A, 17A. Form uniformly narrow in both sexes; apices separately rounded in female, conjointly rounded in male; sutural spine at apex small and distinct; posterior margins minutely microserrulate. Surface finely granulate, impunctate, numerous small, irregular, shiny green or blue green flecks of various sizes scattered over a dull, dark copper brown background; fully developed elytral pattern of broad, bold markings contrasting with the darker elytral ground color; setigerous punctures with short, erect, transparent setae indistinct in subsutural rows on disc, but distinct at elytral base, and at inner humeral angles, each surrounded by a metallic fleck slightl y larger than flecks elsewhere on elytra; surface slightly depressed in humeral area and on disc creating a slight but distinct raised area basally. Elytra dull cupreous ground color with metallic blue and green flecks scattered randomly over the unmarked portions of the surface. The markings faint, nearly translucent elytral markings, forming a complete, but reduced pattern with humeral lunule divided into a humeral spot and a discal dot, a complete middle lunule enlarged near the lateral margin and near the suture, and a complete apical lunule broadly reaching the suture. Elytral epipleura testaceous except for narrow, metallic green to copper green band along dorsal margin. Abdomen.Figs 17B, 17E. Surface of 1st-5th sterna shiny black with green reflections, 6th sternum entirely shiny black to black brown; posterior margins of male 3rd-5th sterna and female 3rd-4th sterna narrowly black; posterior margin female 5th sternum broadly black; 3rd-5th sterna medially smooth with scattered, fine, erect setae in both sexes; male 1st-6th sterna and female 1st-5th sterna laterally mainly covered with dense, scattered, appressed white setae and roughened from setal punctures; male 6th sternum glabrous medially with a broad, deep concave notch; female 5th sternum with moderately raised, transverse wrinkles interrupted by a wide, membranous band along midline extending anteriorly only to middle of sternum from a small, membranous wedge along posterior margin; female 6th sternum entirely glabrous, posterior margin with a row of 6-10 erect spines and a small lateral gibbosity on each side; female 8th sternum median notch shallowly incised. Male Genitalia.Figs 16E, 16F, 16G. Shape narrow near base, uniformly broad in along most of its length, distally narrowed to a short and wide neck, apical hook inner angle acutely rounded and outer angles evenly rounded, tip long and acutely angled to aedeagus. Aedeagus inner sac sclerites: stylet long and straight,, tip pointed; shield unevenly rounded distad; large tooth small and pointed at tip with very long root and large dark fields; arched piece long and thick; spine field within aedeagus neck short and forked. Ecology. The unique holotype male was collected at 130 m in a flight intercept trap in deciduous forest, whereas the unique allotype female specimen was collected in wet, deciduous forest at 730 m. Deciduous forest typical of the lower slopes is shown in Fig. 20F, whereas deciduous forest typical of the middle slopes is shown in Fig. 20D. Although the collection method for the allotype was not specified on its labels, a passive method of either malaise or yellow pan traps were both employed by CMNH staff while collecting at a site for several days. Both specimens were found on the lower south slopes of the Sierra de Baoruco in September. Vegetation in this part of the Sierra de Baoruco gradually transitions from desert-like habitats below 255 m through dry deciduous forest at 255 m to moist tropical forest and eventually grass and pine habitat at higher elevations ( Woodruff 2004 ). Because of this great diversity of habitats, these mountains at the eastern end of the southernmost part of the Dominican Republic have high floristic and faunal endemism ( Woodruff 2004 ). Distribution. Fig. 22. DOMINICAN REPUBLIC: Pedernales Province, 13.5 km to 26 km north of Cabo Rojo on the lower southern slopes of the Sierra de Baoruco in deciduous forest habitats at elevations from 130 to 730 m. Etymology. This Latinized eponym, genitive case, is based on the last name of Chen W. Young, Associate Curator of Invertebrate Zoology, Carnegie Museum of Natural History, Pittsburgh, PA, and world authority on the Diptera family Tipulidae. Chen participated in most of the expeditions by CMNH staff to the Dominican Republic and specifically the one that collected of this new Brasiella species. As a friend and colleague, I have the honor of naming this new species for him. Remarks. Brasiella youngi and Brasiella rawlinsi appear to be closely related species based on their male genitalia, but their distinctiveness as separate species has been established by differences in the external adult male and female characters presented in the key and under their descriptions. Obvious morphological differences between the two species are exhibited by the size of their eyes and the extent to which the pale elytral pattern is infuscated by the darker ground color. Other important distinctions exist between females of these two species in the depth and position of the female coupling sulcus, the extent to which the membranous, longitudinal median band is developed on the 5th abdominal sternum and the form of the emargination on the 8th sternum. Although both species occur in the Sierra de Baoruco at the same time of year, each species occurs at a different elevation in quite different habitats (refer to Ecology for each of these species). Brasiella youngi is less closely related to Brasiella iviei based on their male genitalia; although both of these species apparently occur in more moist habitats than Brasiella rawlinsi . Figure 16. Brasiella youngi, sp. n., male. Holotype 16A Body, dorsal 16B Body, ventral 16C Body, anterior, left lateral 16D Body, anterior, dorsal 16E Aedeagus, dorsal 16F Aedeagus inner sac, ventral aspect, and 16G Aedeagus inner sac, dorsal aspect-AP , arched piece LT large tooth; SH, shield; ST, stylet; SF, spine fields (one displaced from within aedeagus neck and two isolated); RT, root of LT; DF, dark fields. [Scale lines = 1 mm]. Figure 17. Brasiella youngi, sp. n., female. Allotype 17A Body, dorsal 17B Body, lateral 17C Body, anterior, left lateral-CS , coupling sulcus 17D Body, anterior, dorsal 17E Abdomen, sterna, ventral- 5th sternum, MB, membranous band; 5th sternum, MW, membranous wedge; 6th sternum, LG, lateral gibbosity; 8th sternum, MN, median notch 17F Head, anterior. [Scale lines = 1 mm]. Figure 18. Brasiella Habitats on Hispaniola 18ABrasiella bellorum, sp. n., type locality vicinity; aerial view of fields and forests at U-shaped curve in road network near El Convento, Dominican Republic 18BBrasiella bellorum, sp. n., type locality; Robert L. Davidson collecting along eroded pasture road 18CBrasiella bellorum, sp. n., type locality; Robert L. Davidson collecting at eroded clay bank in pasture 18DBrasiella bellorum, sp. n., type locality; breeding site in clay bank with larval burrows (LB) and adults (A) 18EBrasiella darlingtoniana, sp. n., type locality vicinity; aerial view along Highway 101 below La Visite National Park, Massif de la Selle, Haiti 18FBrasiella darlingtoniana, sp. n., type locality; eroded banks and road surface along Highway 101 below La Visite National Park, Massif de la Selle, Haiti. Figure 19. Brasiella Habitats on Hispaniola 19ABrasiella davidsoni, sp. n., type locality vicinity; aerial view of roads and fields in disturbed forests at Morne Formon southern slope of the Massif de la Hotte, Haiti 19BBrasiella dominicana (Mandl), type locality vicinity; aerial view of Highway 2 crossing Rio Bani along east side of Bani, Dominican Republic 19CBrasiella ocoa, sp. n., type locality vicinity; aerial view of Rio Ocoa flood plain south of Highway 2 near Las Carreras, Dominican Republic 19DBrasiella ocoa, sp. n., habitat; sparsely vegetated riverbanks along the sand, gravel and rock river bed of Rio Ocoa at Highway 2 crossing near Carreras, Dominican Republic 19EBrasiella philipi, sp. n., type locality vicinity; aerial view of dirt roads among fields and forests at Cierracita above Mata Grande in the northern foothills of the Cordillera Central, Dominican Republic 19FBrasiella philipi, sp. n., type locality vicinity; clay soil openings in field looking south to the Cordillera Central, Dominican Republic. Figure 20. Brasiella Habitats on Hispaniola 20ABrasiella iviei, sp. n., type locality vicinity; aerial view of forested ravines similar to the Las Abejas type locality, mid-elevations, southern slopes of the Sierra de Baoruco, Dominican Republic 20BBrasiella iviei, sp. n., habitat; eroded, clay banks of watercourse along trail at upper parts of Las Abejas, south side of the Sierra de Baoruco, Dominican Republic 20CBrasiella iviei, sp. n., habitat; trail entering upper parts of Las Abejas, southern slopes of the Sierra de Baoruco, Dominican Republic 20DBrasiella iviei, sp. n., type locality vicinity; tropical malaise trap at lower parts of Las Abejas set by 1987 CMNH Expedition, southern slopes of the Sierra de Baoruco, Dominican Republic 20EBrasiella rawlinsi, sp. n., type locality vicinity; Robert L. Davidson at 1987 CMNH Expedition campsite in compacted clay opening within an old burn regenerating with shrubby regeneration and young Pinus occidentalis Swartz in background, upper elevations, southern slopes of the Sierra de Baoruco, Dominican Republic 20FBrasiella rawlinsi, sp. n., type locality vicinity; Aceitillar with Pinus occidentalis Swartz forest at upper elevations, southern slopes of the Sierra de Baoruco, Dominican Republic.Dominican Republic. Figure 21. Brasiella Habitats on Hispaniola 21ABrasiella rawlinsi, sp. n., type locality vicinity; aerial view of Aceitillar, Andropogon sp. grassland with Pinus occidentalis Swartz and bauxite surface mines at upper elevations, southern slopes of the Sierra de Baoruco, Dominican Republic 21BBrasiella rawlinsi, sp. n., habitat; Aceitillar, Andropogon sp. grassland openings in young Pinus occidentalis Swartz on red clay soils, southern slopes of the Sierra de Baoruco, Dominican Republic 21CBrasiella youngi, sp. n., type locality vicinity; aerial view of wet, deciduous forest near the Haitian Hut along Alcoa Road at mid-elevations on the southern slopes of the Sierra de Baoruco, Dominican Republic 21DBrasiella youngi, sp. n., habitat: wet deciduous forest typical of allotype site of this species was collected 21EBrasiella youngi, sp. n., type locality vicinity; aerial view of deciduous woodlands adjoining desert scrub typical of holotype site at lower elevations on the southern slopes of the Sierra de Baoruco, Dominican Republic 21FBrasiella youngi, sp. n., habitat; dry, deciduous woodlands typical of holotype site on the lower slopes of the Sierra de Baoruco, Dominican Republic. Figure 22. Geographic Distribution of Brasiella species on Hispaniola. The type locality, other localities, country, province or departement and geographic feature associated with each species: (ORANGE) Brasiella bellorum, sp. n., DOMINICAN REPUBLIC: La Vega Province, El Convento, Cordillera Central; (VIOLET) Brasiella darlingtoniana, sp. n., HAITI: Departement du l'Ouest, La Visite, Massif de la Selle; (BLUE) Brasiella davidsoni, sp. n., HAITI: Departement du l'Sud, Ville Formon, Massif de la Hotte; (YELLOW) Brasiella dominicana (Mandl), DOMINICAN REPUBLIC: Peravia Province, Bani, Rio Bani; (GRAY) Brasiella iviei, sp. n., DOMINICAN REPUBLIC: Pedernales Province, Las Abejas, Sierra de Baoruco; (LIME GREEN) Brasiella ocoa, sp. n., DOMINICAN REPUBLIC: Peravia Province, Las Carreras, Rio Ocoa; (GREEN) Brasiella philipi, sp. n., DOMINICAN REPUBLIC: Santiago Province, La Sierrecita, Cordillera Central; (RED) Brasiella rawlinsi, sp. n., DOMINICAN REPUBLIC: Pedernales Province, Aceitillar, Sierra de Baoruco; (PURPLE) Brasiella youngi, sp. n., DOMINICAN REPUBLIC: Pedernales Province, 13.5 km N Cabo Rojo, Sierra de Baoruco. Relationships within the Brasiella virdicollis Species Group of Brasiella Rivalier Although a phylogenetic analysis of Brasiella species on Hispaniola is not part of the present revision, a general discussion of the latest published phylogenetic treatment of Brasiella relative to the Hispaniola species, is considered relevant. Emphasis is placed on the composition of the Brasiella virdicollis species group, along with its current and ancestral geographic distributions in the West Indies. Such a discussion is considered relevant for suggesting further studies to investigate the phylogenetic relationships for the Hispaniola Brasiella species treated in this revision relative to other Western Hemisphere Brasiella . Previous phylogenetic and biogeographic studies of tiger beetle genera in Brazil ( Freitag and Barnes 1989 ) and West Indies ( Freitag 1992 ) recognized that Brasiella sensu Rivalier possessed a unique suite of synapomorphous adult characters not found in related Western Hemisphere taxa of Cicindelinae. However, their studies treated Brasiella Rivalier 1954 subspecifically within Cicindela Linnaeus 1758, a concept held by many North American tiger beetle students, but not by those studying either elsewhere in the world or the higher classifications of Coleoptera. Consequently, in the most recent and comprehensive treatments of the world Geadephaga ( Lorenz 2005 ) and Western Hemisphere Caraboidea ( Erwin and Pearson 2008 ), Brasiella was recognized as distinct among Cicindelina Latreille 1802 lineages and given full generic status. Although treated as a distinctive genus, the latter authors considered the taxonomic status of species within Brasiella to be unstable, likely because further discovery of new species would require changes to the current classification. As the present revision demonstrates, the Neotropical Brasiella fauna is likely to be far richer than indicated by current faunal lists, especially after habitats have been more fully explored in both new and previously visited geographic areas throughout all seasons. Additionally, more detailed studies of genitalic structures, among populations of those species already described, may reveal cryptic species, as the present revision has shown to be the situation on Hispaniola. Indeed, the discovery and description of new species periodically provides the opportunity to reconsider any existing published phylogeny so that the relationships of the new species to those previously described are better understood. One approach would be to use the criteria of the existing classification and simply fit the new species into it. This first approach seems appropriate when a small number of new species are considered from a restricted geographic area where populations have been adequately sampled such that the faunal diversity is considered to be entirely known. Another approach would be a more comprehensive reevaluation of the criteria of the existing classification in order to develop a revised classification. This second approach would be most appropriate when considering a large number of new species originating over broadly diverse geographic areas. Under this second approach, results would be most informative when populations from all these geographic areas have been adequately sampled. Thus, for either approach the question of adequately sampling existing populations is critical to any meaningful reconstructed phylogeny. Because the present revision revealed new Brasiella species with each expedition to Hispaniola, there is a high probability that many additional new species will be discovered on this island. Further collecting is likely to find these, not only within new areas and habitats, but also within those previously visited, especially during different seasons on Hispaniola. Even though the Brasiella populations on Hispaniola may not have been completely sampled, this present revision did provide an opportunity for subjective interpretations of the species relationships within this genus on Hispaniola based on the population samples that were collected thus far and studied by the author. These relationships were discussed earlier in this revision under the Remarks within each Species Account. Such interpretations, while based on the author's observations and experience with the species, should provide a working hypothesis for any future phylogenetic analysis designed to examine more rigorously the relationships among the Hispaniolan Brasiella species, all of which are endemic to this island and superficially appear similar to one another. The current classification into species groups of nearly all Brasiella described at the time was presented by Freitag and Barnes (1989) . Brasiella dominicana (Mandl), described in 1983, unexplainably was overlooked. However, this species later was studied and subsequently assigned to a species group within Brasiella by Freitag (1992) . In their publication, Freitag and Barnes (1989) established seven species groups for 29 species within the genus based on a reconstructed phylogeny involving 50 adult morphological characters. Despite the fact that male genitalic characters of most species were not actually examined and documented by dissection, but rather interpreted only from the published descriptions and illustrations, their results were important because they provided a basis for all subsequent efforts to phylogenetically organize the ever-growing number of Brasiella species found in the Western Hemisphere. The latest catalogue ( Erwin and Pearson 2008 ) treated 39 Brasiella species, and with the eight more added in the present revision alone, Brasiella is now at 47 species. Although Freitag and Barnes (1989) established morphological characters and assigned their polarities--plesiomorphic (ancestral) or apomorphic (derived)--for all 50 adult characters, male and female genitalic characters were given the most value in establishing the species groups within Brasiella phylogeny. While they treated primarily Brazilian species in their publication, they did establish the viridicollis species group for several Brasiella species occurring in the West Indies together with one species from North America. However, this species group was not discussed in detail until Freitag (1992) studied the West Indian Cicindelidae. In that publication, he reassigned two Brasiella species he had originally placed within the viridicollis species group, and added one species he had previously overlooked, so that the species group at that time included the following three species-- Brasiella viridicollis (Dejean 1831) from Cuba, Brasiella dominicana (Mandl) from Hispaniola, and Brasiella wickhami (W. Horn 1903) from northern Mexico and the southwestern United States. The numerous new species described in the present revision have made it necessary to redefine the earlier concept of the Brasiella viridicollis species group. Decisions to include newly described species, or exclude those previously described, would result from applying the revised concept. The present revision suggests that the Brasiella viridicollis species group concept would be strengthened if a previously overlooked adult female structure, a small gibbosity located on each side of the 6th abdominal sternum, was included in its concept. This structure, and two membranous structures on the 5th abdominal sternum, are considered to be apomorphic. The synapomorphic gibbosity character for females of nearly all Hispaniolan Brasiella , as well as, the Cuban Brasiella virdicollis (Dejean). provides further evidence for treating the viridicollis species group as an evolutionary unit within the Brasiella . All three structures appear relevant to the Brasiella viridicollis species group concept and are discussed in detail here. Freitag and Barnes (1989) were the first to report a membraneous structure along the posterior margin of the female 5th sternum. They considered this unusual female abdominal character unique within Western Hemisphere Cicindelinae and termed it an "unpigmented bell-shaped spot". However, this structure is actually membranous as explained above under Methods; thus, in the present revision it has been termed a "membranous wedge." The second membranous structure, also on the same female 5th abdominal sternum, is a longitudinal "membranous band", mesad and anterior to this membranous wedge. The membranous band does not seem to have been considered a separate structure by Freitag and Barnes (1989) . These two structures exist for nearly all Brasiella species on Hispaniola. The exceptions being Brasiella ocoa whose females are unknown, and Brasiella davidsoni whose females plausibly have secondarily lost the 5th abdominal sternum structures. The third structure is the small, raised gibbosity on each side of the female 6th abdominal sternum located mesad on each surface. These two gibbosities have not been previously mentioned in any publication dealing with Brasiella phylogeny ( Freitag and Barnes 1989 )( Freitag 1992 ). As mentioned above, these abdominal gibbosities are considered an important apomorphic character for including or excluding species within the established Brasiella species groups. The gibbosity character is synapomorphic for females of eight of the nine Brasiella species on Hispaniola; Brasiella ocoa is known only from males, but its females are likely to possess this character. Those species in the Brasiella viridicollis species group that do not occur on Hispaniola require consideration. Apparently, two subspecies of Brasiella virdicollis now have been recorded on Cuba; the nominotypic subspecies and Brasiella viridicollis fernandozayasi (Kippenhan, Ivie & Hopp 2009). Cicindela fernandozayasi was proposed as a replacement name ( Kippenhan et al. 2009 ) for Cicindela carbonaria Zayas 1988 ( Zayas 1988 ). Zayas (1988) described his species from a single specimen without any detail as to where it was collected on Cuba. This taxon currently is classified as Cicindela viridicollis fernandozayasi (Kippenhan, Ivie & Hopp 2009), but treated as a Brasiella in this revision. This subspecies has a completely black dorsal body color compared to nominotypic Brasiella viridicollis with its distinctive blue green head and pronotum contrasting with dull brown elytra covered by metallic blue green flecks. Nominotypic Brasiella virdicollis (Dejean) possesses all three of these apomorphic structures on the two last visible sterna based on the author's study of female specimens from Cuba, Cienfuegos Province, Minas Carlota, Sierra de Trinidad, at CMNH. The same cannot be said for the second subspecies because no specimens of Brasiella viridicollis fernandozayasi were available for examination. This subspecies is known only from its holotype deposited in the Fernando Zayas Collection in Havana, Cuba ( Kippenhan et al. 2009 ). There seems little doubt about its placement within the Brasiella viridicollis species group, however, its subspecific status may need to be reconsidered. The color image of the type presented by Kippenhan et al. (2009) exhibits some morphological characters different from the nominotypic Brasiella viridicollis that suggest it may represent a separate species. Although the sex of the holotype of Brasiella viridicollis fernandozayasi was not indicated when it was described as Cicindela carbonaria Zayas 1988 , nor when it received its current replacement name, it appears to be a male. The conclusion is based on the expanded setal pads on the proximal three tarsomeres, the shape of its elytral apices and the protruding abdomen from the illustration and image in these two publications. When compared with a male of Brasiella viridicollis , the most obvious difference between these two subspecies, other than their strikingly different dorsal body color, lies in the large, laterally bulging eyes of Brasiella viridicollis fernandozayasi compared with the smaller, less bulging eyes of the nominotypic Brasiella viridicollis . Final resolution about the specific status of each subspecies must await additional examination of their types, as well as, the discovery of more male specimens of Brasiella viridicollis fernandozayasi whose genitalia can be dissected. Evidence is presented here that Brasiella wickhami (W. Horn) should not be considered part of the Brasiella viridicollis species group. This species has been included within the Brasiella viridicollis species group ( Freitag and Barnes 1989 , Freitag 1992 ) on the basis of its aedeagus having a shield rounded at the apex and slightly protruding from the male genitalia. However, the shield for Brasiella wickhami is actually a more complex sclerite than the simple shield possessed by the other species within the Brasiella viridicollis species group. Rivalier (1955) depicted the shield of Brasiella wickhami as a large sclerite with both a rounded knob and a pointed section at the apex. My examination of specimens from Arizona at CMNH, confirm this more complex shield shape. Moreover, Brasiella wickhami females lack the lateral gibbosities on the 6th abdominal sternum. From these findings, it is suggested that Brasiella wickhami should not be a part of the Brasiella viridicollis species group. The assignment of Brasiella wickhami to another species group within Brasiella must await a phylogenetic analysis of the other mainland species in the Neotropics of North and Central America. As the previous discussion suggests, the revised Brasiella viridicollis species group concept includes only the following ten species and two subspecies arranged here alphabetically by country and species within each country: Brasiella viridicollis fernandozayasi (Kippenhan, Ivie & Hopp) - Cuba Brasiella viridicollis viridicollis (Dejean) - Cuba Brasiella bellorum , new species, Acciavatti - Dominican Republic Brasiella dominicana (Mandl) - Dominican Republic Brasiella iviei , new species, Acciavatti - Dominican Republic Brasiella ocoa , new species, Acciavatti - Dominican Republic Brasiella philipi , new species, Acciavatti - Dominican Republic Brasiella rawlinsi , new species, Acciavatti - Dominican Republic Brasiella youngi , new species, Acciavatti - Dominican Republic Brasiella darlingtoniana , new species, Acciavatti - Haiti Brasiella davidsoni , new species, Acciavatti - Haiti How these current species distributions within the Brasiella viridicollis species group may have been influenced by their ancestral geographic distributions should be briefly considered. Previous biogeographic studies of tiger beetle genera in the West Indies provided only a partial explanation of the possible pathways by which the Brasiella viridicollis species group may have arrived on Cuba and Hispaniola from its ancestral distributions. According to Freitag (1992) , the dispersal pathway accounted for the ancestors of the Brasiella on Cuba and Hispaniola. He suggested that these Brasiella ancestors where originally distributed in Middle America and arrived on these islands most likely through biogeographical patterns of dispersal common to tiger beetles. He did not distinguish between active and passive dispersal mechanisms for the large, actively flying Cicindela species compared with the small, primarily cursorial Brasiella species. Dispersal as a pathway implied movement of species from ancestral distributions in the Yucatan Peninsula or Nicaragua Highlands to the Greater Antilles over wide water gaps. Such dispersal might be rare, but likely has occurred over time even for small Brasiella species, as documented recently for Brasiella viridicollis from Cuba to the Florida Keys( Schiefer 2004 ) . Although the dispersal pathway undoubtedly has influenced the currrent distribution of the Brasiella viridicollis species group, the vicariance pathway also should be considered in explaining current geographic distributions. Woodruff (2004) provided insight into the likely influence of both dispersal and vicariance on the current geographic distributions for Phyllophaga species (Coleoptera: Scarabaeidae: Melolonthinae) on Hispaniola. He summarized the concepts about the geologic history of Hispaniola prevalent at the time of his revision. Brasiella species divergence on Hispaniola apparently occurred after the arrival of the ancestral members of the Brasiella viridicollis species group. Species divergence likely was greatly influenced by the extent and duration to which populations were isolated, especially during the Pleistocene. During interglacial episodes in that geologic time period, the Caribbean Sea would have risen from melting continental glaciers and inundated the lowlands. Lowland inundation over a long time could have kept populations isolated the longest in the highest and most widely separated mountain masses, such as the Massif de la Hotte in southwestern Haiti, where the most divergent Brasiella davidsoni presently exists on Hispaniola. All these biogeographic concepts discussed here should be considered, and discussed in greater detail, when interpreting future phylogenetic studies as to how the Brasiella species on Hispaniola relate to each other, as well as, to the other Brasiella species in the Neotropics.